mirror of
https://github.com/geekwenjie/SmartJavaAI.git
synced 2026-09-11 20:28:56 +00:00
Compare commits
56 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
f137644e46 | ||
|
|
553c195f2b | ||
|
|
7c220d02e8 | ||
|
|
25ab62ea3b | ||
|
|
1620efd9da | ||
|
|
90c4924a9f | ||
|
|
29e2a8e491 | ||
|
|
8a6f671703 | ||
|
|
c8cda3f240 | ||
|
|
9f50a9396c | ||
|
|
ac6a31b132 | ||
|
|
26f014e46e | ||
|
|
90acc76887 | ||
|
|
b40dcac4cb | ||
|
|
d4a6180687 | ||
|
|
5ae7670564 | ||
|
|
13805e1a87 | ||
|
|
36a19ffa87 | ||
|
|
0946f69f09 | ||
|
|
40c7cfbbcf | ||
|
|
191c6976d9 | ||
|
|
11f80ee4c4 | ||
|
|
f00e1d0d83 | ||
|
|
8b52ea1e3e | ||
|
|
171361459a | ||
|
|
1e854fa7ad | ||
|
|
6f3fc456e5 | ||
|
|
1249f7b22a | ||
|
|
7898087a83 | ||
|
|
213d6bfd08 | ||
|
|
4d589f0f88 | ||
|
|
f3f3a94aa3 | ||
|
|
b26b747d80 | ||
|
|
98337b081e | ||
|
|
2d4b710b9c | ||
|
|
5314d70fa0 | ||
|
|
42c30491ec | ||
|
|
14286729ec | ||
|
|
91a2f95ab8 | ||
|
|
b979320e16 | ||
|
|
d17d6546bc | ||
|
|
a7c67c2813 | ||
|
|
798864ed7b | ||
|
|
78271e20ed | ||
|
|
a549bc0a9d | ||
|
|
8a9400993a | ||
|
|
c37f2871af | ||
|
|
8ab216317b | ||
|
|
42479ef031 | ||
|
|
8abc70fdca | ||
|
|
20586cf7c3 | ||
|
|
37aa6d3bf6 | ||
|
|
bd3abab1b9 | ||
|
|
12db5d47d7 | ||
|
|
616a346d79 | ||
|
|
e8172670cd |
88
README.md
88
README.md
@@ -6,7 +6,7 @@
|
||||
<em>像Hutool一样简单易用的Java AI工具箱</em>
|
||||
</p>
|
||||
<p align="center">
|
||||
👉 <a href="http://smartjavaai.cn/">http://smartjavaai.cn/</a> 👈
|
||||
👉 <a href="http://numberone.ink/">http://numberone.ink/</a> 👈
|
||||
</p>
|
||||
<p align="center">
|
||||
<a target="_blank" href="https://central.sonatype.com/artifact/ink.numberone/smartjavaai-all">
|
||||
@@ -31,7 +31,7 @@
|
||||
|
||||
-------------------------------------------------------------------------------
|
||||
|
||||
[**开发文档**](http://doc.smartjavaai.cn)
|
||||
[**开发文档**](http://doc.numberone.ink)
|
||||
|
||||
-------------------------------------------------------------------------------
|
||||
|
||||
@@ -68,7 +68,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face5point.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face/face4.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -81,7 +81,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face1-1.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face1-1.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -93,7 +93,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/idcard.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/idcard.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -109,7 +109,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face1-n.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face1-n.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -126,7 +126,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face_attribute.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/face_attribute.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -139,7 +139,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/liveness2.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/liveness2.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -152,7 +152,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/emotion.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/emotion.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -174,15 +174,10 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
<p>目标检测(Object Detection)</p>
|
||||
- 视频流目标检测：rtsp、摄像头、视频文件等 <br>
|
||||
</div>
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/objectdect/object_detect_1.jpeg" width = "500px"/>
|
||||
</div>
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/objectdect/object_detection_detected.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/objectdect/object_detection_detected.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -294,7 +289,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/general_ocr_002_recognized.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/general_ocr_002_recognized.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -307,7 +302,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/table.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/table.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -321,14 +316,44 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/plate_recognized.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/plate_recognized.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/plate_recognized2.jpg" width = "500px"/>
|
||||
<div align="left">
|
||||
<p>身份证识别<br>(IDCard Recognition)</p>
|
||||
- 正面识别 <br>
|
||||
</div>
|
||||
</td>
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/idcard/idcard_front.png" width="500px"/>
|
||||
<pre align="left">{
|
||||
"name": "小氧",
|
||||
"gender": "女",
|
||||
"ethnicity": "汉",
|
||||
"idNumber": "430602200010108888",
|
||||
"birthday": "2000-10-10",
|
||||
"address": "湖南省岳阳市岳阳楼区金鄂中路456号"
|
||||
}</pre>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>
|
||||
<div align="left">
|
||||
<p>身份证识别<br>(IDCard Recognition)</p>
|
||||
- 反面识别 <br>
|
||||
</div>
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/ocr/idcard/idcard_back.png" width="500px"/>
|
||||
<pre align="left">{"issuingAuthority":"杭州市公安局江干分局","validFrom":"2015-11-05","validTo":"2025-11-05"}</pre>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
<tr>
|
||||
<td>
|
||||
@@ -339,7 +364,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/translate/translate.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/translate/translate.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -353,7 +378,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/speech/asr.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/speech/asr.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -365,7 +390,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/speech/tts.jpg" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/speech/tts.jpg" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -380,7 +405,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
</td>
|
||||
<td>
|
||||
<div align="center">
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/vision/clip.png" width = "500px"/>
|
||||
<img src="https://cdn.jsdelivr.net/gh/geekwenjie/SmartJavaAI-Site/images/vision/clip.png" width = "400px"/>
|
||||
</div>
|
||||
</td>
|
||||
</tr>
|
||||
@@ -447,6 +472,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
- 支持通用文字识别，通用手写字识别
|
||||
- 支持表格识别
|
||||
- 支持中文车牌识别：单层/双层检测，颜色识别，支持12种中文车牌
|
||||
- 支持身份证识别：支持身份证正反面字段提取、方向矫正与结构化解析
|
||||
- **机器翻译**
|
||||
- 集成NLLB-200模型：支持200+语言互相翻译
|
||||
- **语音识别(ASR)**
|
||||
@@ -500,21 +526,22 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
### 1、环境要求
|
||||
|
||||
- Java 版本：**JDK 8或更高版本**
|
||||
- 操作系统：不同模型支持的系统不一样，具体请查看[文档](http://doc.smartjavaai.cn)
|
||||
- 操作系统：Windows 64 位 / Linux / macOS M1
|
||||
- CPU架构：x86_64、ARM64（aarch64）
|
||||
|
||||
### 2、Maven
|
||||
|
||||
在项目的 `pom.xml` 的 `dependencies` 中可以一次性引入全部功能（如下所示）。
|
||||
|
||||
|
||||
⚠️ **注意：不推荐直接引入全部依赖**，更推荐根据实际需求，按功能模块单独引入，避免引入不必要的包。
|
||||
|
||||
详细引入方式请查看 [文档](http://doc.smartjavaai.cn/install.html)、或查看[示例代码](https://gitee.com/dengwenjie/SmartJavaAI/tree/master/examples)
|
||||
详细引入方式请查看 [文档](http://doc.numberone.ink/install.html)、或查看[示例代码](https://gitee.com/dengwenjie/SmartJavaAI/tree/master/examples)
|
||||
|
||||
```xml
|
||||
<dependency>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>all</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</dependency>
|
||||
```
|
||||
|
||||
@@ -548,7 +575,7 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
|
||||
### 4、文档地址
|
||||
|
||||
[开发文档](http://doc.smartjavaai.cn)
|
||||
[开发文档](http://doc.numberone.ink)
|
||||
|
||||
### 5、模型简介及下载
|
||||
|
||||
@@ -876,4 +903,3 @@ SmartJavaAI是专为JAVA 开发者打造的一个功能丰富、开箱即用的
|
||||
6、等待维护者合并
|
||||
|
||||
|
||||
|
||||
|
||||
@@ -6,11 +6,11 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>all</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>${project.artifactId}</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
@@ -6,10 +6,10 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<artifactId>bom</artifactId>
|
||||
<name>bom</name>
|
||||
<description>统一版本管理的 BOM 包，同时支持 import 和全量依赖</description>
|
||||
|
||||
@@ -6,7 +6,7 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<name>common</name>
|
||||
|
||||
@@ -59,6 +59,6 @@ src/main/java/smartai/examples/face/
|
||||
## 📄 文档
|
||||
|
||||
有关完整使用说明，请查阅 SmartJavaAI 官方文档：
|
||||
[http://doc.smartjavaai.cn](http://doc.smartjavaai.cn)
|
||||
[http://doc.numberone.ink](http://doc.numberone.ink)
|
||||
|
||||
---
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
<maven.compiler.source>11</maven.compiler.source>
|
||||
<maven.compiler.target>11</maven.compiler.target>
|
||||
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||
<smartjavaai.version>1.1.1</smartjavaai.version>
|
||||
<smartjavaai.version>1.1.2</smartjavaai.version>
|
||||
<!--如果打包运行，需要替换成你的main-->
|
||||
<exec.mainClass>smartai.examples.face.FaceDemo</exec.mainClass>
|
||||
|
||||
|
||||
@@ -33,7 +33,7 @@ import java.util.List;
|
||||
/**
|
||||
* 人脸属性检测demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -51,7 +51,7 @@ import java.util.List;
|
||||
* 表情识别demo
|
||||
* 支持识别7种表情：neutral（中性）、happy（高兴）、sad（悲伤）、surprise（惊讶）、fear（恐惧）、disgust（厌恶）、anger（愤怒）
|
||||
* 模型下载地址：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -43,7 +43,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 人脸检测模型demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/1d2YlJ2YOdGn3Y-AegyAhmQ?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -37,7 +37,7 @@ import java.util.List;
|
||||
/**
|
||||
* 人脸识别模型demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -78,7 +78,7 @@ public class FaceRecDemo {
|
||||
* 获取人脸检测模型（高精度模型）
|
||||
* 注意事项：
|
||||
* 1、高精度模型，识别准确度高，速度慢
|
||||
* 2、具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
* 2、具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
* @return
|
||||
*/
|
||||
public FaceDetModel getProFaceDetModel(){
|
||||
@@ -98,7 +98,7 @@ public class FaceRecDemo {
|
||||
* 获取人脸检测模型（极速模型）
|
||||
* 注意事项：
|
||||
* 1、极速模型，识别准确度低，速度快
|
||||
* 2、具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
* 2、具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
* @return
|
||||
*/
|
||||
public FaceDetModel getFastFaceDetModel(){
|
||||
@@ -118,7 +118,7 @@ public class FaceRecDemo {
|
||||
/**
|
||||
* 获取人脸识别模型（高精度，速度慢）
|
||||
* 追求准确度可以使用
|
||||
* 也可以使用其他模型，具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
* 也可以使用其他模型，具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
* @return
|
||||
*/
|
||||
public FaceRecModel getFaceRecModel(){
|
||||
@@ -140,7 +140,7 @@ public class FaceRecDemo {
|
||||
/**
|
||||
* 获取人脸识别模型（高速模型，精度一般）
|
||||
* 追求速度可以使用
|
||||
* 也可以使用其他模型，具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
* 也可以使用其他模型，具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
* @return
|
||||
*/
|
||||
public FaceRecModel getHighSpeedFaceRecModel(){
|
||||
@@ -165,7 +165,7 @@ public class FaceRecDemo {
|
||||
*/
|
||||
public FaceRecModel getFaceRecModelWithDbConfig(){
|
||||
FaceRecConfig config = new FaceRecConfig();
|
||||
//高精度模型，速度慢,追求速度请更换高速模型，具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
//高精度模型，速度慢,追求速度请更换高速模型，具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
config.setModelEnum(FaceRecModelEnum.INSIGHT_FACE_IRSE50_MODEL);
|
||||
//模型路径，请下载模型并替换为本地路径：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
config.setModelPath("/Users/wenjie/Documents/develop/model/face_model/recognition/InsightFace/model_ir_se50.pt");
|
||||
@@ -173,7 +173,7 @@ public class FaceRecDemo {
|
||||
config.setCropFace(true);
|
||||
//开启人脸对齐：适用于人脸不正的场景，开启将提升人脸特征准确度，关闭可以提升性能
|
||||
config.setAlign(true);
|
||||
//指定人脸检测模型，可切换人脸检测模型（极速：getFastFaceDetModel，高精度：getProFaceDetModel），具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
//指定人脸检测模型，可切换人脸检测模型（极速：getFastFaceDetModel，高精度：getProFaceDetModel），具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
config.setDetectModel(getFaceDetModel());
|
||||
config.setDevice(device);
|
||||
|
||||
@@ -198,7 +198,7 @@ public class FaceRecDemo {
|
||||
*/
|
||||
public FaceRecModel getFaceRecModelWithSQLiteConfig(){
|
||||
FaceRecConfig config = new FaceRecConfig();
|
||||
//高精度模型，速度慢, 追求速度请更换高速模型，具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
//高精度模型，速度慢, 追求速度请更换高速模型，具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
config.setModelEnum(FaceRecModelEnum.INSIGHT_FACE_IRSE50_MODEL);
|
||||
//模型路径，请下载模型并替换为本地路径：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
config.setModelPath("/Users/wenjie/Documents/develop/model/face_model/recognition/InsightFace/model_ir_se50.pt");
|
||||
@@ -206,7 +206,7 @@ public class FaceRecDemo {
|
||||
config.setCropFace(true);
|
||||
//开启人脸对齐：适用于人脸不正的场景，开启将提升人脸特征准确度，关闭可以提升性能
|
||||
config.setAlign(true);
|
||||
//指定人脸检测模型，可切换人脸检测模型（极速：getFastFaceDetModel，高精度：getProFaceDetModel），具体其他模型参数可以查看文档：http://doc.smartjavaai.cn/face.html
|
||||
//指定人脸检测模型，可切换人脸检测模型（极速：getFastFaceDetModel，高精度：getProFaceDetModel），具体其他模型参数可以查看文档：http://doc.numberone.ink/face.html
|
||||
config.setDetectModel(getFaceDetModel());
|
||||
config.setDevice(device);
|
||||
|
||||
|
||||
@@ -52,7 +52,7 @@ import java.util.List;
|
||||
/**
|
||||
* 静态活体检测demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -38,7 +38,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 人脸质量评估 demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/10l22x5fRz_gwLr8EAHa1Jg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -61,6 +61,6 @@ src
|
||||
## 📄 文档
|
||||
|
||||
有关完整使用说明，请查阅 SmartJavaAI 官方文档：
|
||||
[http://doc.smartjavaai.cn](http://doc.smartjavaai.cn)
|
||||
[http://doc.numberone.ink](http://doc.numberone.ink)
|
||||
|
||||
---
|
||||
|
||||
@@ -12,9 +12,9 @@
|
||||
<maven.compiler.source>11</maven.compiler.source>
|
||||
<maven.compiler.target>11</maven.compiler.target>
|
||||
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||
<smartjavaai.version>1.1.1</smartjavaai.version>
|
||||
<smartjavaai.version>1.1.2</smartjavaai.version>
|
||||
<!--如果打包运行，需要替换成你的main-->
|
||||
<exec.mainClass>smartai.examples.ocr.common.OcrRecognizeDemo</exec.mainClass>
|
||||
<exec.mainClass>smartai.examples.ocr.common.OcrDetectionDemo</exec.mainClass>
|
||||
|
||||
<javacv.version>1.5.10</javacv.version>
|
||||
|
||||
@@ -90,19 +90,33 @@
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>ocr</artifactId>
|
||||
<exclusions>
|
||||
<exclusion>
|
||||
<groupId>com.microsoft.onnxruntime</groupId>
|
||||
<artifactId>onnxruntime</artifactId>
|
||||
</exclusion>
|
||||
<!-- <exclusion>-->
|
||||
<!-- <groupId>org.openpnp</groupId>-->
|
||||
<!-- <artifactId>opencv</artifactId>-->
|
||||
<!-- </exclusion>-->
|
||||
<!-- <exclusion>-->
|
||||
<!-- <groupId>com.microsoft.onnxruntime</groupId>-->
|
||||
<!-- <artifactId>onnxruntime</artifactId>-->
|
||||
<!-- </exclusion>-->
|
||||
<!-- <exclusion>-->
|
||||
<!-- <groupId>org.bytedeco</groupId>-->
|
||||
<!-- <artifactId>javacv</artifactId>-->
|
||||
<!-- </exclusion>-->
|
||||
</exclusions>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>com.microsoft.onnxruntime</groupId>
|
||||
<artifactId>onnxruntime</artifactId>
|
||||
<version>1.20.0</version>
|
||||
<scope>runtime</scope>
|
||||
</dependency>
|
||||
<!-- <dependency>-->
|
||||
<!-- <groupId>com.microsoft.onnxruntime</groupId>-->
|
||||
<!-- <artifactId>onnxruntime</artifactId>-->
|
||||
<!-- <version>1.16.3</version>-->
|
||||
<!-- <scope>compile</scope>-->
|
||||
<!-- </dependency>-->
|
||||
|
||||
<!-- <dependency>-->
|
||||
<!-- <groupId>org.openpnp</groupId>-->
|
||||
<!-- <artifactId>opencv</artifactId>-->
|
||||
<!-- <version>3.4.2-2</version>-->
|
||||
<!-- </dependency>-->
|
||||
|
||||
|
||||
<dependency>
|
||||
@@ -113,79 +127,10 @@
|
||||
</dependency>
|
||||
|
||||
|
||||
<!-- windows平台 (保留对应平台的配置，可以减小包大小)-->
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>javacpp</artifactId>
|
||||
<version>${javacv.version}</version>
|
||||
<classifier>${javacv.platform.windows-x86_64}</classifier>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>ffmpeg</artifactId>
|
||||
<version>6.1.1-1.5.10</version>
|
||||
<classifier>${javacv.platform.windows-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>openblas</artifactId>
|
||||
<version>0.3.26-1.5.10</version>
|
||||
<classifier>${javacv.platform.windows-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>opencv</artifactId>
|
||||
<version>4.9.0-1.5.10</version>
|
||||
<classifier>${javacv.platform.windows-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>ai.djl.pytorch</groupId>
|
||||
<artifactId>pytorch-native-cpu</artifactId>
|
||||
<classifier>${djl.platform.windows-x86_64}</classifier>
|
||||
<version>2.7.1</version>
|
||||
<scope>runtime</scope>
|
||||
</dependency>
|
||||
|
||||
|
||||
|
||||
<!-- linux x86 平台 (保留对应平台的配置，可以减小包大小)-->
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>javacpp</artifactId>
|
||||
<version>${javacv.version}</version>
|
||||
<classifier>${javacv.platform.linux-x86_64}</classifier>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>ffmpeg</artifactId>
|
||||
<version>6.1.1-1.5.10</version>
|
||||
<classifier>${javacv.platform.linux-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>openblas</artifactId>
|
||||
<version>0.3.26-1.5.10</version>
|
||||
<classifier>${javacv.platform.linux-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>opencv</artifactId>
|
||||
<version>4.9.0-1.5.10</version>
|
||||
<classifier>${javacv.platform.linux-x86_64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>ai.djl.pytorch</groupId>
|
||||
<artifactId>pytorch-native-cpu</artifactId>
|
||||
<classifier>${djl.platform.linux-x86_64}</classifier>
|
||||
<version>2.7.1</version>
|
||||
<scope>runtime</scope>
|
||||
</dependency>
|
||||
|
||||
<!-- linux aarch64 平台 (保留对应平台的配置，可以减小包大小)-->
|
||||
<dependency>
|
||||
@@ -223,42 +168,16 @@
|
||||
<scope>runtime</scope>
|
||||
</dependency>
|
||||
|
||||
<!-- <dependency>-->
|
||||
<!-- <groupId>ai.djl.pytorch</groupId>-->
|
||||
<!-- <artifactId>pytorch-native-cpu-precxx11</artifactId>-->
|
||||
<!-- <classifier>linux-aarch64</classifier>-->
|
||||
<!-- <version>2.5.1</version>-->
|
||||
<!-- <scope>runtime</scope>-->
|
||||
<!-- </dependency>-->
|
||||
|
||||
<!-- macOS M系列 平台 (保留对应平台的配置，可以减小包大小)-->
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>javacpp</artifactId>
|
||||
<version>${javacv.version}</version>
|
||||
<classifier>${javacv.platform.macosx-arm64}</classifier>
|
||||
</dependency>
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>ffmpeg</artifactId>
|
||||
<version>6.1.1-1.5.10</version>
|
||||
<classifier>${javacv.platform.macosx-arm64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>openblas</artifactId>
|
||||
<version>0.3.26-1.5.10</version>
|
||||
<classifier>${javacv.platform.macosx-arm64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>org.bytedeco</groupId>
|
||||
<artifactId>opencv</artifactId>
|
||||
<version>4.9.0-1.5.10</version>
|
||||
<classifier>${javacv.platform.macosx-arm64}</classifier>
|
||||
</dependency>
|
||||
|
||||
<dependency>
|
||||
<groupId>ai.djl.pytorch</groupId>
|
||||
<artifactId>pytorch-native-cpu</artifactId>
|
||||
<classifier>${djl.platform.osx-aarch64}</classifier>
|
||||
<version>2.7.1</version>
|
||||
<scope>runtime</scope>
|
||||
</dependency>
|
||||
|
||||
|
||||
</dependencies>
|
||||
|
||||
@@ -30,7 +30,7 @@ import java.util.List;
|
||||
/**
|
||||
* OCR 文本检测 示例
|
||||
* 模型下载地址：https://pan.baidu.com/s/15Noz2xHQzqMQSl1B19BobQ?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -25,7 +25,7 @@ import java.util.List;
|
||||
/**
|
||||
* OCR 行文本方向检测 示例
|
||||
* 模型下载地址：https://pan.baidu.com/s/1MLfd73Vjdpnuls9-oqc9uw?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -33,12 +33,18 @@ import java.io.File;
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
import java.util.concurrent.CountDownLatch;
|
||||
import java.util.concurrent.ExecutorService;
|
||||
import java.util.concurrent.Executors;
|
||||
import java.util.concurrent.TimeUnit;
|
||||
import java.util.concurrent.atomic.AtomicInteger;
|
||||
|
||||
/**
|
||||
* OCR 文本识别 示例
|
||||
* 模型下载地址：https://pan.baidu.com/s/1MLfd73Vjdpnuls9-oqc9uw?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -272,6 +278,78 @@ public class OcrRecognizeDemo {
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* 多线程文本识别
|
||||
* 注意事项：
|
||||
* 1、模型在外层统一创建，多个线程共享同一个模型实例，避免重复加载模型。
|
||||
* 2、每个线程内部单独创建 Image 对象，避免共享图片对象带来的线程安全问题。
|
||||
*/
|
||||
@Test
|
||||
public void multiThreadRecognize() {
|
||||
ExecutorService executorService = null;
|
||||
try {
|
||||
final OcrCommonRecModel recModel = getFastRecModel();
|
||||
final OcrRecOptions options = new OcrRecOptions(false, true);
|
||||
final String imagePath = "src/main/resources/ocr_1.jpg";
|
||||
|
||||
int threadCount = 5;
|
||||
int taskCount = 20;
|
||||
CountDownLatch countDownLatch = new CountDownLatch(taskCount);
|
||||
AtomicInteger successCount = new AtomicInteger(0);
|
||||
AtomicInteger failCount = new AtomicInteger(0);
|
||||
List<String> errorMessages = new ArrayList<>();
|
||||
executorService = Executors.newFixedThreadPool(threadCount);
|
||||
|
||||
long startTime = System.currentTimeMillis();
|
||||
for (int i = 0; i < taskCount; i++) {
|
||||
final int taskIndex = i;
|
||||
executorService.submit(() -> {
|
||||
try {
|
||||
Image image = SmartImageFactory.getInstance().fromFile(imagePath);
|
||||
OcrInfo ocrInfo = recModel.recognize(image, options);
|
||||
successCount.incrementAndGet();
|
||||
log.info("线程：{}，任务：{}，识别结果：{}",
|
||||
Thread.currentThread().getName(),
|
||||
taskIndex,
|
||||
JSONObject.toJSONString(ocrInfo));
|
||||
} catch (Exception e) {
|
||||
failCount.incrementAndGet();
|
||||
String errorMsg = "任务" + taskIndex + "识别失败：" + e.getMessage();
|
||||
synchronized (errorMessages) {
|
||||
errorMessages.add(errorMsg);
|
||||
}
|
||||
log.error(errorMsg, e);
|
||||
} finally {
|
||||
countDownLatch.countDown();
|
||||
}
|
||||
});
|
||||
}
|
||||
|
||||
countDownLatch.await();
|
||||
long costTime = System.currentTimeMillis() - startTime;
|
||||
log.info("多线程识别完成，总任务数：{}，成功：{}，失败：{}，耗时：{} ms",
|
||||
taskCount, successCount.get(), failCount.get(), costTime);
|
||||
|
||||
if (!errorMessages.isEmpty()) {
|
||||
log.error("失败详情：{}", JsonUtils.toJson(errorMessages));
|
||||
}
|
||||
} catch (Exception e) {
|
||||
e.printStackTrace();
|
||||
} finally {
|
||||
if (executorService != null) {
|
||||
executorService.shutdown();
|
||||
try {
|
||||
if (!executorService.awaitTermination(10, TimeUnit.SECONDS)) {
|
||||
executorService.shutdownNow();
|
||||
}
|
||||
} catch (InterruptedException e) {
|
||||
executorService.shutdownNow();
|
||||
Thread.currentThread().interrupt();
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
@@ -0,0 +1,216 @@
|
||||
package smartai.examples.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import ai.djl.util.JsonUtils;
|
||||
import cn.smartjavaai.common.config.Config;
|
||||
import cn.smartjavaai.common.cv.SmartImageFactory;
|
||||
import cn.smartjavaai.common.enums.DeviceEnum;
|
||||
import cn.smartjavaai.common.utils.ImageUtils;
|
||||
import cn.smartjavaai.ocr.config.DirectionModelConfig;
|
||||
import cn.smartjavaai.ocr.config.OcrDetModelConfig;
|
||||
import cn.smartjavaai.ocr.config.OcrRecModelConfig;
|
||||
import cn.smartjavaai.ocr.config.OcrRecOptions;
|
||||
import cn.smartjavaai.ocr.entity.IdCardBackInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardFrontInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
import cn.smartjavaai.ocr.enums.CommonDetModelEnum;
|
||||
import cn.smartjavaai.ocr.enums.CommonRecModelEnum;
|
||||
import cn.smartjavaai.ocr.enums.DirectionModelEnum;
|
||||
import cn.smartjavaai.ocr.factory.OcrModelFactory;
|
||||
import cn.smartjavaai.ocr.idcard.DefaultIdCardRecognizer;
|
||||
import cn.smartjavaai.ocr.idcard.IdCardPreprocessListener;
|
||||
import cn.smartjavaai.ocr.model.common.detect.OcrCommonDetModel;
|
||||
import cn.smartjavaai.ocr.model.common.direction.OcrDirectionModel;
|
||||
import cn.smartjavaai.ocr.model.common.recognize.OcrCommonRecModel;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
import org.junit.BeforeClass;
|
||||
import org.junit.Test;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.nio.file.Files;
|
||||
import java.nio.file.Path;
|
||||
import java.nio.file.Paths;
|
||||
|
||||
/**
|
||||
* 身份证识别 demo
|
||||
* 使用说明：
|
||||
* 1、先下载 OCR 模型
|
||||
* 2、把下面的模型路径改成你自己的本地路径
|
||||
* 3、把身份证图片路径改成你自己的图片路径
|
||||
* 4、优先运行 recognizeFront() / recognizeBack() 查看结构化结果
|
||||
*
|
||||
* 模型下载地址：https://pan.baidu.com/s/1MLfd73Vjdpnuls9-oqc9uw?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
public class IdCardRecDemo {
|
||||
|
||||
// 设备类型
|
||||
public static DeviceEnum device = DeviceEnum.CPU;
|
||||
|
||||
// 下载模型后，请替换成你自己的模型路径
|
||||
private static final String DET_MODEL_PATH =
|
||||
"/Users/wenjie/Documents/develop/model/ocr/PP-OCRv5_server_det_infer/PP-OCRv5_server_det.onnx";
|
||||
private static final String REC_MODEL_PATH =
|
||||
"/Users/wenjie/Documents/develop/model/ocr/PP-OCRv5_server_rec_infer/PP-OCRv5_server_rec.onnx";
|
||||
private static final String DIRECTION_MODEL_PATH =
|
||||
"/Users/wenjie/Documents/develop/model/ocr/PP-LCNet_x0_25_textline_ori_infer/PP-LCNet_x0_25_textline_ori_infer.onnx";
|
||||
|
||||
// 这里改成你自己的身份证图片路径
|
||||
private static final String FRONT_IMAGE_PATH = "src/main/resources/idcard/idcard_front1.png";
|
||||
private static final String BACK_IMAGE_PATH = "src/main/resources/idcard/idcard_back1.png";
|
||||
|
||||
@BeforeClass
|
||||
public static void beforeAll() throws IOException {
|
||||
// 修改缓存路径
|
||||
// Config.setCachePath("/Users/xxx/smartjavaai_cache");
|
||||
}
|
||||
|
||||
/**
|
||||
* 获取文本检测模型
|
||||
* @return
|
||||
*/
|
||||
public OcrCommonDetModel getDetectionModel() {
|
||||
OcrDetModelConfig config = new OcrDetModelConfig();
|
||||
// 文本检测模型，切换模型需要同时修改 modelEnum 及 modelPath
|
||||
config.setModelEnum(CommonDetModelEnum.PP_OCR_V4_SERVER_DET_MODEL);
|
||||
// 下载模型并替换本地路径
|
||||
config.setDetModelPath(DET_MODEL_PATH);
|
||||
config.setDevice(device);
|
||||
return OcrModelFactory.getInstance().getDetModel(config);
|
||||
}
|
||||
|
||||
/**
|
||||
* 获取方向检测模型
|
||||
* @return
|
||||
*/
|
||||
public OcrDirectionModel getDirectionModel() {
|
||||
DirectionModelConfig directionModelConfig = new DirectionModelConfig();
|
||||
// 行文本方向检测模型，切换模型需要同时修改 modelEnum 及 modelPath
|
||||
directionModelConfig.setModelEnum(DirectionModelEnum.PP_LCNET_X0_25);
|
||||
// 下载模型并替换本地路径
|
||||
directionModelConfig.setModelPath(DIRECTION_MODEL_PATH);
|
||||
directionModelConfig.setTextDetModel(getDetectionModel());
|
||||
directionModelConfig.setDevice(device);
|
||||
return OcrModelFactory.getInstance().getDirectionModel(directionModelConfig);
|
||||
}
|
||||
|
||||
/**
|
||||
* 获取 OCR 识别模型
|
||||
* @return
|
||||
*/
|
||||
public OcrCommonRecModel getRecModel() {
|
||||
OcrRecModelConfig recModelConfig = new OcrRecModelConfig();
|
||||
// 文本识别模型，切换模型需要同时修改 modelEnum 及 modelPath
|
||||
recModelConfig.setRecModelEnum(CommonRecModelEnum.PP_OCR_V5_MOBILE_REC_MODEL);
|
||||
// 下载模型并替换本地路径
|
||||
recModelConfig.setRecModelPath(REC_MODEL_PATH);
|
||||
recModelConfig.setTextDetModel(getDetectionModel());
|
||||
recModelConfig.setDevice(device);
|
||||
return OcrModelFactory.getInstance().getRecModel(recModelConfig);
|
||||
}
|
||||
|
||||
/**
|
||||
* 创建身份证识别器
|
||||
* @return
|
||||
*/
|
||||
public DefaultIdCardRecognizer getIdCardRecognizer() {
|
||||
return new DefaultIdCardRecognizer()
|
||||
.setRecModel(getRecModel())
|
||||
// 身份证识别建议按行返回，方向矫正交给预处理逻辑处理
|
||||
.setRecOptions(new OcrRecOptions(false, true))
|
||||
.setDirectionModel(getDirectionModel())
|
||||
.setEnablePreprocess(true);
|
||||
}
|
||||
|
||||
/**
|
||||
* 身份证正面识别
|
||||
* 可识别：姓名、性别、民族、出生日期、住址、身份证号
|
||||
*/
|
||||
@Test
|
||||
public void recognizeFront() {
|
||||
try {
|
||||
DefaultIdCardRecognizer recognizer = getIdCardRecognizer();
|
||||
Image image = SmartImageFactory.getInstance().fromFile(FRONT_IMAGE_PATH);
|
||||
IdCardFrontInfo result = recognizer.recognizeFront(image);
|
||||
log.info("身份证正面识别结果：{}", JsonUtils.toJson(result));
|
||||
} catch (Exception e) {
|
||||
log.error("身份证正面识别失败", e);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* 身份证反面识别
|
||||
* 可识别：签发机关、有效期开始时间、有效期结束时间
|
||||
*/
|
||||
@Test
|
||||
public void recognizeBack() {
|
||||
try {
|
||||
DefaultIdCardRecognizer recognizer = getIdCardRecognizer();
|
||||
Image image = SmartImageFactory.getInstance().fromFile(BACK_IMAGE_PATH);
|
||||
IdCardBackInfo result = recognizer.recognizeBack(image);
|
||||
log.info("身份证反面识别结果：{}", JsonUtils.toJson(result));
|
||||
} catch (Exception e) {
|
||||
log.error("身份证反面识别失败", e);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* 同时识别身份证正反面
|
||||
*/
|
||||
@Test
|
||||
public void recognizeBoth() {
|
||||
try {
|
||||
DefaultIdCardRecognizer recognizer = getIdCardRecognizer();
|
||||
Image frontImage = SmartImageFactory.getInstance().fromFile(FRONT_IMAGE_PATH);
|
||||
Image backImage = SmartImageFactory.getInstance().fromFile(BACK_IMAGE_PATH);
|
||||
IdCardInfo result = recognizer.recognizeBoth(frontImage, backImage);
|
||||
log.info("身份证正反面识别结果：{}", JsonUtils.toJson(result));
|
||||
} catch (Exception e) {
|
||||
log.error("身份证正反面识别失败", e);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* 身份证识别并输出调试图片
|
||||
* 适合排查“为什么没识别出来”，不适合生产环境使用
|
||||
*/
|
||||
@Test
|
||||
public void recognizeFrontWithDebugImages() {
|
||||
try {
|
||||
Path debugDir = Paths.get("output/idcard_debug");
|
||||
Files.createDirectories(debugDir);
|
||||
String filePrefix = "idcard_front";
|
||||
DefaultIdCardRecognizer recognizer = getIdCardRecognizer()
|
||||
// .setPreprocessListener(createDebugListener(debugDir, "idcard_front"));
|
||||
.setPreprocessListener(new IdCardPreprocessListener() {
|
||||
@Override
|
||||
public void onAfterDirection(Image image) {
|
||||
saveDebugImage(image, debugDir.resolve(filePrefix + "_step1_direction.png"));
|
||||
}
|
||||
|
||||
@Override
|
||||
public void onAfterRecognize(Image image, OcrInfo ocrInfo) {
|
||||
saveDebugImage(image, debugDir.resolve(filePrefix + "_step2_ocr_boxes.png"));
|
||||
}
|
||||
});
|
||||
|
||||
Image image = SmartImageFactory.getInstance().fromFile(FRONT_IMAGE_PATH);
|
||||
IdCardFrontInfo result = recognizer.recognizeFront(image);
|
||||
log.info("身份证正面识别结果：{}", JsonUtils.toJson(result));
|
||||
log.info("调试图片已输出到：{}", debugDir.toAbsolutePath());
|
||||
} catch (Exception e) {
|
||||
log.error("身份证调试识别失败", e);
|
||||
}
|
||||
}
|
||||
|
||||
private void saveDebugImage(Image image, Path outputPath) {
|
||||
try {
|
||||
ImageUtils.save(image, outputPath, "png");
|
||||
} catch (IOException e) {
|
||||
throw new RuntimeException(e);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -29,7 +29,7 @@ import java.util.List;
|
||||
/**
|
||||
* 车牌识别demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/1YEP56UqYcL-Op80M6JAreA?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -39,7 +39,7 @@ import java.util.List;
|
||||
/**
|
||||
* OCR 表格识别 示例
|
||||
* 模型下载地址：https://pan.baidu.com/s/1MLfd73Vjdpnuls9-oqc9uw?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
BIN
examples/ocr-examples/src/main/resources/idcard/idcard_back1.png
Normal file
BIN
examples/ocr-examples/src/main/resources/idcard/idcard_back1.png
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 997 KiB |
Binary file not shown.
|
After Width: | Height: | Size: 418 KiB |
@@ -33,6 +33,6 @@ src
|
||||
## 📄 文档
|
||||
|
||||
有关完整使用说明，请查阅 SmartJavaAI 官方文档：
|
||||
[http://doc.smartjavaai.cn](http://doc.smartjavaai.cn)
|
||||
[http://doc.numberone.ink](http://doc.numberone.ink)
|
||||
|
||||
---
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
<maven.compiler.source>11</maven.compiler.source>
|
||||
<maven.compiler.target>11</maven.compiler.target>
|
||||
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||
<smartjavaai.version>1.1.1</smartjavaai.version>
|
||||
<smartjavaai.version>1.1.2</smartjavaai.version>
|
||||
<!--如果打包运行，需要替换成你的main-->
|
||||
<exec.mainClass>smartai.examples.speech.asr.common.OcrRecognizeDemo</exec.mainClass>
|
||||
|
||||
|
||||
@@ -35,7 +35,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 语音识别demo（Vosk、Whisper）
|
||||
* 模型下载网盘：https://pan.baidu.com/s/1kiMF5MF641R7LTn1GpB2lQ?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -40,6 +40,6 @@
|
||||
## 📄 文档
|
||||
|
||||
有关完整使用说明，请查阅 SmartJavaAI 官方文档：
|
||||
[http://doc.smartjavaai.cn](http://doc.smartjavaai.cn)
|
||||
[http://doc.numberone.ink](http://doc.numberone.ink)
|
||||
|
||||
---
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
<maven.compiler.source>11</maven.compiler.source>
|
||||
<maven.compiler.target>11</maven.compiler.target>
|
||||
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||
<smartjavaai.version>1.1.1</smartjavaai.version>
|
||||
<smartjavaai.version>1.1.2</smartjavaai.version>
|
||||
<!--如果打包运行，需要替换成你的main-->
|
||||
<exec.mainClass>smartai.examples.nlp.translation.TranslationDemo</exec.mainClass>
|
||||
|
||||
|
||||
@@ -20,7 +20,7 @@ import java.io.IOException;
|
||||
/**
|
||||
* 机器翻译Demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/1wf7btnb4cyBFv7DB7baHnw?pwd=1234 提取码: 1234
|
||||
* 开发文档：http://doc.smartjavaai.cn/
|
||||
* 开发文档：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -96,6 +96,6 @@ vision-example/
|
||||
## 📄 文档
|
||||
|
||||
有关完整使用说明，请查阅 SmartJavaAI 官方文档：
|
||||
[http://doc.smartjavaai.cn](http://doc.smartjavaai.cn)
|
||||
[http://doc.numberone.ink](http://doc.numberone.ink)
|
||||
|
||||
---
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
<maven.compiler.source>11</maven.compiler.source>
|
||||
<maven.compiler.target>11</maven.compiler.target>
|
||||
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
|
||||
<smartjavaai.version>1.1.1</smartjavaai.version>
|
||||
<smartjavaai.version>1.1.2</smartjavaai.version>
|
||||
<!--如果打包运行，需要替换成你的main-->
|
||||
<exec.mainClass>smartai.examples.vision.ObjectDetectionDemo</exec.mainClass>
|
||||
|
||||
|
||||
@@ -21,7 +21,7 @@ import java.util.Arrays;
|
||||
/**
|
||||
* 动作识别Demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/17doY4pgZM9EbtSIaoCWWCA?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -41,7 +41,7 @@ public class ActionRecognizeDemo {
|
||||
/**
|
||||
* 获取动作识别模型
|
||||
* 注意事项：
|
||||
* 1、不同模型支持的动作类别不同，请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、不同模型支持的动作类别不同，请查看文档：http://doc.numberone.ink
|
||||
*/
|
||||
public ActionRecModel getModel(){
|
||||
ActionRecModelConfig config = new ActionRecModelConfig();
|
||||
@@ -62,7 +62,7 @@ public class ActionRecognizeDemo {
|
||||
/**
|
||||
* 动作识别
|
||||
* 注意事项：
|
||||
* 1、不同模型支持的动作类别不同，请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、不同模型支持的动作类别不同，请查看文档：http://doc.numberone.ink
|
||||
* 2、图片中应该只包含单一动作人物
|
||||
* 3、动作识别，只做图片分类，并不做人物定位
|
||||
*/
|
||||
|
||||
@@ -28,7 +28,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 实例分割 Demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/12nRRY9JFNDwLeg63jfBerA?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -50,7 +50,7 @@ public class InstanceSegDemo {
|
||||
/**
|
||||
* 获取实例分割模型
|
||||
* 注意事项：
|
||||
* 1、更多模型请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、更多模型请查看文档：http://doc.numberone.ink
|
||||
*/
|
||||
public InstanceSegModel getModel(){
|
||||
InstanceSegModelConfig config = new InstanceSegModelConfig();
|
||||
|
||||
@@ -25,7 +25,7 @@ import java.util.Arrays;
|
||||
/**
|
||||
* obb旋转框检测demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/1-tC0u-aha3tnMQwy8FKy1Q?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -45,7 +45,7 @@ public class ObbDetDemo {
|
||||
/**
|
||||
* 获取旋转框检测模型
|
||||
* 注意事项：
|
||||
* 1、更多模型请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、更多模型请查看文档：http://doc.numberone.ink
|
||||
* 2、模型可检测物体请查看：模型同目录文件synset.txt
|
||||
*/
|
||||
public ObbDetModel getModel(){
|
||||
|
||||
@@ -45,7 +45,7 @@ import java.util.concurrent.CountDownLatch;
|
||||
/**
|
||||
* 目标检测模型demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/10aTOLBlR6EG-sq6g0OkAWg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -66,7 +66,7 @@ public class ObjectDetectionDemo {
|
||||
/**
|
||||
* 获取目标检测模型
|
||||
* 注意事项：
|
||||
* 1、更多模型请查看文档：http://doc.smartjavaai.cn/objectdetect.html
|
||||
* 1、更多模型请查看文档：http://doc.numberone.ink/objectdetect.html
|
||||
*/
|
||||
public DetectorModel getModel(){
|
||||
DetectorModelConfig config = new DetectorModelConfig();
|
||||
@@ -147,7 +147,7 @@ public class ObjectDetectionDemo {
|
||||
DetectorModelConfig config = new DetectorModelConfig();
|
||||
//目标检测模型，切换模型需要同时修改modelEnum及modelPath
|
||||
config.setModelEnum(DetectorModelEnum.YOLOV12_CUSTOM_ONNX);
|
||||
//模型所在路径，synset.txt也需要放在同目录下(分类文件，具体请看文档：http://doc.smartjavaai.cn/objectdetect.html#%E4%BD%BF%E7%94%A8%E8%87%AA%E5%B7%B1%E8%AE%AD%E7%BB%83%E7%9A%84%E6%A8%A1%E5%9E%8B%E6%A3%80%E6%B5%8B)
|
||||
//模型所在路径，synset.txt也需要放在同目录下(分类文件，具体请看文档：http://doc.numberone.ink/objectdetect.html#%E4%BD%BF%E7%94%A8%E8%87%AA%E5%B7%B1%E8%AE%AD%E7%BB%83%E7%9A%84%E6%A8%A1%E5%9E%8B%E6%A3%80%E6%B5%8B)
|
||||
config.setModelPath("/Users/xxx/Documents/develop/fire_model/best.onnx");
|
||||
//模型训练时图片宽度
|
||||
config.putCustomParam("width", 640);//resize 宽
|
||||
|
||||
@@ -21,7 +21,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 行人检测案例
|
||||
* 模型下载地址：https://pan.baidu.com/s/1EWfExw7pYjKEH5uR5wf3Rw?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
|
||||
@@ -22,7 +22,7 @@ import java.nio.file.Paths;
|
||||
/**
|
||||
* 姿态估计demo
|
||||
* 模型下载地址：https://pan.baidu.com/s/1pPYyl1V2CpcMYCO8CJQHGg?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -42,7 +42,7 @@ public class PoseDetDemo {
|
||||
/**
|
||||
* 获取姿态估计模型
|
||||
* 注意事项：
|
||||
* 1、更多模型请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、更多模型请查看文档：http://doc.numberone.ink
|
||||
*/
|
||||
public PoseModel getModel(){
|
||||
PoseModelConfig config = new PoseModelConfig();
|
||||
|
||||
@@ -23,7 +23,7 @@ import java.util.Arrays;
|
||||
/**
|
||||
* 语义分割 Demo 通过网盘分享的文件：语义分割（semantic_segmentation）
|
||||
* 模型下载地址：https://pan.baidu.com/s/18gs9E5h_d9imPmNLHuDo9A?pwd=1234 提取码: 1234
|
||||
* 文档地址：http://doc.smartjavaai.cn/
|
||||
* 文档地址：http://doc.numberone.ink/
|
||||
* @author dwj
|
||||
*/
|
||||
@Slf4j
|
||||
@@ -45,7 +45,7 @@ public class SemSegDemo {
|
||||
/**
|
||||
* 获取语义分割模型
|
||||
* 注意事项：
|
||||
* 1、更多模型请查看文档：http://doc.smartjavaai.cn
|
||||
* 1、更多模型请查看文档：http://doc.numberone.ink
|
||||
*/
|
||||
public SemSegModel getModel(){
|
||||
SemSegModelConfig config = new SemSegModelConfig();
|
||||
|
||||
@@ -6,11 +6,11 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>face</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>face</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
@@ -6,7 +6,7 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>ocr</artifactId>
|
||||
@@ -42,7 +42,7 @@
|
||||
</dependency>
|
||||
</dependencies>
|
||||
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>ocr</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
@@ -21,6 +21,7 @@ public class OcrRecOptions {
|
||||
private boolean enableLineSplit = true;
|
||||
|
||||
|
||||
|
||||
public OcrRecOptions(boolean enableDirectionCorrect, boolean enableLineSplit) {
|
||||
this.enableDirectionCorrect = enableDirectionCorrect;
|
||||
this.enableLineSplit = enableLineSplit;
|
||||
|
||||
@@ -0,0 +1,26 @@
|
||||
package cn.smartjavaai.ocr.entity;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
/**
|
||||
* 身份证反面信息
|
||||
*/
|
||||
@Data
|
||||
public class IdCardBackInfo {
|
||||
|
||||
/**
|
||||
* 签发机关
|
||||
*/
|
||||
private String issuingAuthority;
|
||||
|
||||
/**
|
||||
* 有效期开始日期，格式：yyyy-MM-dd
|
||||
*/
|
||||
private String validFrom;
|
||||
|
||||
/**
|
||||
* 有效期结束日期，格式：yyyy-MM-dd 或 "长期"
|
||||
*/
|
||||
private String validTo;
|
||||
}
|
||||
|
||||
@@ -0,0 +1,41 @@
|
||||
package cn.smartjavaai.ocr.entity;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
/**
|
||||
* 身份证正面信息
|
||||
*/
|
||||
@Data
|
||||
public class IdCardFrontInfo {
|
||||
|
||||
/**
|
||||
* 姓名
|
||||
*/
|
||||
private String name;
|
||||
|
||||
/**
|
||||
* 性别（男/女）
|
||||
*/
|
||||
private String gender;
|
||||
|
||||
/**
|
||||
* 民族（如：汉、满、回等，不带“族”字）
|
||||
*/
|
||||
private String ethnicity;
|
||||
|
||||
/**
|
||||
* 公民身份号码
|
||||
*/
|
||||
private String idNumber;
|
||||
|
||||
/**
|
||||
* 出生日期，格式：yyyy-MM-dd
|
||||
*/
|
||||
private String birthday;
|
||||
|
||||
/**
|
||||
* 住址
|
||||
*/
|
||||
private String address;
|
||||
}
|
||||
|
||||
@@ -1,12 +1,26 @@
|
||||
package cn.smartjavaai.ocr.entity;
|
||||
|
||||
import lombok.Data;
|
||||
|
||||
/**
|
||||
* 身份证信息
|
||||
* 身份证信息（聚合正反面）
|
||||
*
|
||||
* 当前主要使用正面信息，预留反面字段，方便后续扩展。
|
||||
*
|
||||
* @author dwj
|
||||
* @date 2025/5/22
|
||||
*/
|
||||
@Data
|
||||
public class IdCardInfo {
|
||||
|
||||
/**
|
||||
* 身份证正面信息
|
||||
*/
|
||||
private IdCardFrontInfo front;
|
||||
|
||||
|
||||
/**
|
||||
* 身份证反面信息
|
||||
*/
|
||||
private IdCardBackInfo back;
|
||||
}
|
||||
|
||||
|
||||
@@ -0,0 +1,203 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import ai.djl.util.JsonUtils;
|
||||
import cn.smartjavaai.common.utils.ImageUtils;
|
||||
import cn.smartjavaai.ocr.config.OcrRecOptions;
|
||||
import cn.smartjavaai.ocr.entity.IdCardBackInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardFrontInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrItem;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
import cn.smartjavaai.ocr.model.common.direction.OcrDirectionModel;
|
||||
import cn.smartjavaai.ocr.model.common.recognize.OcrCommonRecModel;
|
||||
import cn.smartjavaai.ocr.utils.OcrUtils;
|
||||
import lombok.Data;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
import lombok.experimental.Accessors;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
/**
|
||||
* 身份证 OCR 识别服务实现
|
||||
*
|
||||
* - 支持从 Image 开始：内部完成预处理（方向矫正 OcrDirectionModel）+ OCR + 解析
|
||||
* - 支持从 OcrInfo 开始：只做结构化解析
|
||||
*
|
||||
* 通过 Lombok @Accessors(chain = true) 支持链式设置依赖。
|
||||
*/
|
||||
@Slf4j
|
||||
@Data
|
||||
@Accessors(chain = true)
|
||||
public class DefaultIdCardRecognizer implements IdCardRecognizer {
|
||||
|
||||
private static final int DEBUG_DRAW_FONT_SIZE = 20;
|
||||
|
||||
/**
|
||||
* 文本识别模型（可选；仅当从 Image 开始时需要）
|
||||
*/
|
||||
private OcrCommonRecModel recModel;
|
||||
|
||||
/**
|
||||
* 识别选项（如是否方向矫正、是否按行返回等）
|
||||
*/
|
||||
private OcrRecOptions recOptions;
|
||||
|
||||
/**
|
||||
* 方向检测模型（可选；预处理时用于 0/90/180/270° 整图方向矫正）。
|
||||
* 未设置则跳过方向矫正。
|
||||
*/
|
||||
private OcrDirectionModel directionModel;
|
||||
|
||||
/**
|
||||
* 图像预处理器（方向矫正，可选，默认按需 lazy-init）
|
||||
*/
|
||||
private IdCardPreprocessor preprocessor;
|
||||
|
||||
/**
|
||||
* 正面解析器（默认按需 lazy-init 为 IdCardFrontParser）
|
||||
*/
|
||||
private IdCardParser<IdCardFrontInfo> frontParser;
|
||||
|
||||
/**
|
||||
* 反面解析器（默认按需 lazy-init 为 IdCardBackParser）
|
||||
*/
|
||||
private IdCardParser<IdCardBackInfo> backParser;
|
||||
|
||||
/**
|
||||
* 校验器（可选，默认按需 lazy-init 为 IdCardValidator）
|
||||
*/
|
||||
private IdCardValidator validator;
|
||||
|
||||
/**
|
||||
* 是否进行图像预处理（方向矫正）。默认 true。
|
||||
* 若图片已校正，可设为 false 以跳过预处理、节省时间。
|
||||
*/
|
||||
private boolean enablePreprocess = true;
|
||||
|
||||
/**
|
||||
* 预处理调试监听器（可选）。
|
||||
* 若设置，则在方向矫正阶段结束后回调当前图像，
|
||||
* 方便最终用户在需要时保存中间结果进行排查和可视化调试。
|
||||
* 默认 null，不影响正常业务逻辑。
|
||||
*/
|
||||
private IdCardPreprocessListener preprocessListener;
|
||||
|
||||
@Override
|
||||
public IdCardFrontInfo recognizeFront(Image image) {
|
||||
if (recModel == null) {
|
||||
throw new IllegalStateException("recModel 未配置，无法从 Image 执行身份证识别，请先设置 recModel 或改用 recognizeFront(OcrInfo)。");
|
||||
}
|
||||
long totalStart = System.nanoTime();
|
||||
Image toRecognize = image;
|
||||
IdCardPreprocessResult preprocessResult = null;
|
||||
if (enablePreprocess) {
|
||||
long preprocessStart = System.nanoTime();
|
||||
IdCardPreprocessor pp = (preprocessor != null) ? preprocessor : (preprocessor = new IdCardPreprocessor());
|
||||
preprocessResult = pp.preprocess(image, recModel.getTextDetModel(), directionModel, preprocessListener);
|
||||
toRecognize = preprocessResult.getProcessedImage();
|
||||
log.debug("身份证正面预处理耗时={}ms", elapsedMillis(preprocessStart));
|
||||
}
|
||||
long recognizeStart = System.nanoTime();
|
||||
OcrInfo ocrInfo = preprocessResult != null
|
||||
&& !preprocessResult.isRotated()
|
||||
&& preprocessResult.getReusableBoxes() != null
|
||||
&& !preprocessResult.getReusableBoxes().isEmpty()
|
||||
? recModel.recognize(toRecognize, preprocessResult.getReusableBoxes(), recOptions)
|
||||
: recModel.recognize(toRecognize, recOptions);
|
||||
log.debug("身份证正面OCR模型调用耗时={}ms", elapsedMillis(recognizeStart));
|
||||
notifyAfterRecognize(toRecognize, ocrInfo);
|
||||
IdCardFrontInfo info = recognizeFront(ocrInfo);
|
||||
log.debug("身份证正面总耗时={}ms", elapsedMillis(totalStart));
|
||||
return info;
|
||||
}
|
||||
|
||||
@Override
|
||||
public IdCardBackInfo recognizeBack(Image image) {
|
||||
if (recModel == null) {
|
||||
throw new IllegalStateException("recModel 未配置，无法从 Image 执行身份证识别，请先设置 recModel 或改用 recognizeBack(OcrInfo)。");
|
||||
}
|
||||
long totalStart = System.nanoTime();
|
||||
Image toRecognize = image;
|
||||
IdCardPreprocessResult preprocessResult = null;
|
||||
if (enablePreprocess) {
|
||||
long preprocessStart = System.nanoTime();
|
||||
IdCardPreprocessor pp = (preprocessor != null) ? preprocessor : (preprocessor = new IdCardPreprocessor());
|
||||
preprocessResult = pp.preprocess(image, recModel.getTextDetModel(), directionModel, preprocessListener);
|
||||
toRecognize = preprocessResult.getProcessedImage();
|
||||
log.debug("身份证反面预处理耗时={}ms", elapsedMillis(preprocessStart));
|
||||
}
|
||||
long recognizeStart = System.nanoTime();
|
||||
OcrInfo ocrInfo = preprocessResult != null
|
||||
&& !preprocessResult.isRotated()
|
||||
&& preprocessResult.getReusableBoxes() != null
|
||||
&& !preprocessResult.getReusableBoxes().isEmpty()
|
||||
? recModel.recognize(toRecognize, preprocessResult.getReusableBoxes(), recOptions)
|
||||
: recModel.recognize(toRecognize, recOptions);
|
||||
log.debug("身份证反面OCR模型调用耗时={}ms", elapsedMillis(recognizeStart));
|
||||
notifyAfterRecognize(toRecognize, ocrInfo);
|
||||
IdCardBackInfo info = recognizeBack(ocrInfo);
|
||||
log.debug("身份证反面总耗时={}ms", elapsedMillis(totalStart));
|
||||
return info;
|
||||
}
|
||||
|
||||
private void notifyAfterRecognize(Image toRecognize, OcrInfo ocrInfo) {
|
||||
if (preprocessListener == null || toRecognize == null || ocrInfo == null) {
|
||||
return;
|
||||
}
|
||||
try {
|
||||
//打印
|
||||
log.debug("身份证 OCR 调试结果：{}", JsonUtils.toJson(ocrInfo));
|
||||
Image drawImage = ImageUtils.copy(toRecognize);
|
||||
List<OcrItem> ocrItems = ocrInfo.getOcrItemList();
|
||||
if ((ocrItems == null || ocrItems.isEmpty()) && ocrInfo.getLineList() != null) {
|
||||
ocrItems = ocrInfo.flattenLines();
|
||||
}
|
||||
if (ocrItems != null && !ocrItems.isEmpty()) {
|
||||
OcrUtils.drawOcrResult(drawImage, ocrItems, DEBUG_DRAW_FONT_SIZE);
|
||||
}
|
||||
preprocessListener.onAfterRecognize(drawImage, ocrInfo);
|
||||
} catch (Exception e) {
|
||||
log.warn("身份证 OCR 调试结果绘制失败，跳过 onAfterRecognize 回调。", e);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@Override
|
||||
public IdCardFrontInfo recognizeFront(OcrInfo ocrInfo) {
|
||||
long start = System.nanoTime();
|
||||
IdCardParser<IdCardFrontInfo> parser =
|
||||
(frontParser != null) ? frontParser : (frontParser = new IdCardFrontParser());
|
||||
IdCardFrontInfo info = parser.parse(ocrInfo);
|
||||
IdCardValidator v = (validator != null) ? validator : (validator = new IdCardValidator());
|
||||
v.validateFront(info);
|
||||
return info;
|
||||
}
|
||||
|
||||
@Override
|
||||
public IdCardBackInfo recognizeBack(OcrInfo ocrInfo) {
|
||||
IdCardParser<IdCardBackInfo> parser =
|
||||
(backParser != null) ? backParser : (backParser = new IdCardBackParser());
|
||||
IdCardBackInfo info = parser.parse(ocrInfo);
|
||||
IdCardValidator v = (validator != null) ? validator : (validator = new IdCardValidator());
|
||||
v.validateBack(info);
|
||||
return info;
|
||||
}
|
||||
|
||||
@Override
|
||||
public IdCardInfo recognizeBoth(Image frontImage, Image backImage) {
|
||||
long totalStart = System.nanoTime();
|
||||
IdCardInfo info = new IdCardInfo();
|
||||
info.setFront(recognizeFront(frontImage));
|
||||
info.setBack(recognizeBack(backImage));
|
||||
IdCardValidator v = (validator != null) ? validator : (validator = new IdCardValidator());
|
||||
v.validate(info);
|
||||
log.debug("身份证正反面识别总耗时={}ms", elapsedMillis(totalStart));
|
||||
return info;
|
||||
}
|
||||
|
||||
private long elapsedMillis(long startNanos) {
|
||||
return (System.nanoTime() - startNanos) / 1_000_000;
|
||||
}
|
||||
|
||||
}
|
||||
@@ -0,0 +1,335 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import cn.smartjavaai.ocr.entity.IdCardBackInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrItem;
|
||||
import cn.smartjavaai.ocr.utils.BoxUtils;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.IdentityHashMap;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Objects;
|
||||
import java.util.regex.Matcher;
|
||||
import java.util.regex.Pattern;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
/**
|
||||
* 身份证反面解析逻辑。
|
||||
*
|
||||
* 解析：
|
||||
* - 签发机关
|
||||
* - 有效期限（起止日期 / 长期）
|
||||
*/
|
||||
public class IdCardBackParser implements IdCardParser<IdCardBackInfo> {
|
||||
|
||||
private static final Pattern VALID_PERIOD_PATTERN = Pattern.compile(
|
||||
"((?:19|20)\\d{2})[./-]?((?:1[0-2])|(?:0[1-9]))[./-]?((?:3[01])|(?:[12]\\d)|(?:0[1-9]))" +
|
||||
"\\s*[-一至到~]\\s*" +
|
||||
"(((?:19|20)\\d{2})[./-]?((?:1[0-2])|(?:0[1-9]))[./-]?((?:3[01])|(?:[12]\\d)|(?:0[1-9]))|长期)"
|
||||
);
|
||||
|
||||
@Override
|
||||
public IdCardBackInfo parse(OcrInfo ocrInfo) {
|
||||
IdCardBackInfo info = new IdCardBackInfo();
|
||||
if (ocrInfo == null || ocrInfo.getLineList() == null || ocrInfo.getLineList().isEmpty()) {
|
||||
return info;
|
||||
}
|
||||
|
||||
List<OcrItem> allItems = ocrInfo.getLineList().stream()
|
||||
.filter(Objects::nonNull)
|
||||
.flatMap(List::stream)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
allItems = filterSmallBoxes(allItems);
|
||||
|
||||
if (allItems.isEmpty()) {
|
||||
return info;
|
||||
}
|
||||
Map<OcrBox, OcrItem> itemByBox = buildBoxItemMap(allItems);
|
||||
|
||||
OcrItem authorityLabel = findLabel(allItems, Arrays.asList("签发机关", "签发機关", "签发机関", "签发"));
|
||||
if (authorityLabel != null) {
|
||||
String authority = findIssuingAuthority(authorityLabel, allItems, itemByBox);
|
||||
info.setIssuingAuthority(authority);
|
||||
}
|
||||
|
||||
OcrItem validLabel = findLabel(allItems, Arrays.asList("有效期限", "有效期", "有效期限:", "有效期限："));
|
||||
if (validLabel != null) {
|
||||
ValidPeriod validPeriod = findValidPeriod(validLabel, allItems, itemByBox);
|
||||
if (validPeriod != null) {
|
||||
info.setValidFrom(validPeriod.validFrom);
|
||||
info.setValidTo(validPeriod.validTo);
|
||||
}
|
||||
}
|
||||
|
||||
if (info.getIssuingAuthority() == null || info.getIssuingAuthority().isEmpty()) {
|
||||
info.setIssuingAuthority(findIssuingAuthorityByGlobalFallback(allItems));
|
||||
}
|
||||
if (info.getValidFrom() == null || info.getValidTo() == null) {
|
||||
ValidPeriod fallback = findValidPeriodByGlobalFallback(allItems);
|
||||
if (fallback != null) {
|
||||
if (info.getValidFrom() == null) {
|
||||
info.setValidFrom(fallback.validFrom);
|
||||
}
|
||||
if (info.getValidTo() == null) {
|
||||
info.setValidTo(fallback.validTo);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return info;
|
||||
}
|
||||
|
||||
private String findIssuingAuthority(OcrItem authorityLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
String selfAuthority = extractIssuingAuthority(authorityLabel.getText());
|
||||
if (selfAuthority != null) {
|
||||
return selfAuthority;
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox anchorBox = authorityLabel.getOcrBox();
|
||||
String authority = findAuthorityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.RIGHT, 3);
|
||||
if (authority != null) {
|
||||
return authority;
|
||||
}
|
||||
authority = findAuthorityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.DOWN, 3);
|
||||
if (authority != null) {
|
||||
return authority;
|
||||
}
|
||||
authority = findAuthorityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.UP, 2);
|
||||
if (authority != null) {
|
||||
return authority;
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private String findAuthorityByDirection(OcrBox anchorBox,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
BoxUtils.Direction direction,
|
||||
int limit) {
|
||||
List<OcrBox> neighbors = BoxUtils.findNearestBoxes(anchorBox, boxList, direction, limit);
|
||||
if (neighbors == null || neighbors.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
List<OcrItem> neighborItems = neighbors.stream()
|
||||
.map(itemByBox::get)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
for (OcrItem item : neighborItems) {
|
||||
String authority = extractIssuingAuthority(item.getText());
|
||||
if (authority != null) {
|
||||
return authority;
|
||||
}
|
||||
}
|
||||
|
||||
String merged = neighborItems.stream()
|
||||
.map(OcrItem::getText)
|
||||
.filter(Objects::nonNull)
|
||||
.map(this::normalizeText)
|
||||
.collect(Collectors.joining());
|
||||
return extractIssuingAuthority(merged);
|
||||
}
|
||||
|
||||
private String extractIssuingAuthority(String text) {
|
||||
if (text == null || text.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
String clean = normalizeText(text)
|
||||
.replace("签发机关", "")
|
||||
.replace("签发機关", "")
|
||||
.replace("签发机関", "")
|
||||
.replace("签发", "");
|
||||
if (clean.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
if (clean.contains("有效期限") || clean.contains("有效期")) {
|
||||
return null;
|
||||
}
|
||||
return clean;
|
||||
}
|
||||
|
||||
private ValidPeriod findValidPeriod(OcrItem validLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
ValidPeriod selfPeriod = extractValidPeriod(validLabel.getText());
|
||||
if (selfPeriod != null) {
|
||||
return selfPeriod;
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox anchorBox = validLabel.getOcrBox();
|
||||
ValidPeriod period = findValidPeriodByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.RIGHT, 4);
|
||||
if (period != null) {
|
||||
return period;
|
||||
}
|
||||
period = findValidPeriodByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.DOWN, 4);
|
||||
if (period != null) {
|
||||
return period;
|
||||
}
|
||||
period = findValidPeriodByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.UP, 2);
|
||||
if (period != null) {
|
||||
return period;
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private ValidPeriod findValidPeriodByDirection(OcrBox anchorBox,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
BoxUtils.Direction direction,
|
||||
int limit) {
|
||||
List<OcrBox> neighbors = BoxUtils.findNearestBoxes(anchorBox, boxList, direction, limit);
|
||||
if (neighbors == null || neighbors.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
List<OcrItem> neighborItems = neighbors.stream()
|
||||
.map(itemByBox::get)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
for (OcrItem item : neighborItems) {
|
||||
ValidPeriod period = extractValidPeriod(item.getText());
|
||||
if (period != null) {
|
||||
return period;
|
||||
}
|
||||
}
|
||||
|
||||
String merged = neighborItems.stream()
|
||||
.map(OcrItem::getText)
|
||||
.filter(Objects::nonNull)
|
||||
.map(this::normalizeText)
|
||||
.collect(Collectors.joining());
|
||||
return extractValidPeriod(merged);
|
||||
}
|
||||
|
||||
private ValidPeriod extractValidPeriod(String text) {
|
||||
if (text == null || text.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
String clean = normalizeText(text)
|
||||
.replace("有效期限", "")
|
||||
.replace("有效期", "");
|
||||
|
||||
Matcher matcher = VALID_PERIOD_PATTERN.matcher(clean);
|
||||
if (!matcher.find()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
String from = normalizeDate(matcher.group(1), matcher.group(2), matcher.group(3));
|
||||
if (from == null) {
|
||||
return null;
|
||||
}
|
||||
|
||||
String toRaw = matcher.group(4);
|
||||
String to = "长期".equals(toRaw)
|
||||
? "长期"
|
||||
: normalizeDate(matcher.group(5), matcher.group(6), matcher.group(7));
|
||||
if (to == null) {
|
||||
return null;
|
||||
}
|
||||
|
||||
return new ValidPeriod(from, to);
|
||||
}
|
||||
|
||||
private String normalizeDate(String year, String month, String day) {
|
||||
if (year == null || month == null || day == null) {
|
||||
return null;
|
||||
}
|
||||
return year + "-" + month + "-" + day;
|
||||
}
|
||||
|
||||
private String findIssuingAuthorityByGlobalFallback(List<OcrItem> allItems) {
|
||||
for (OcrItem item : allItems) {
|
||||
String authority = extractIssuingAuthority(item.getText());
|
||||
if (authority != null && (authority.contains("公安局") || authority.contains("分局") || authority.contains("机关"))) {
|
||||
return authority;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private ValidPeriod findValidPeriodByGlobalFallback(List<OcrItem> allItems) {
|
||||
List<String> texts = new ArrayList<>();
|
||||
for (OcrItem item : allItems) {
|
||||
if (item != null && item.getText() != null) {
|
||||
ValidPeriod period = extractValidPeriod(item.getText());
|
||||
if (period != null) {
|
||||
return period;
|
||||
}
|
||||
texts.add(normalizeText(item.getText()));
|
||||
}
|
||||
}
|
||||
return extractValidPeriod(String.join("", texts));
|
||||
}
|
||||
|
||||
private OcrItem findLabel(List<OcrItem> allItems, List<String> keywords) {
|
||||
for (OcrItem item : allItems) {
|
||||
String text = item.getText();
|
||||
if (text == null) {
|
||||
continue;
|
||||
}
|
||||
String clean = normalizeText(text);
|
||||
for (String keyword : keywords) {
|
||||
if (clean.contains(keyword)) {
|
||||
return item;
|
||||
}
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private String normalizeText(String text) {
|
||||
if (text == null) {
|
||||
return "";
|
||||
}
|
||||
return text.replace(" ", "")
|
||||
.replace("一", "-")
|
||||
.replace("到", "-")
|
||||
.replace("至", "-")
|
||||
.replace("~", "-")
|
||||
.replace("—", "-")
|
||||
.replace("－", "-");
|
||||
}
|
||||
|
||||
private List<OcrItem> filterSmallBoxes(List<OcrItem> items) {
|
||||
List<OcrItem> result = IdCardOcrUtils.filterSmallBoxes(items, 0.15);
|
||||
return result;
|
||||
}
|
||||
|
||||
private Map<OcrBox, OcrItem> buildBoxItemMap(List<OcrItem> items) {
|
||||
Map<OcrBox, OcrItem> itemByBox = new IdentityHashMap<>();
|
||||
if (items == null) {
|
||||
return itemByBox;
|
||||
}
|
||||
for (OcrItem item : items) {
|
||||
if (item != null && item.getOcrBox() != null) {
|
||||
itemByBox.put(item.getOcrBox(), item);
|
||||
}
|
||||
}
|
||||
return itemByBox;
|
||||
}
|
||||
|
||||
private static class ValidPeriod {
|
||||
private final String validFrom;
|
||||
private final String validTo;
|
||||
|
||||
private ValidPeriod(String validFrom, String validTo) {
|
||||
this.validFrom = validFrom;
|
||||
this.validTo = validTo;
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,930 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import cn.smartjavaai.common.entity.Point;
|
||||
import cn.smartjavaai.ocr.entity.IdCardFrontInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrItem;
|
||||
import cn.smartjavaai.ocr.utils.BoxUtils;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
import java.time.DateTimeException;
|
||||
import java.time.LocalDate;
|
||||
import java.util.*;
|
||||
import java.util.function.Function;
|
||||
import java.util.regex.Matcher;
|
||||
import java.util.regex.Pattern;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
/**
|
||||
* 身份证正面解析逻辑。
|
||||
*
|
||||
* 从 OCR 结果中解析：
|
||||
* - 姓名（兼容“姓名+值在同一检测框”与“键值分框”的情况）
|
||||
* - 性别
|
||||
* - 民族
|
||||
* - 出生日期
|
||||
* - 公民身份号码
|
||||
* - 住址
|
||||
*/
|
||||
@Slf4j
|
||||
public class IdCardFrontParser implements IdCardParser<IdCardFrontInfo> {
|
||||
|
||||
private static final Pattern BIRTHDAY_WITH_SEPARATORS = Pattern.compile(
|
||||
"((?:19|20)\\d{2})[年/._-]?((?:1[0-2])|(?:0?[1-9]))[月/._-]?((?:3[01])|(?:[12]\\d)|(?:0?[1-9]))日?"
|
||||
);
|
||||
|
||||
private static final Pattern BIRTHDAY_COMPACT = Pattern.compile(
|
||||
"((?:19|20)\\d{2})((?:1[0-2])|(?:0[1-9]))((?:3[01])|(?:[12]\\d)|(?:0[1-9]))"
|
||||
);
|
||||
|
||||
@Override
|
||||
public IdCardFrontInfo parse(OcrInfo ocrInfo) {
|
||||
IdCardFrontInfo info = new IdCardFrontInfo();
|
||||
if (ocrInfo == null || ocrInfo.getLineList() == null || ocrInfo.getLineList().isEmpty()) {
|
||||
return info;
|
||||
}
|
||||
|
||||
// 扁平化所有 item
|
||||
List<OcrItem> allItems = ocrInfo.getLineList().stream()
|
||||
.filter(Objects::nonNull)
|
||||
.flatMap(List::stream)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
// 过滤明显过小的检测框（噪声）
|
||||
allItems = filterSmallBoxes(allItems);
|
||||
|
||||
// 可用的检测框列表：解析出字段后逐步移除，减少后续遍历量
|
||||
List<OcrItem> remainingItems = new ArrayList<>(allItems);
|
||||
Map<OcrBox, OcrItem> itemByBox = buildBoxItemMap(remainingItems);
|
||||
|
||||
// 1. 姓名（兼容“标签+值同框”以及右侧单独值框）
|
||||
OcrItem nameLabel = findLabel(remainingItems, Arrays.asList("姓名", "姓名:", "姓名："));
|
||||
if (nameLabel != null) {
|
||||
StringBuilder nameSb = new StringBuilder();
|
||||
if (nameLabel.getText() != null) {
|
||||
nameSb.append(nameLabel.getText());
|
||||
}
|
||||
// 右侧最近一个框（可能是姓名值）
|
||||
OcrItem rightItem = findNearestRightItem(nameLabel, remainingItems, itemByBox);
|
||||
if (rightItem != null && rightItem.getText() != null) {
|
||||
nameSb.append(rightItem.getText());
|
||||
}
|
||||
|
||||
// 过滤：只保留中文字符
|
||||
String nameText = nameSb.toString()
|
||||
.replace(" ", "")
|
||||
.replaceAll("[^\\u4e00-\\u9fa5]", "");
|
||||
|
||||
// 提取"姓名"后面的内容作为名字
|
||||
String name = extractNameAfterLabel(nameText);
|
||||
if (name != null && !name.isEmpty()) {
|
||||
info.setName(name);
|
||||
remainingItems.remove(nameLabel);
|
||||
if (rightItem != null) {
|
||||
remainingItems.remove(rightItem);
|
||||
}
|
||||
itemByBox = buildBoxItemMap(remainingItems);
|
||||
}
|
||||
}
|
||||
|
||||
// 2. 性别（优先右 → 上 → 下 → 全局）
|
||||
OcrItem genderLabel = findLabel(remainingItems, Arrays.asList("性别", "性别:", "性别："));
|
||||
if (genderLabel != null) {
|
||||
String gender = findGenderAroundLabel(genderLabel, remainingItems, itemByBox);
|
||||
if (gender != null) {
|
||||
info.setGender(gender);
|
||||
final String g = gender;
|
||||
// 移除性别相关检测框（若同时包含民族信息则保留用于民族解析）
|
||||
remainingItems.removeIf(it -> {
|
||||
String t = it.getText();
|
||||
if (t == null) {
|
||||
return false;
|
||||
}
|
||||
String clean = t.replace(" ", "");
|
||||
if (clean.contains("性别") && !clean.contains("民族")) {
|
||||
return true;
|
||||
}
|
||||
String cg = cleanGender(t);
|
||||
return g.equals(cg) && !clean.contains("民族");
|
||||
});
|
||||
itemByBox = buildBoxItemMap(remainingItems);
|
||||
}
|
||||
}
|
||||
|
||||
// 3. 民族（策略同性别）
|
||||
OcrItem ethnicLabel = findLabel(remainingItems, Arrays.asList("民族", "民族:", "民族："));
|
||||
if (ethnicLabel != null) {
|
||||
String ethnicity = findEthnicityAroundLabel(ethnicLabel, remainingItems, itemByBox);
|
||||
if (ethnicity != null) {
|
||||
info.setEthnicity(ethnicity);
|
||||
final String eth = ethnicity;
|
||||
remainingItems.removeIf(it -> {
|
||||
String t = it.getText();
|
||||
if (t == null) {
|
||||
return false;
|
||||
}
|
||||
String clean = t.replace(" ", "");
|
||||
if (clean.contains("民族")) {
|
||||
return true;
|
||||
}
|
||||
String e = extractEthnicity(t);
|
||||
return eth.equals(e);
|
||||
});
|
||||
itemByBox = buildBoxItemMap(remainingItems);
|
||||
}
|
||||
}
|
||||
|
||||
// 4. 公民身份号码
|
||||
OcrItem idSourceItem = null;
|
||||
OcrItem idRightItem = null;
|
||||
for (OcrItem item : remainingItems) {
|
||||
String text = normalizeText(item.getText());
|
||||
if (text.contains("公民身份号码") || text.contains("公民身份號碼") || text.toUpperCase().contains("ID")) {
|
||||
String idInBox = extractIdNumber(text);
|
||||
if (idInBox == null) {
|
||||
OcrItem idRight = findNearestRightItem(item, remainingItems, itemByBox);
|
||||
if (idRight != null) {
|
||||
idInBox = extractIdNumber(normalizeText(idRight.getText()));
|
||||
idRightItem = idRight;
|
||||
}
|
||||
}
|
||||
if (idInBox != null) {
|
||||
info.setIdNumber(idInBox);
|
||||
idSourceItem = item;
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if (idSourceItem != null) {
|
||||
remainingItems.remove(idSourceItem);
|
||||
}
|
||||
if (idRightItem != null) {
|
||||
remainingItems.remove(idRightItem);
|
||||
}
|
||||
if (idSourceItem != null || idRightItem != null) {
|
||||
itemByBox = buildBoxItemMap(remainingItems);
|
||||
}
|
||||
|
||||
// 5. 出生日期：考虑多框拆分 & 与标签同框等多种情况
|
||||
OcrItem birthLabel = findLabel(remainingItems, Arrays.asList("出生", "出生日期", "出生:", "出生："));
|
||||
if (birthLabel != null) {
|
||||
String birth = findBirthdayAroundLabel(birthLabel, remainingItems, itemByBox);
|
||||
if (birth != null) {
|
||||
info.setBirthday(birth);
|
||||
final String bFinal = birth;
|
||||
remainingItems.removeIf(it -> {
|
||||
String t = it.getText();
|
||||
if (t == null) {
|
||||
return false;
|
||||
}
|
||||
String clean = t.replace(" ", "");
|
||||
if (clean.contains("出生")) {
|
||||
return true;
|
||||
}
|
||||
String ex = extractBirthday(t);
|
||||
return bFinal.equals(ex);
|
||||
});
|
||||
itemByBox = buildBoxItemMap(remainingItems);
|
||||
}
|
||||
}
|
||||
|
||||
// 如果生日没识别到，则尝试从身份证号码中推断
|
||||
if (info.getBirthday() == null && info.getIdNumber() != null && info.getIdNumber().length() >= 14) {
|
||||
String id = info.getIdNumber();
|
||||
String yyyyMMdd = id.substring(6, 14);
|
||||
info.setBirthday(formatBirthdayFromId(yyyyMMdd));
|
||||
}
|
||||
|
||||
// 6. 住址（从“住址”右侧开始，向下若干行拼接）
|
||||
OcrItem addressLabel = findAddressLabel(remainingItems);
|
||||
if (addressLabel != null) {
|
||||
String addr = collectAddress(addressLabel, remainingItems, itemByBox);
|
||||
info.setAddress(addr);
|
||||
}
|
||||
|
||||
return info;
|
||||
}
|
||||
|
||||
// ----------------- 文本与 Box 相关的工具方法 -----------------
|
||||
|
||||
private String normalizeText(String text) {
|
||||
if (text == null) {
|
||||
return "";
|
||||
}
|
||||
return text.replace(" ", "")
|
||||
.replaceAll("[^0-9Xx\\u4e00-\\u9fa5]", "");
|
||||
}
|
||||
|
||||
/**
|
||||
* 从包含"姓名"的文本中提取姓名值（提取"姓名"后面的所有中文字符）
|
||||
*/
|
||||
private String extractNameAfterLabel(String text) {
|
||||
if (text == null || text.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
int nameIndex = text.indexOf("姓名");
|
||||
if (nameIndex >= 0) {
|
||||
String name = text.substring(nameIndex + 2);
|
||||
return name.isEmpty() ? null : name;
|
||||
}
|
||||
int surnameIndex = text.indexOf("姓");
|
||||
if (surnameIndex >= 0 && surnameIndex < text.length() - 1) {
|
||||
String name = text.substring(surnameIndex + 1);
|
||||
return name.isEmpty() ? null : name;
|
||||
}
|
||||
int givenNameIndex = text.indexOf("名");
|
||||
if (givenNameIndex >= 0 && givenNameIndex < text.length() - 1) {
|
||||
String name = text.substring(givenNameIndex + 1);
|
||||
return name.isEmpty() ? null : name;
|
||||
}
|
||||
// 如果都没找到，返回原文本（可能已经是纯姓名）
|
||||
return text;
|
||||
}
|
||||
|
||||
private String cleanGender(String text) {
|
||||
if (text == null) {
|
||||
return null;
|
||||
}
|
||||
String clean = text.replace(" ", "");
|
||||
if (clean.contains("男")) {
|
||||
return "男";
|
||||
}
|
||||
if (clean.contains("女")) {
|
||||
return "女";
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 按“右 → 上 → 下 → 全局”顺序查找性别信息
|
||||
*/
|
||||
private String findGenderAroundLabel(OcrItem genderLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
if (genderLabel == null || genderLabel.getOcrBox() == null || allItems == null || allItems.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
// 情况一：标签所在框本身包含“性别+男/女”
|
||||
String selfGender = cleanGender(genderLabel.getText());
|
||||
if (selfGender != null) {
|
||||
return selfGender;
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox anchorBox = genderLabel.getOcrBox();
|
||||
|
||||
String gender = findGenderByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.RIGHT, 3);
|
||||
if (gender != null) {
|
||||
return gender;
|
||||
}
|
||||
gender = findGenderByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.UP, 3);
|
||||
if (gender != null) {
|
||||
return gender;
|
||||
}
|
||||
gender = findGenderByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.DOWN, 3);
|
||||
if (gender != null) {
|
||||
return gender;
|
||||
}
|
||||
|
||||
return findNearbyFallback(anchorBox, allItems, this::cleanGender);
|
||||
}
|
||||
|
||||
/**
|
||||
* 按指定方向在若干个最近框中查找性别
|
||||
*/
|
||||
private String findGenderByDirection(OcrBox anchorBox,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
BoxUtils.Direction direction,
|
||||
int limit) {
|
||||
List<OcrBox> neighbors = BoxUtils.findNearestBoxes(anchorBox, boxList, direction, limit);
|
||||
if (neighbors == null || neighbors.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
for (OcrBox neighbor : neighbors) {
|
||||
OcrItem item = itemByBox.get(neighbor);
|
||||
if (item == null) {
|
||||
continue;
|
||||
}
|
||||
String g = cleanGender(item.getText());
|
||||
if (g != null) {
|
||||
return g;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 按“右 → 上 → 下 → 全局”顺序查找民族信息
|
||||
*/
|
||||
private String findEthnicityAroundLabel(OcrItem ethnicLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
if (ethnicLabel == null || ethnicLabel.getOcrBox() == null || allItems == null || allItems.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
String selfEthnic = extractEthnicity(ethnicLabel.getText());
|
||||
if (selfEthnic != null) {
|
||||
return selfEthnic;
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox anchorBox = ethnicLabel.getOcrBox();
|
||||
|
||||
String ethnicity = findEthnicityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.RIGHT, 3);
|
||||
if (ethnicity != null) {
|
||||
return ethnicity;
|
||||
}
|
||||
ethnicity = findEthnicityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.UP, 3);
|
||||
if (ethnicity != null) {
|
||||
return ethnicity;
|
||||
}
|
||||
ethnicity = findEthnicityByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.DOWN, 3);
|
||||
if (ethnicity != null) {
|
||||
return ethnicity;
|
||||
}
|
||||
|
||||
return findNearbyFallback(anchorBox, allItems, this::extractEthnicity);
|
||||
}
|
||||
|
||||
/**
|
||||
* 按指定方向在若干个最近框中查找民族
|
||||
*/
|
||||
private String findEthnicityByDirection(OcrBox anchorBox,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
BoxUtils.Direction direction,
|
||||
int limit) {
|
||||
List<OcrBox> neighbors = BoxUtils.findNearestBoxes(anchorBox, boxList, direction, limit);
|
||||
if (neighbors == null || neighbors.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
for (OcrBox neighbor : neighbors) {
|
||||
OcrItem item = itemByBox.get(neighbor);
|
||||
if (item == null) {
|
||||
continue;
|
||||
}
|
||||
String e = extractEthnicity(item.getText());
|
||||
if (e != null) {
|
||||
return e;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 从文本中提取并纠正民族信息
|
||||
*/
|
||||
private String extractEthnicity(String ocrText) {
|
||||
if (ocrText == null || ocrText.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
String cleanText = ocrText.replaceAll("[^\\u4e00-\\u9fa5]", "");
|
||||
cleanText = cleanText.replaceAll("^民族", "");
|
||||
for (String ethnic : IdCardOcrUtils.ETHNIC_SET) {
|
||||
if (cleanText.contains(ethnic)) {
|
||||
return ethnic;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 兜底时仅在标签附近寻找合法候选，避免被远处噪声或少数民族文字误带偏。
|
||||
*/
|
||||
private String findNearbyFallback(OcrBox anchorBox,
|
||||
List<OcrItem> allItems,
|
||||
Function<String, String> extractor) {
|
||||
if (anchorBox == null || allItems == null || allItems.isEmpty() || extractor == null) {
|
||||
return null;
|
||||
}
|
||||
|
||||
double anchorWidth = boxWidth(anchorBox);
|
||||
double anchorHeight = boxHeight(anchorBox);
|
||||
Point anchorCenter = boxCenter(anchorBox);
|
||||
|
||||
OcrItem bestItem = null;
|
||||
double bestScore = Double.MAX_VALUE;
|
||||
|
||||
for (OcrItem item : allItems) {
|
||||
if (item == null || item.getOcrBox() == null) {
|
||||
continue;
|
||||
}
|
||||
String extracted = extractor.apply(item.getText());
|
||||
if (extracted == null) {
|
||||
continue;
|
||||
}
|
||||
|
||||
Point candidateCenter = boxCenter(item.getOcrBox());
|
||||
double dx = Math.abs(candidateCenter.getX() - anchorCenter.getX());
|
||||
double dy = Math.abs(candidateCenter.getY() - anchorCenter.getY());
|
||||
|
||||
// 性别/民族通常紧邻标签，限制在局部窗口内做兜底。
|
||||
if (dx > anchorWidth * 5.0 || dy > anchorHeight * 2.5) {
|
||||
continue;
|
||||
}
|
||||
|
||||
double score = dy * 2.0 + dx;
|
||||
if (score < bestScore) {
|
||||
bestScore = score;
|
||||
bestItem = item;
|
||||
}
|
||||
}
|
||||
|
||||
return bestItem == null ? null : extractor.apply(bestItem.getText());
|
||||
}
|
||||
|
||||
/**
|
||||
* 按“自身 → 右 → 下 → 上 → 全局”顺序查找出生日期
|
||||
*/
|
||||
|
||||
private String findBirthdayAroundLabel(OcrItem birthLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
if (birthLabel == null || birthLabel.getOcrBox() == null || allItems == null || allItems.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
String fromSelf = extractBirthday(birthLabel.getText());
|
||||
if (fromSelf != null) {
|
||||
return fromSelf;
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox anchorBox = birthLabel.getOcrBox();
|
||||
|
||||
String fromRight = findBirthdayByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.RIGHT, 4);
|
||||
if (fromRight != null) {
|
||||
return fromRight;
|
||||
}
|
||||
|
||||
String fromDown = findBirthdayByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.DOWN, 4);
|
||||
if (fromDown != null) {
|
||||
return fromDown;
|
||||
}
|
||||
|
||||
String fromUp = findBirthdayByDirection(anchorBox, itemByBox, boxList, BoxUtils.Direction.UP, 3);
|
||||
if (fromUp != null) {
|
||||
return fromUp;
|
||||
}
|
||||
|
||||
for (OcrItem item : allItems) {
|
||||
String b = extractBirthday(item.getText());
|
||||
if (b != null) {
|
||||
return b;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 按指定方向在若干最近框中尝试组合/单独解析出生日期
|
||||
*/
|
||||
private String findBirthdayByDirection(OcrBox anchorBox,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
BoxUtils.Direction direction,
|
||||
int limit) {
|
||||
List<OcrBox> neighbors = BoxUtils.findNearestBoxes(anchorBox, boxList, direction, limit);
|
||||
if (neighbors == null || neighbors.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
List<OcrItem> neighborItems = neighbors.stream()
|
||||
.map(itemByBox::get)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
if (neighborItems.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
|
||||
for (OcrItem item : neighborItems) {
|
||||
String b = extractBirthday(item.getText());
|
||||
if (b != null) {
|
||||
return b;
|
||||
}
|
||||
}
|
||||
|
||||
StringBuilder sb = new StringBuilder();
|
||||
for (OcrItem item : neighborItems) {
|
||||
if (item.getText() != null) {
|
||||
sb.append(item.getText().replace(" ", ""));
|
||||
}
|
||||
}
|
||||
String merged = sb.toString();
|
||||
if (!merged.isEmpty()) {
|
||||
String b = extractBirthday(merged);
|
||||
if (b != null) {
|
||||
return b;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 从文本中提取身份证号（优先 18 位二代证）
|
||||
*/
|
||||
private String extractIdNumber(String text) {
|
||||
if (text == null) {
|
||||
return null;
|
||||
}
|
||||
String candidate = text.toUpperCase();
|
||||
Matcher m = Pattern.compile("[0-9X]{15,18}").matcher(candidate);
|
||||
while (m.find()) {
|
||||
String id = m.group();
|
||||
if (id.length() == 18) {
|
||||
return id;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
/**
|
||||
* 从文本中提取生日（标准化为 yyyy-MM-dd）
|
||||
*/
|
||||
private String extractBirthday(String text) {
|
||||
if (text == null) {
|
||||
return null;
|
||||
}
|
||||
String clean = text.replace(" ", "");
|
||||
|
||||
Matcher m1 = BIRTHDAY_WITH_SEPARATORS.matcher(clean);
|
||||
if (m1.find()) {
|
||||
String birthday = normalizeBirthday(m1.group(1), m1.group(2), m1.group(3));
|
||||
if (birthday != null) {
|
||||
return birthday;
|
||||
}
|
||||
}
|
||||
|
||||
Matcher m2 = BIRTHDAY_COMPACT.matcher(clean);
|
||||
if (m2.find()) {
|
||||
String birthday = normalizeBirthday(m2.group(1), m2.group(2), m2.group(3));
|
||||
if (birthday != null) {
|
||||
return birthday;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private String normalizeBirthday(String raw) {
|
||||
String clean = raw.replace("年", "-")
|
||||
.replace("月", "-")
|
||||
.replace("日", "")
|
||||
.replace("/", "-")
|
||||
.replace(".", "-");
|
||||
String[] parts = clean.split("-");
|
||||
if (parts.length != 3) {
|
||||
return null;
|
||||
}
|
||||
String y = parts[0];
|
||||
String m = parts[1].length() == 1 ? "0" + parts[1] : parts[1];
|
||||
String d = parts[2].length() == 1 ? "0" + parts[2] : parts[2];
|
||||
return y + "-" + m + "-" + d;
|
||||
}
|
||||
|
||||
private String normalizeBirthday(String year, String month, String day) {
|
||||
if (year == null || month == null || day == null) {
|
||||
return null;
|
||||
}
|
||||
String normalizedMonth = month.length() == 1 ? "0" + month : month;
|
||||
String normalizedDay = day.length() == 1 ? "0" + day : day;
|
||||
try {
|
||||
LocalDate date = LocalDate.of(
|
||||
Integer.parseInt(year),
|
||||
Integer.parseInt(normalizedMonth),
|
||||
Integer.parseInt(normalizedDay)
|
||||
);
|
||||
return date.toString();
|
||||
} catch (DateTimeException | NumberFormatException e) {
|
||||
return null;
|
||||
}
|
||||
}
|
||||
|
||||
private String formatBirthdayFromId(String yyyymmdd) {
|
||||
if (yyyymmdd == null || yyyymmdd.length() != 8) {
|
||||
return null;
|
||||
}
|
||||
String y = yyyymmdd.substring(0, 4);
|
||||
String m = yyyymmdd.substring(4, 6);
|
||||
String d = yyyymmdd.substring(6, 8);
|
||||
return y + "-" + m + "-" + d;
|
||||
}
|
||||
|
||||
private Point boxCenter(OcrBox box) {
|
||||
float[] pts = box.toFloatArray();
|
||||
float cx = (pts[0] + pts[2] + pts[4] + pts[6]) / 4;
|
||||
float cy = (pts[1] + pts[3] + pts[5] + pts[7]) / 4;
|
||||
return new Point(cx, cy);
|
||||
}
|
||||
|
||||
private double boxWidth(OcrBox box) {
|
||||
float[] pts = box.toFloatArray();
|
||||
float minX = Math.min(Math.min(pts[0], pts[2]), Math.min(pts[4], pts[6]));
|
||||
float maxX = Math.max(Math.max(pts[0], pts[2]), Math.max(pts[4], pts[6]));
|
||||
return Math.max(1.0, maxX - minX);
|
||||
}
|
||||
|
||||
private double boxHeight(OcrBox box) {
|
||||
float[] pts = box.toFloatArray();
|
||||
float minY = Math.min(Math.min(pts[1], pts[3]), Math.min(pts[5], pts[7]));
|
||||
float maxY = Math.max(Math.max(pts[1], pts[3]), Math.max(pts[5], pts[7]));
|
||||
return Math.max(1.0, maxY - minY);
|
||||
}
|
||||
|
||||
/**
|
||||
* 过滤掉面积明显小于整体的检测框，粗略去噪
|
||||
*/
|
||||
private List<OcrItem> filterSmallBoxes(List<OcrItem> items) {
|
||||
List<OcrItem> result = IdCardOcrUtils.filterSmallBoxes(items, 0.15);
|
||||
log.debug("身份证正面解析：过滤小框完成，原始数量={}，过滤后数量={}", items == null ? 0 : items.size(), result == null ? 0 : result.size());
|
||||
return result;
|
||||
}
|
||||
|
||||
/**
|
||||
* 基于 BoxUtils，从锚点右侧找到最近的一个文本框
|
||||
*/
|
||||
private OcrItem findNearestRightItem(OcrItem anchor, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
if (anchor == null || anchor.getOcrBox() == null) {
|
||||
return null;
|
||||
}
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
OcrBox nearest = BoxUtils.findNearestBox(anchor.getOcrBox(), boxList, BoxUtils.Direction.RIGHT);
|
||||
if (nearest == null) {
|
||||
return null;
|
||||
}
|
||||
return itemByBox.get(nearest);
|
||||
}
|
||||
|
||||
/**
|
||||
* 查找包含标签关键字的 item
|
||||
*/
|
||||
private OcrItem findLabel(List<OcrItem> allItems, List<String> keywords) {
|
||||
for (OcrItem item : allItems) {
|
||||
String text = item.getText();
|
||||
if (text == null) {
|
||||
continue;
|
||||
}
|
||||
String clean = text.replace(" ", "");
|
||||
for (String kw : keywords) {
|
||||
if (clean.contains(kw)) {
|
||||
return item;
|
||||
}
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private OcrItem findAddressLabel(List<OcrItem> allItems) {
|
||||
OcrItem direct = findLabel(allItems, Arrays.asList("住址", "地址", "住址:", "住址："));
|
||||
if (direct != null) {
|
||||
return direct;
|
||||
}
|
||||
if (allItems == null) {
|
||||
return null;
|
||||
}
|
||||
for (OcrItem item : allItems) {
|
||||
String text = item.getText();
|
||||
if (text == null) {
|
||||
continue;
|
||||
}
|
||||
String clean = text.replace(" ", "");
|
||||
if (looksLikeAddressLabel(clean)) {
|
||||
return item;
|
||||
}
|
||||
}
|
||||
return null;
|
||||
}
|
||||
|
||||
private boolean looksLikeAddressLabel(String text) {
|
||||
if (text == null || text.length() < 2) {
|
||||
return false;
|
||||
}
|
||||
if (text.contains("址") && (text.startsWith("住") || text.startsWith("佳") || text.startsWith("往"))) {
|
||||
return true;
|
||||
}
|
||||
return false;
|
||||
}
|
||||
|
||||
/**
|
||||
* 拼接地址：
|
||||
* 1. 从“住址”右侧开始，整行向右拼接多个检测框
|
||||
* 2. 继续向下拼接若干行同一列附近的文本
|
||||
* 3. 若某一行疑似“公民身份号码”行，则终止
|
||||
*/
|
||||
private String collectAddress(OcrItem addressLabel, List<OcrItem> allItems, Map<OcrBox, OcrItem> itemByBox) {
|
||||
StringBuilder sb = new StringBuilder();
|
||||
if (addressLabel == null || addressLabel.getOcrBox() == null || allItems == null || allItems.isEmpty()) {
|
||||
return "";
|
||||
}
|
||||
|
||||
List<OcrBox> boxList = allItems.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.collect(Collectors.toList());
|
||||
|
||||
Set<OcrBox> usedBoxes = new HashSet<>();
|
||||
|
||||
String labelAddressText = extractAddressAfterLabel(addressLabel.getText());
|
||||
if (!labelAddressText.isEmpty()) {
|
||||
appendMergedText(sb, labelAddressText);
|
||||
usedBoxes.add(addressLabel.getOcrBox());
|
||||
}
|
||||
|
||||
OcrItem firstLineItem = findNearestRightItem(addressLabel, allItems, itemByBox);
|
||||
if (firstLineItem != null && firstLineItem.getOcrBox() != null) {
|
||||
String firstLineText = buildAddressLine(firstLineItem.getOcrBox(), itemByBox, boxList, usedBoxes);
|
||||
if (!firstLineText.isEmpty() && !isIdNumberLine(firstLineText)) {
|
||||
appendMergedText(sb, firstLineText);
|
||||
} else if (isIdNumberLine(firstLineText)) {
|
||||
return sb.toString();
|
||||
}
|
||||
}
|
||||
|
||||
OcrBox current = firstLineItem != null && firstLineItem.getOcrBox() != null
|
||||
? firstLineItem.getOcrBox()
|
||||
: addressLabel.getOcrBox();
|
||||
|
||||
for (int i = 0; i < 3; i++) {
|
||||
List<OcrBox> downs = BoxUtils.findNearestBoxes(
|
||||
current,
|
||||
boxList,
|
||||
BoxUtils.Direction.DOWN,
|
||||
1
|
||||
);
|
||||
if (downs == null || downs.isEmpty()) {
|
||||
break;
|
||||
}
|
||||
OcrBox down = downs.get(0);
|
||||
if (!isLikelyAddressContinuation(current, down, itemByBox)) {
|
||||
break;
|
||||
}
|
||||
String lineText = buildAddressLine(down, itemByBox, boxList, usedBoxes);
|
||||
if (lineText.isEmpty()) {
|
||||
break;
|
||||
}
|
||||
if (isIdNumberLine(lineText)) {
|
||||
break;
|
||||
}
|
||||
appendMergedText(sb, lineText);
|
||||
current = down;
|
||||
}
|
||||
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
private String extractAddressAfterLabel(String text) {
|
||||
if (text == null || text.isEmpty()) {
|
||||
return "";
|
||||
}
|
||||
String clean = text.replace(" ", "");
|
||||
int idx = clean.indexOf("住址");
|
||||
if (idx >= 0) {
|
||||
return clean.substring(idx + 2);
|
||||
}
|
||||
idx = clean.indexOf("佳址");
|
||||
if (idx >= 0) {
|
||||
return clean.substring(idx + 2);
|
||||
}
|
||||
idx = clean.indexOf("往址");
|
||||
if (idx >= 0) {
|
||||
return clean.substring(idx + 2);
|
||||
}
|
||||
idx = clean.indexOf("地址");
|
||||
if (idx >= 0) {
|
||||
return clean.substring(idx + 2);
|
||||
}
|
||||
return "";
|
||||
}
|
||||
|
||||
private void appendMergedText(StringBuilder sb, String nextText) {
|
||||
if (nextText == null || nextText.isEmpty()) {
|
||||
return;
|
||||
}
|
||||
if (sb.length() == 0) {
|
||||
sb.append(nextText);
|
||||
return;
|
||||
}
|
||||
String existing = sb.toString();
|
||||
int maxOverlap = Math.min(existing.length(), nextText.length());
|
||||
for (int overlap = maxOverlap; overlap > 0; overlap--) {
|
||||
if (existing.regionMatches(existing.length() - overlap, nextText, 0, overlap)) {
|
||||
sb.append(nextText.substring(overlap));
|
||||
return;
|
||||
}
|
||||
}
|
||||
sb.append(nextText);
|
||||
}
|
||||
|
||||
private boolean isLikelyAddressContinuation(OcrBox current, OcrBox candidate, Map<OcrBox, OcrItem> itemByBox) {
|
||||
if (current == null || candidate == null) {
|
||||
return false;
|
||||
}
|
||||
OcrItem candidateItem = itemByBox.get(candidate);
|
||||
if (candidateItem == null) {
|
||||
return false;
|
||||
}
|
||||
String text = candidateItem.getText();
|
||||
if (text == null || text.trim().isEmpty()) {
|
||||
return false;
|
||||
}
|
||||
String compact = text.replace(" ", "");
|
||||
if (!compact.matches(".*[\\u4e00-\\u9fa5].*")) {
|
||||
return false;
|
||||
}
|
||||
if (compact.matches("[A-Za-z]+")) {
|
||||
return false;
|
||||
}
|
||||
|
||||
double currentBottom = Math.max(Math.max(current.getBottomLeft().getY(), current.getBottomRight().getY()),
|
||||
Math.max(current.getTopLeft().getY(), current.getTopRight().getY()));
|
||||
double candidateTop = Math.min(Math.min(candidate.getTopLeft().getY(), candidate.getTopRight().getY()),
|
||||
Math.min(candidate.getBottomLeft().getY(), candidate.getBottomRight().getY()));
|
||||
double verticalGap = candidateTop - currentBottom;
|
||||
double allowedGap = Math.max(boxHeight(current), boxHeight(candidate)) * 1.2;
|
||||
if (verticalGap > allowedGap) {
|
||||
return false;
|
||||
}
|
||||
|
||||
double currentCenterX = boxCenter(current).getX();
|
||||
double candidateCenterX = boxCenter(candidate).getX();
|
||||
double centerDeltaX = Math.abs(candidateCenterX - currentCenterX);
|
||||
double allowedDeltaX = Math.max(boxWidth(current), boxWidth(candidate)) * 0.9;
|
||||
return centerDeltaX <= allowedDeltaX;
|
||||
}
|
||||
|
||||
/**
|
||||
* 根据某一行的起始框，向右拼接同一行上的多个检测框文本
|
||||
*/
|
||||
private String buildAddressLine(OcrBox rowAnchor,
|
||||
Map<OcrBox, OcrItem> itemByBox,
|
||||
List<OcrBox> boxList,
|
||||
Set<OcrBox> usedBoxes) {
|
||||
if (rowAnchor == null) {
|
||||
return "";
|
||||
}
|
||||
StringBuilder sb = new StringBuilder();
|
||||
|
||||
OcrItem anchorItem = itemByBox.get(rowAnchor);
|
||||
if (anchorItem != null && anchorItem.getText() != null && !usedBoxes.contains(rowAnchor)) {
|
||||
sb.append(anchorItem.getText().replace(" ", ""));
|
||||
usedBoxes.add(rowAnchor);
|
||||
}
|
||||
|
||||
List<OcrBox> rightBoxes = BoxUtils.findNearestBoxes(rowAnchor, boxList, BoxUtils.Direction.RIGHT, 8);
|
||||
if (rightBoxes != null && !rightBoxes.isEmpty()) {
|
||||
for (OcrBox rb : rightBoxes) {
|
||||
if (usedBoxes.contains(rb)) {
|
||||
continue;
|
||||
}
|
||||
OcrItem item = itemByBox.get(rb);
|
||||
if (item != null && item.getText() != null) {
|
||||
sb.append(item.getText().replace(" ", ""));
|
||||
usedBoxes.add(rb);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return sb.toString();
|
||||
}
|
||||
|
||||
/**
|
||||
* 判断一行文本是否疑似“公民身份号码”行
|
||||
*/
|
||||
private boolean isIdNumberLine(String text) {
|
||||
if (text == null || text.isEmpty()) {
|
||||
return false;
|
||||
}
|
||||
String norm = normalizeText(text);
|
||||
if (norm.contains("公民身份号码") || norm.contains("公民身份號碼") || norm.toUpperCase().contains("ID")) {
|
||||
return true;
|
||||
}
|
||||
return extractIdNumber(norm) != null;
|
||||
}
|
||||
|
||||
private Map<OcrBox, OcrItem> buildBoxItemMap(List<OcrItem> items) {
|
||||
Map<OcrBox, OcrItem> itemByBox = new IdentityHashMap<>();
|
||||
if (items == null) {
|
||||
return itemByBox;
|
||||
}
|
||||
for (OcrItem item : items) {
|
||||
if (item != null && item.getOcrBox() != null) {
|
||||
itemByBox.put(item.getOcrBox(), item);
|
||||
}
|
||||
}
|
||||
return itemByBox;
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,68 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
import cn.smartjavaai.ocr.entity.OcrItem;
|
||||
|
||||
import java.util.Arrays;
|
||||
import java.util.Collections;
|
||||
import java.util.HashSet;
|
||||
import java.util.List;
|
||||
import java.util.Objects;
|
||||
import java.util.Set;
|
||||
import java.util.stream.Collectors;
|
||||
|
||||
/**
|
||||
* 身份证 OCR 解析相关的通用工具。
|
||||
*/
|
||||
final class IdCardOcrUtils {
|
||||
|
||||
static final Set<String> ETHNIC_SET = Collections.unmodifiableSet(new HashSet<>(Arrays.asList(
|
||||
"汉", "蒙古", "回", "藏", "维吾尔", "苗", "彝", "壮", "布依", "朝鲜", "满", "侗",
|
||||
"瑶", "白", "土家", "哈尼", "哈萨克", "傣", "黎", "傈僳", "佤", "畲", "高山",
|
||||
"拉祜", "水", "东乡", "纳西", "景颇", "柯尔克孜", "土", "达斡尔", "仫佬",
|
||||
"羌", "布朗", "撒拉", "毛南", "仡佬", "锡伯", "阿昌", "普米", "塔吉克", "怒",
|
||||
"乌孜别克", "俄罗斯", "鄂温克", "德昂", "保安", "裕固", "京", "塔塔尔", "独龙",
|
||||
"鄂伦春", "赫哲", "门巴", "珞巴", "基诺"
|
||||
)));
|
||||
|
||||
private IdCardOcrUtils() {
|
||||
}
|
||||
|
||||
static List<OcrItem> filterSmallBoxes(List<OcrItem> items, double factor) {
|
||||
if (items == null || items.isEmpty()) {
|
||||
return items;
|
||||
}
|
||||
|
||||
List<Double> areas = items.stream()
|
||||
.map(OcrItem::getOcrBox)
|
||||
.filter(Objects::nonNull)
|
||||
.map(IdCardOcrUtils::estimateBoxArea)
|
||||
.sorted()
|
||||
.collect(Collectors.toList());
|
||||
|
||||
if (areas.size() < 5) {
|
||||
return items;
|
||||
}
|
||||
|
||||
double median = areas.get(areas.size() / 2);
|
||||
double threshold = median * factor;
|
||||
List<OcrItem> result = items.stream()
|
||||
.filter(it -> it.getOcrBox() == null || estimateBoxArea(it.getOcrBox()) >= threshold)
|
||||
.collect(Collectors.toList());
|
||||
return result.isEmpty() ? items : result;
|
||||
}
|
||||
|
||||
static double estimateBoxArea(OcrBox box) {
|
||||
if (box == null) {
|
||||
return 0.0;
|
||||
}
|
||||
float[] pts = box.toFloatArray();
|
||||
float minX = Math.min(Math.min(pts[0], pts[2]), Math.min(pts[4], pts[6]));
|
||||
float minY = Math.min(Math.min(pts[1], pts[3]), Math.min(pts[5], pts[7]));
|
||||
float maxX = Math.max(Math.max(pts[0], pts[2]), Math.max(pts[4], pts[6]));
|
||||
float maxY = Math.max(Math.max(pts[1], pts[3]), Math.max(pts[5], pts[7]));
|
||||
double width = Math.max(1.0, maxX - minX);
|
||||
double height = Math.max(1.0, maxY - minY);
|
||||
return width * height;
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,15 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
|
||||
/**
|
||||
* 身份证解析接口，用于从 OCR 结果中解析结构化字段。
|
||||
*/
|
||||
public interface IdCardParser<T> {
|
||||
|
||||
/**
|
||||
* 从 OCR 结果中解析出指定类型的身份证信息
|
||||
*/
|
||||
T parse(OcrInfo ocrInfo);
|
||||
}
|
||||
|
||||
@@ -0,0 +1,29 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
|
||||
/**
|
||||
* 身份证预处理调试监听器。
|
||||
*
|
||||
* 在预处理或识别流程中，只有当某个阶段真的产出了中间结果时，才会回调当前图像，
|
||||
* 便于测试环境中将中间结果保存到磁盘进行对比和调试。
|
||||
*
|
||||
* 生产环境可不设置监听器，完全不影响正常流程。
|
||||
*/
|
||||
public interface IdCardPreprocessListener {
|
||||
|
||||
/**
|
||||
* 方向矫正完成后回调。
|
||||
* 如果方向检测结果为 0 且未发生旋转，则不会调用。
|
||||
*/
|
||||
default void onAfterDirection(Image image) {}
|
||||
|
||||
/**
|
||||
* OCR 识别完成后回调。
|
||||
*
|
||||
* 传入的 image 为预处理后的图片副本，已绘制 OCR 检测框，便于直接保存调试结果。
|
||||
* 如果识别流程失败或未进入识别阶段，则不会调用。
|
||||
*/
|
||||
default void onAfterRecognize(Image image, OcrInfo ocrInfo) {}
|
||||
}
|
||||
@@ -0,0 +1,26 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
import lombok.Data;
|
||||
import lombok.experimental.Accessors;
|
||||
|
||||
import java.util.List;
|
||||
|
||||
/**
|
||||
* 身份证预处理结果。
|
||||
*
|
||||
* - processedImage: 预处理后的图像；若无需旋转，则通常与原图一致
|
||||
* - reusableBoxes: 可复用的文本检测框；仅在无需旋转时可直接复用
|
||||
* - rotated: 是否进行了整图旋转
|
||||
*/
|
||||
@Data
|
||||
@Accessors(chain = true)
|
||||
public class IdCardPreprocessResult {
|
||||
|
||||
private Image processedImage;
|
||||
|
||||
private List<OcrBox> reusableBoxes;
|
||||
|
||||
private boolean rotated;
|
||||
}
|
||||
@@ -0,0 +1,227 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
import cn.smartjavaai.ocr.entity.OcrItem;
|
||||
import cn.smartjavaai.ocr.enums.AngleEnum;
|
||||
import cn.smartjavaai.ocr.model.common.detect.OcrCommonDetModel;
|
||||
import cn.smartjavaai.ocr.model.common.direction.OcrDirectionModel;
|
||||
import cn.smartjavaai.ocr.utils.OcrUtils;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
import org.opencv.core.Mat;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.List;
|
||||
|
||||
/**
|
||||
* 身份证图像预处理：方向矫正（0/90/180/270°）。
|
||||
*
|
||||
* - 方向矫正：使用 OcrDirectionModel 检测整图方向并旋转至正向；
|
||||
*/
|
||||
@Slf4j
|
||||
public class IdCardPreprocessor {
|
||||
|
||||
/**
|
||||
* 预处理：
|
||||
* 1. 方向矫正（若提供 directionModel）：纠正 0/90/180/270° 整体方向；
|
||||
* 2. 若提供 listener，则在阶段结束后回调当前图像，便于测试保存中间结果。
|
||||
*/
|
||||
public Image preprocess(Image src,
|
||||
OcrDirectionModel directionModel,
|
||||
IdCardPreprocessListener listener) {
|
||||
if (src == null) {
|
||||
return src;
|
||||
}
|
||||
if (directionModel != null) {
|
||||
return applyDirectionCorrection(src, directionModel, listener);
|
||||
}
|
||||
return src;
|
||||
}
|
||||
|
||||
/**
|
||||
* 预处理（不关心调试监听器的常规调用入口）。
|
||||
*/
|
||||
public Image preprocess(Image src,
|
||||
OcrDirectionModel directionModel) {
|
||||
return preprocess(src, directionModel, null);
|
||||
}
|
||||
|
||||
/**
|
||||
* 预处理增强版：
|
||||
* 1. 先做文本检测
|
||||
* 2. 过滤明显过小的文本框，降低方向判断噪声
|
||||
* 3. 使用已有文本框做方向检测
|
||||
* 4. 若无需旋转，则直接返回可复用的检测框
|
||||
* 5. 若需要旋转，则返回旋转后的图像，检测框不复用
|
||||
*/
|
||||
public IdCardPreprocessResult preprocess(Image src,
|
||||
OcrCommonDetModel textDetModel,
|
||||
OcrDirectionModel directionModel,
|
||||
IdCardPreprocessListener listener) {
|
||||
IdCardPreprocessResult result = new IdCardPreprocessResult()
|
||||
.setProcessedImage(src)
|
||||
.setRotated(false);
|
||||
if (src == null) {
|
||||
return result;
|
||||
}
|
||||
if (textDetModel == null || directionModel == null) {
|
||||
if (directionModel != null) {
|
||||
result.setProcessedImage(preprocess(src, directionModel, listener));
|
||||
result.setRotated(result.getProcessedImage() != src);
|
||||
}
|
||||
return result;
|
||||
}
|
||||
Mat srcMat = null;
|
||||
try {
|
||||
srcMat = cn.smartjavaai.common.utils.ImageUtils.toMat(src).clone();
|
||||
List<OcrBox> detectedBoxes = textDetModel.detect(src);
|
||||
List<OcrBox> filteredBoxes = filterSmallBoxes(detectedBoxes);
|
||||
if (filteredBoxes.isEmpty()) {
|
||||
result.setReusableBoxes(detectedBoxes);
|
||||
return result;
|
||||
}
|
||||
List<OcrItem> directionItems = directionModel.detect(filteredBoxes, srcMat);
|
||||
AngleEnum dominant = dominantAngle(directionItems);
|
||||
if (dominant != null && dominant != AngleEnum.ANGLE_0) {
|
||||
log.debug("方向检测结果为 {}，执行整图旋转矫正。", dominant);
|
||||
Image rotated = OcrUtils.rotateImg(src, dominant);
|
||||
if (listener != null) {
|
||||
listener.onAfterDirection(rotated);
|
||||
}
|
||||
return result.setProcessedImage(rotated)
|
||||
.setReusableBoxes(null)
|
||||
.setRotated(true);
|
||||
}
|
||||
return result.setReusableBoxes(detectedBoxes);
|
||||
} catch (Exception e) {
|
||||
log.warn("身份证方向矫正失败，使用原图继续。", e);
|
||||
return result;
|
||||
} finally {
|
||||
if (srcMat != null) {
|
||||
srcMat.release();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* 使用方向模型检测整图方向（0/90/180/270°），并将图片旋转至正向。
|
||||
*/
|
||||
private Image applyDirectionCorrection(Image src,
|
||||
OcrDirectionModel directionModel,
|
||||
IdCardPreprocessListener listener) {
|
||||
if (src == null || directionModel == null) {
|
||||
return src;
|
||||
}
|
||||
try {
|
||||
long detectStart = System.nanoTime();
|
||||
List<OcrItem> items = directionModel.detect(src);
|
||||
log.debug("身份证方向模型调用耗时={}ms, 检测框数量={}", elapsedMillis(detectStart), items == null ? 0 : items.size());
|
||||
log.debug("方向检测结果：{}", items);
|
||||
AngleEnum dominant = dominantAngle(items);
|
||||
if (dominant != null && dominant != AngleEnum.ANGLE_0) {
|
||||
log.info("方向检测结果为 {}，执行整图旋转矫正。", dominant);
|
||||
long rotateStart = System.nanoTime();
|
||||
Image rotated = OcrUtils.rotateImg(src, dominant);
|
||||
log.debug("身份证整图旋转耗时={}ms", elapsedMillis(rotateStart));
|
||||
if (listener != null) {
|
||||
listener.onAfterDirection(rotated);
|
||||
}
|
||||
return rotated;
|
||||
}
|
||||
} catch (Exception e) {
|
||||
log.warn("身份证方向矫正失败，使用原图继续。", e);
|
||||
}
|
||||
return src;
|
||||
}
|
||||
|
||||
/**
|
||||
* 从方向检测结果中取主角度：
|
||||
* 1. 先过滤掉明显的小框，减少碎框噪声；
|
||||
* 2. 再按 面积 * score 做加权投票；
|
||||
* 3. 若过滤后为空，则退化为对全部框做加权投票。
|
||||
*/
|
||||
private AngleEnum dominantAngle(List<OcrItem> items) {
|
||||
if (items == null || items.isEmpty()) {
|
||||
return null;
|
||||
}
|
||||
double maxArea = 0.0;
|
||||
for (OcrItem item : items) {
|
||||
if (item == null || item.getAngle() == null || item.getOcrBox() == null) {
|
||||
continue;
|
||||
}
|
||||
maxArea = Math.max(maxArea, estimateBoxArea(item.getOcrBox()));
|
||||
}
|
||||
if (maxArea <= 0.0) {
|
||||
return weightedVote(items, 0.0);
|
||||
}
|
||||
|
||||
double minArea = maxArea * 0.15;
|
||||
AngleEnum dominant = weightedVote(items, minArea);
|
||||
if (dominant != null) {
|
||||
return dominant;
|
||||
}
|
||||
return weightedVote(items, 0.0);
|
||||
}
|
||||
|
||||
private AngleEnum weightedVote(List<OcrItem> items, double minArea) {
|
||||
AngleEnum first = null;
|
||||
double weight0 = 0.0, weight90 = 0.0, weight180 = 0.0, weight270 = 0.0;
|
||||
for (OcrItem item : items) {
|
||||
if (item == null || item.getAngle() == null || item.getOcrBox() == null) {
|
||||
continue;
|
||||
}
|
||||
double area = estimateBoxArea(item.getOcrBox());
|
||||
if (area < minArea) {
|
||||
continue;
|
||||
}
|
||||
double score = item.getScore() > 0 ? item.getScore() : 1.0;
|
||||
double weight = area * score;
|
||||
AngleEnum angle = item.getAngle();
|
||||
if (first == null) {
|
||||
first = angle;
|
||||
}
|
||||
switch (angle) {
|
||||
case ANGLE_0: weight0 += weight; break;
|
||||
case ANGLE_90: weight90 += weight; break;
|
||||
case ANGLE_180: weight180 += weight; break;
|
||||
case ANGLE_270: weight270 += weight; break;
|
||||
}
|
||||
}
|
||||
double maxWeight = Math.max(Math.max(weight0, weight90), Math.max(weight180, weight270));
|
||||
if (maxWeight <= 0.0) {
|
||||
return first;
|
||||
}
|
||||
if (Double.compare(weight90, maxWeight) == 0) return AngleEnum.ANGLE_90;
|
||||
if (Double.compare(weight270, maxWeight) == 0) return AngleEnum.ANGLE_270;
|
||||
if (Double.compare(weight180, maxWeight) == 0) return AngleEnum.ANGLE_180;
|
||||
if (Double.compare(weight0, maxWeight) == 0) return AngleEnum.ANGLE_0;
|
||||
return first;
|
||||
}
|
||||
|
||||
private List<OcrBox> filterSmallBoxes(List<OcrBox> boxes) {
|
||||
if (boxes == null || boxes.isEmpty()) {
|
||||
return boxes;
|
||||
}
|
||||
List<OcrItem> items = new ArrayList<>(boxes.size());
|
||||
for (OcrBox box : boxes) {
|
||||
items.add(new OcrItem(box, AngleEnum.ANGLE_0, 1.0f));
|
||||
}
|
||||
List<OcrItem> filteredItems = IdCardOcrUtils.filterSmallBoxes(items, 0.15);
|
||||
List<OcrBox> filteredBoxes = new ArrayList<>(filteredItems.size());
|
||||
for (OcrItem item : filteredItems) {
|
||||
if (item != null && item.getOcrBox() != null) {
|
||||
filteredBoxes.add(item.getOcrBox());
|
||||
}
|
||||
}
|
||||
return filteredBoxes.isEmpty() ? boxes : filteredBoxes;
|
||||
}
|
||||
|
||||
private double estimateBoxArea(OcrBox box) {
|
||||
return IdCardOcrUtils.estimateBoxArea(box);
|
||||
}
|
||||
|
||||
private long elapsedMillis(long startNanos) {
|
||||
return (System.nanoTime() - startNanos) / 1_000_000;
|
||||
}
|
||||
|
||||
}
|
||||
@@ -0,0 +1,44 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import ai.djl.modality.cv.Image;
|
||||
import cn.smartjavaai.ocr.entity.IdCardBackInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardFrontInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardInfo;
|
||||
import cn.smartjavaai.ocr.entity.OcrInfo;
|
||||
|
||||
/**
|
||||
* 身份证 OCR 识别服务接口
|
||||
*
|
||||
* 支持：
|
||||
* 1. 直接传入图片，由服务内部完成预处理 + OCR + 解析；
|
||||
* 2. 已有 OCR 结果的情况下，仅做单面结构化解析。
|
||||
*/
|
||||
public interface IdCardRecognizer {
|
||||
|
||||
/**
|
||||
* 识别身份证正面（从图片开始）
|
||||
*/
|
||||
IdCardFrontInfo recognizeFront(Image image);
|
||||
|
||||
/**
|
||||
* 识别身份证反面（从图片开始）
|
||||
*/
|
||||
IdCardBackInfo recognizeBack(Image image);
|
||||
|
||||
/**
|
||||
* 识别身份证正面（仅结构化已有 OCR 结果）
|
||||
*/
|
||||
IdCardFrontInfo recognizeFront(OcrInfo ocrInfo);
|
||||
|
||||
/**
|
||||
* 识别身份证反面（仅结构化已有 OCR 结果）
|
||||
*/
|
||||
IdCardBackInfo recognizeBack(OcrInfo ocrInfo);
|
||||
|
||||
/**
|
||||
* 同时识别身份证正反面（分别传入两面图片）
|
||||
*/
|
||||
IdCardInfo recognizeBoth(Image frontImage, Image backImage);
|
||||
|
||||
|
||||
}
|
||||
@@ -0,0 +1,79 @@
|
||||
package cn.smartjavaai.ocr.idcard;
|
||||
|
||||
import cn.smartjavaai.ocr.entity.IdCardBackInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardFrontInfo;
|
||||
import cn.smartjavaai.ocr.entity.IdCardInfo;
|
||||
import lombok.extern.slf4j.Slf4j;
|
||||
|
||||
/**
|
||||
* 身份证字段校验工具。
|
||||
*
|
||||
* 提供：
|
||||
* - 正面字段完整性与格式校验
|
||||
* - 反面字段简单格式校验（预留）
|
||||
* - 身份证号码校验位校验
|
||||
*/
|
||||
@Slf4j
|
||||
public class IdCardValidator {
|
||||
|
||||
public boolean validateFront(IdCardFrontInfo info) {
|
||||
if (info == null) {
|
||||
return false;
|
||||
}
|
||||
boolean nameOk = info.getName() != null
|
||||
&& info.getName().length() >= 2
|
||||
&& info.getName().length() <= 4;
|
||||
boolean genderOk = "男".equals(info.getGender()) || "女".equals(info.getGender());
|
||||
boolean ethnicOk = info.getEthnicity() != null && IdCardOcrUtils.ETHNIC_SET.contains(info.getEthnicity());
|
||||
boolean idOk = validateIdNumber(info.getIdNumber());
|
||||
|
||||
log.debug("身份证正面字段校验详情：nameOk={}, genderOk={}, ethnicOk={}, idOk={}",
|
||||
nameOk, genderOk, ethnicOk, idOk);
|
||||
return nameOk && genderOk && ethnicOk && idOk;
|
||||
}
|
||||
|
||||
public boolean validateBack(IdCardBackInfo info) {
|
||||
// 目前只做存在性校验，后续可以根据业务需要增加日期格式、范围等更严格校验
|
||||
if (info == null) {
|
||||
return false;
|
||||
}
|
||||
boolean authorityOk = info.getIssuingAuthority() != null && !info.getIssuingAuthority().isEmpty();
|
||||
boolean validFromOk = info.getValidFrom() != null && !info.getValidFrom().isEmpty();
|
||||
boolean validToOk = info.getValidTo() != null && !info.getValidTo().isEmpty();
|
||||
log.info("身份证反面字段校验详情：authorityOk={}, validFromOk={}, validToOk={}",
|
||||
authorityOk, validFromOk, validToOk);
|
||||
return authorityOk && validFromOk && validToOk;
|
||||
}
|
||||
|
||||
public boolean validate(IdCardInfo info) {
|
||||
if (info == null) {
|
||||
return false;
|
||||
}
|
||||
boolean frontOk = info.getFront() == null || validateFront(info.getFront());
|
||||
boolean backOk = info.getBack() == null || validateBack(info.getBack());
|
||||
return frontOk && backOk;
|
||||
}
|
||||
|
||||
/**
|
||||
* 简单的身份证号码格式 + 校验位校验
|
||||
*/
|
||||
public boolean validateIdNumber(String id) {
|
||||
if (id == null) {
|
||||
return false;
|
||||
}
|
||||
String upper = id.toUpperCase();
|
||||
if (!upper.matches("^[1-9]\\d{5}(19|20)\\d{2}(0[1-9]|1[0-2])(0[1-9]|[12]\\d|3[01])\\d{3}[0-9X]$")) {
|
||||
return false;
|
||||
}
|
||||
// 校验位
|
||||
char[] chars = upper.toCharArray();
|
||||
int[] weight = {7, 9, 10, 5, 8, 4, 2, 1, 6, 3, 7, 9, 10, 5, 8, 4, 2};
|
||||
char[] validateCode = {'1', '0', 'X', '9', '8', '7', '6', '5', '4', '3', '2'};
|
||||
int sum = 0;
|
||||
for (int i = 0; i < 17; i++) {
|
||||
sum += (chars[i] - '0') * weight[i];
|
||||
}
|
||||
int mod = sum % 11;
|
||||
return validateCode[mod] == chars[17];
|
||||
}
|
||||
}
|
||||
@@ -94,8 +94,10 @@ public class OcrCommonDetModelImpl implements OcrCommonDetModel{
|
||||
|
||||
@Override
|
||||
public List<OcrBox> detect(Image image){
|
||||
long start = System.nanoTime();
|
||||
List<Image> imageList = Collections.singletonList(image);
|
||||
List<List<OcrBox>> result = batchDetectDJLImage(imageList);
|
||||
log.debug("文本检测模型单图调用耗时={}ms, 检测框数量={}", elapsedMillis(start), result.isEmpty() ? 0 : result.get(0).size());
|
||||
return result.get(0);
|
||||
}
|
||||
|
||||
@@ -192,12 +194,19 @@ public class OcrCommonDetModelImpl implements OcrCommonDetModel{
|
||||
if(!ImageUtils.isAllImageSizeEqual(imageList)){
|
||||
throw new OcrException("图片尺寸不一致");
|
||||
}
|
||||
long totalStart = System.nanoTime();
|
||||
Predictor<Image, NDList> predictor = null;
|
||||
try (NDManager manager = NDManager.newBaseManager()) {
|
||||
predictor = detPredictorPool.borrowObject();
|
||||
List<NDList> result = predictor.batchPredict(imageList);
|
||||
result.forEach(ndList -> ndList.attach(manager));
|
||||
return OcrUtils.convertToOcrBox(result);
|
||||
List<List<OcrBox>> boxes = OcrUtils.convertToOcrBox(result);
|
||||
int totalBoxes = 0;
|
||||
for (List<OcrBox> boxList : boxes) {
|
||||
totalBoxes += boxList == null ? 0 : boxList.size();
|
||||
}
|
||||
log.debug("文本检测总耗时={}ms, batchSize={}, totalBoxes={}", elapsedMillis(totalStart), imageList.size(), totalBoxes);
|
||||
return boxes;
|
||||
} catch (Exception e) {
|
||||
throw new OcrException("OCR检测错误", e);
|
||||
}finally {
|
||||
@@ -257,5 +266,9 @@ public class OcrCommonDetModelImpl implements OcrCommonDetModel{
|
||||
return fromFactory;
|
||||
}
|
||||
|
||||
private long elapsedMillis(long startNanos) {
|
||||
return (System.nanoTime() - startNanos) / 1_000_000;
|
||||
}
|
||||
|
||||
|
||||
}
|
||||
|
||||
@@ -108,13 +108,13 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
if(Objects.isNull(textDetModel)){
|
||||
throw new OcrException("textDetModel is null");
|
||||
}
|
||||
//检测文本
|
||||
List<OcrBox> boxeList = textDetModel.detect(image);
|
||||
if(Objects.isNull(boxeList) || boxeList.isEmpty()){
|
||||
throw new OcrException("未检测到文本");
|
||||
}
|
||||
Mat srcMat = ImageUtils.toMat(image);
|
||||
return detect(boxeList, srcMat);
|
||||
List<OcrItem> result = detect(boxeList, srcMat);
|
||||
return result;
|
||||
}
|
||||
|
||||
|
||||
@@ -274,7 +274,9 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
List<Image> imageList = new ArrayList<Image>();
|
||||
List<Boolean> isRotatedList = new ArrayList<Boolean>();
|
||||
int index = 0;
|
||||
long totalStart = System.nanoTime();
|
||||
try (NDManager manager = model.getNDManager().newSubManager()){
|
||||
long prepareStart = System.nanoTime();
|
||||
for(int i = 0; i < srcMatList.size(); i++){
|
||||
for (int j = 0; j < boxList.get(i).size(); j++){
|
||||
//透视变换及裁剪
|
||||
@@ -292,8 +294,11 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
index++;
|
||||
}
|
||||
}
|
||||
log.debug("方向模型裁剪预处理耗时={}ms, batchSize={}, textBlocks={}", elapsedMillis(prepareStart), srcMatList.size(), imageList.size());
|
||||
List<List<OcrItem>> result = new ArrayList<>();
|
||||
long predictStart = System.nanoTime();
|
||||
List<DirectionInfo> directionInfos = batchDetect(imageList);
|
||||
log.debug("方向分类模型调用耗时={}ms, textBlocks={}", elapsedMillis(predictStart), imageList.size());
|
||||
//释放
|
||||
imageList.forEach(image -> ImageUtils.releaseOpenCVMat(image));
|
||||
if(CollectionUtils.isEmpty(directionInfos)){
|
||||
@@ -327,6 +332,7 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
}
|
||||
result.add(ocrItemList);
|
||||
}
|
||||
log.debug("方向模型总耗时={}ms, batchSize={}, textBlocks={}", elapsedMillis(totalStart), srcMatList.size(), index);
|
||||
return result;
|
||||
}
|
||||
}
|
||||
@@ -335,7 +341,8 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
Predictor<Image, DirectionInfo> predictor = null;
|
||||
try {
|
||||
predictor = predictorPool.borrowObject();
|
||||
return predictor.batchPredict(imageList);
|
||||
List<DirectionInfo> result = predictor.batchPredict(imageList);
|
||||
return result;
|
||||
} catch (Exception e) {
|
||||
throw new OcrException("OCR检测错误", e);
|
||||
}finally {
|
||||
@@ -399,4 +406,8 @@ public class PPOCRMobileV2ClsModel implements OcrDirectionModel {
|
||||
public boolean isFromFactory() {
|
||||
return fromFactory;
|
||||
}
|
||||
|
||||
private long elapsedMillis(long startNanos) {
|
||||
return (System.nanoTime() - startNanos) / 1_000_000;
|
||||
}
|
||||
}
|
||||
|
||||
@@ -61,6 +61,19 @@ public interface OcrCommonRecModel extends AutoCloseable{
|
||||
throw new UnsupportedOperationException("默认不支持该功能");
|
||||
}
|
||||
|
||||
/**
|
||||
* 基于已有文本检测框执行文本识别。
|
||||
* 适合调用方已经完成文本检测，想避免重复检测的场景。
|
||||
*
|
||||
* @param image 原图
|
||||
* @param boxList 已有文本检测框
|
||||
* @param options 识别选项
|
||||
* @return OCR 结果
|
||||
*/
|
||||
default OcrInfo recognize(Image image, List<OcrBox> boxList, OcrRecOptions options) {
|
||||
throw new UnsupportedOperationException("默认不支持该功能");
|
||||
}
|
||||
|
||||
|
||||
/**
|
||||
* 文本检测
|
||||
@@ -141,6 +154,18 @@ public interface OcrCommonRecModel extends AutoCloseable{
|
||||
throw new UnsupportedOperationException("默认不支持该功能");
|
||||
}
|
||||
|
||||
/**
|
||||
* 基于已有文本检测框执行批量文本识别。
|
||||
*
|
||||
* @param imageList 原图列表
|
||||
* @param boxList 每张图对应的文本检测框列表
|
||||
* @param options 识别选项
|
||||
* @return OCR 结果
|
||||
*/
|
||||
default List<OcrInfo> batchRecognizeDJLImage(List<Image> imageList, List<List<OcrBox>> boxList, OcrRecOptions options) {
|
||||
throw new UnsupportedOperationException("默认不支持该功能");
|
||||
}
|
||||
|
||||
default GenericObjectPool<Predictor<Image, String>> getPool() {
|
||||
throw new UnsupportedOperationException("默认不支持该功能");
|
||||
}
|
||||
|
||||
@@ -47,6 +47,8 @@ import java.util.stream.Collectors;
|
||||
@Slf4j
|
||||
public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
|
||||
private static final int REC_CHUNK_SIZE = 64;
|
||||
|
||||
private GenericObjectPool<Predictor<Image, String>> recPredictorPool;
|
||||
|
||||
private OcrRecModelConfig config;
|
||||
@@ -111,10 +113,27 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
*/
|
||||
@Override
|
||||
public OcrInfo recognize(Image image, OcrRecOptions options) {
|
||||
long start = System.nanoTime();
|
||||
List<OcrInfo> result = batchRecognizeDJLImage(Collections.singletonList(image), options);
|
||||
if (CollectionUtils.isEmpty(result)) {
|
||||
throw new OcrException("OCR识别结果为空");
|
||||
}
|
||||
log.debug("OCR识别单图总耗时={}ms", elapsedMillis(start));
|
||||
return result.get(0);
|
||||
}
|
||||
|
||||
@Override
|
||||
public OcrInfo recognize(Image image, List<OcrBox> boxList, OcrRecOptions options) {
|
||||
long start = System.nanoTime();
|
||||
List<OcrInfo> result = batchRecognizeDJLImage(
|
||||
Collections.singletonList(image),
|
||||
Collections.singletonList(boxList),
|
||||
options
|
||||
);
|
||||
if (CollectionUtils.isEmpty(result)) {
|
||||
throw new OcrException("OCR识别结果为空");
|
||||
}
|
||||
log.debug("OCR识别单图总耗时={}ms", elapsedMillis(start));
|
||||
return result.get(0);
|
||||
}
|
||||
|
||||
@@ -212,10 +231,8 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
//分行判断
|
||||
for (int i = 1; i < initList.size(); i++) {
|
||||
RotatedBoxCompX tmpBox = new RotatedBoxCompX(initList.get(i).getBox(), initList.get(i).getText());
|
||||
float y1 = firstBox.getBox().toFloatArray()[1];
|
||||
float y2 = tmpBox.getBox().toFloatArray()[1];
|
||||
float dis = Math.abs(y2 - y1);
|
||||
if (dis < 20) { // 认为是同 1 行 - Considered to be in the same line
|
||||
boolean isSameRow = OcrUtils.isSameRow(firstBox.getBox(), tmpBox.getBox());
|
||||
if (isSameRow) {
|
||||
line.add(tmpBox);
|
||||
} else { // 换行 - Line break
|
||||
firstBox = tmpBox;
|
||||
@@ -346,6 +363,11 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
|
||||
@Override
|
||||
public List<OcrInfo> batchRecognizeDJLImage(List<Image> imageList, OcrRecOptions options) {
|
||||
return batchRecognizeDJLImage(imageList, null, options);
|
||||
}
|
||||
|
||||
@Override
|
||||
public List<OcrInfo> batchRecognizeDJLImage(List<Image> imageList, List<List<OcrBox>> boxeList, OcrRecOptions options) {
|
||||
if (Objects.isNull(textDetModel)) {
|
||||
throw new OcrException("textDetModel is null");
|
||||
}
|
||||
@@ -356,64 +378,82 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
if (CollectionUtils.isEmpty(imageList)) {
|
||||
throw new OcrException("imageList is empty");
|
||||
}
|
||||
//检测文本
|
||||
List<List<OcrBox>> boxeList = textDetModel.batchDetectDJLImage(imageList);
|
||||
if (CollectionUtils.isEmpty(boxeList) || boxeList.size() != imageList.size()) {
|
||||
long totalStart = System.nanoTime();
|
||||
List<List<OcrBox>> effectiveBoxList = boxeList;
|
||||
if (CollectionUtils.isEmpty(effectiveBoxList)) {
|
||||
effectiveBoxList = textDetModel.batchDetectDJLImage(imageList);
|
||||
}
|
||||
if (CollectionUtils.isEmpty(effectiveBoxList) || effectiveBoxList.size() != imageList.size()) {
|
||||
throw new OcrException("未检测到文本");
|
||||
}
|
||||
Predictor<Image, String> predictor = null;
|
||||
List<OcrInfo> ocrInfoList = new ArrayList<OcrInfo>();
|
||||
try (NDManager manager = NDManager.newBaseManager()) {
|
||||
predictor = recPredictorPool.borrowObject();
|
||||
List<Image> allImageAlignList = new ArrayList<Image>();
|
||||
//检测方向
|
||||
if (ocrRecOptions.isEnableDirectionCorrect()) {
|
||||
if (Objects.isNull(directionModel)) {
|
||||
throw new OcrException("请配置方向模型");
|
||||
}
|
||||
List<Mat> matList = imageList.stream()
|
||||
.map(image -> ImageUtils.toMat(image))
|
||||
.collect(Collectors.toList());
|
||||
List<List<OcrItem>> ocrItemList = directionModel.batchDetect(boxeList, matList);
|
||||
if (CollectionUtils.isEmpty(ocrItemList) || ocrItemList.size() != imageList.size()) {
|
||||
throw new OcrException("方向检测失败");
|
||||
}
|
||||
allImageAlignList = new ArrayList<Image>();
|
||||
for (int i = 0; i < ocrItemList.size(); i++) {
|
||||
Mat srcMat = ImageUtils.toMat(imageList.get(i));
|
||||
List<Image> imageAlignList = batchAlignWithDirection(ocrItemList.get(i), srcMat, manager);
|
||||
// for(int j = 0; j < imageAlignList.size(); j++){
|
||||
// ImageUtils.saveImage(imageAlignList.get(j),"dir-"+i+"-"+j+".png","/Users/xxx/Downloads/testing33");
|
||||
// }
|
||||
allImageAlignList.addAll(imageAlignList);
|
||||
long directionStart = System.nanoTime();
|
||||
List<Mat> matList = new ArrayList<>(imageList.size());
|
||||
try {
|
||||
for (Image image : imageList) {
|
||||
matList.add(ImageUtils.toMat(image));
|
||||
}
|
||||
List<List<OcrItem>> ocrItemList = directionModel.batchDetect(effectiveBoxList, matList);
|
||||
log.debug("OCR流程-文本方向分类耗时={}ms, batchSize={}", elapsedMillis(directionStart), imageList.size());
|
||||
if (CollectionUtils.isEmpty(ocrItemList) || ocrItemList.size() != imageList.size()) {
|
||||
throw new OcrException("方向检测失败");
|
||||
}
|
||||
long alignStart = System.nanoTime();
|
||||
List<String> textList = new ArrayList<>();
|
||||
List<Image> chunkImages = new ArrayList<>(REC_CHUNK_SIZE);
|
||||
for (int i = 0; i < ocrItemList.size(); i++) {
|
||||
Mat srcMat = matList.get(i);
|
||||
List<Image> imageAlignList = batchAlignWithDirection(ocrItemList.get(i), srcMat, manager);
|
||||
for (Image alignImage : imageAlignList) {
|
||||
chunkImages.add(alignImage);
|
||||
if (chunkImages.size() >= REC_CHUNK_SIZE) {
|
||||
textList.addAll(batchRecognizeChunk(predictor, chunkImages));
|
||||
}
|
||||
}
|
||||
}
|
||||
log.debug("OCR流程-方向矫正裁剪耗时={}ms, textBlocks={}", elapsedMillis(alignStart), textList.size() + chunkImages.size());
|
||||
long recStart = System.nanoTime();
|
||||
if (!chunkImages.isEmpty()) {
|
||||
textList.addAll(batchRecognizeChunk(predictor, chunkImages));
|
||||
}
|
||||
log.debug("OCR流程-识别模型调用耗时={}ms, textBlocks={}", elapsedMillis(recStart), textList.size());
|
||||
return buildOcrInfoList(effectiveBoxList, ocrRecOptions, manager, textList, imageList.size(), totalStart);
|
||||
} finally {
|
||||
releaseTemporaryMats(imageList, matList);
|
||||
}
|
||||
} else {
|
||||
for (int i = 0; i < boxeList.size(); i++) {
|
||||
Mat srcMat = ImageUtils.toMat(imageList.get(i));
|
||||
List<Image> imageAlignList = batchAlign(boxeList.get(i), srcMat, manager);
|
||||
// for(int j = 0; j < imageAlignList.size(); j++){
|
||||
// ImageUtils.saveImage(imageAlignList.get(j),i+"-"+j+".png","/Users/wenjie/Downloads/testing33");
|
||||
// }
|
||||
allImageAlignList.addAll(imageAlignList);
|
||||
}
|
||||
}
|
||||
List<String> textList = batchRecognize(allImageAlignList);
|
||||
int textIndex = 0;
|
||||
for (int i = 0; i < boxeList.size(); i++) {
|
||||
List<RotatedBox> rotatedBoxes = new ArrayList<>();
|
||||
for (int j = 0; j < boxeList.get(i).size(); j++) {
|
||||
if (textIndex >= textList.size()) {
|
||||
throw new OcrException("识别失败: 第" + i + "张图片, 第" + j + "个文本块，未识别到文本");
|
||||
List<String> textList = new ArrayList<>();
|
||||
List<Image> chunkImages = new ArrayList<>(REC_CHUNK_SIZE);
|
||||
for (int i = 0; i < effectiveBoxList.size(); i++) {
|
||||
Mat srcMat = null;
|
||||
try {
|
||||
srcMat = ImageUtils.toMat(imageList.get(i));
|
||||
List<Image> imageAlignList = batchAlign(effectiveBoxList.get(i), srcMat, manager);
|
||||
for (Image alignImage : imageAlignList) {
|
||||
chunkImages.add(alignImage);
|
||||
if (chunkImages.size() >= REC_CHUNK_SIZE) {
|
||||
textList.addAll(batchRecognizeChunk(predictor, chunkImages));
|
||||
}
|
||||
}
|
||||
} finally {
|
||||
releaseTemporaryMat(imageList.get(i), srcMat);
|
||||
}
|
||||
OcrBox box = boxeList.get(i).get(j);
|
||||
NDArray pointsArray = manager.create(box.toFloatArray());
|
||||
rotatedBoxes.add(new RotatedBox(pointsArray, textList.get(textIndex)));
|
||||
textIndex++;
|
||||
}
|
||||
OcrInfo ocrInfo = postProcessOcrResult(rotatedBoxes, ocrRecOptions);
|
||||
ocrInfoList.add(ocrInfo);
|
||||
long recStart = System.nanoTime();
|
||||
if (!chunkImages.isEmpty()) {
|
||||
textList.addAll(batchRecognizeChunk(predictor, chunkImages));
|
||||
}
|
||||
log.debug("OCR流程-识别模型调用耗时={}ms, textBlocks={}", elapsedMillis(recStart), textList.size());
|
||||
return buildOcrInfoList(effectiveBoxList, ocrRecOptions, manager, textList, imageList.size(), totalStart);
|
||||
}
|
||||
return ocrInfoList;
|
||||
} catch (Exception e) {
|
||||
throw new OcrException("OCR检测错误", e);
|
||||
} finally {
|
||||
@@ -432,28 +472,41 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
}
|
||||
}
|
||||
|
||||
private List<String> batchRecognize(List<Image> imageAlignList) {
|
||||
Predictor<Image, String> predictor = null;
|
||||
private List<OcrInfo> buildOcrInfoList(List<List<OcrBox>> effectiveBoxList,
|
||||
OcrRecOptions ocrRecOptions,
|
||||
NDManager manager,
|
||||
List<String> textList,
|
||||
int batchSize,
|
||||
long totalStart) {
|
||||
int textIndex = 0;
|
||||
List<OcrInfo> ocrInfoList = new ArrayList<>();
|
||||
for (int i = 0; i < effectiveBoxList.size(); i++) {
|
||||
List<RotatedBox> rotatedBoxes = new ArrayList<>();
|
||||
for (int j = 0; j < effectiveBoxList.get(i).size(); j++) {
|
||||
if (textIndex >= textList.size()) {
|
||||
throw new OcrException("识别失败: 第" + i + "张图片, 第" + j + "个文本块，未识别到文本");
|
||||
}
|
||||
OcrBox box = effectiveBoxList.get(i).get(j);
|
||||
NDArray pointsArray = manager.create(box.toFloatArray());
|
||||
rotatedBoxes.add(new RotatedBox(pointsArray, textList.get(textIndex)));
|
||||
textIndex++;
|
||||
}
|
||||
OcrInfo ocrInfo = postProcessOcrResult(rotatedBoxes, ocrRecOptions);
|
||||
ocrInfoList.add(ocrInfo);
|
||||
}
|
||||
log.debug("OCR流程总耗时={}ms, batchSize={}", elapsedMillis(totalStart), batchSize);
|
||||
return ocrInfoList;
|
||||
}
|
||||
|
||||
private List<String> batchRecognizeChunk(Predictor<Image, String> predictor, List<Image> imageAlignList) {
|
||||
try {
|
||||
predictor = recPredictorPool.borrowObject();
|
||||
List<String> textList = predictor.batchPredict(imageAlignList);
|
||||
imageAlignList.forEach(subImg -> ImageUtils.releaseOpenCVMat(subImg));
|
||||
return textList;
|
||||
} catch (Exception e) {
|
||||
throw new OcrException("OCR检测错误", e);
|
||||
} finally {
|
||||
if (predictor != null) {
|
||||
try {
|
||||
recPredictorPool.returnObject(predictor); //归还
|
||||
} catch (Exception e) {
|
||||
log.warn("归还Predictor失败", e);
|
||||
try {
|
||||
predictor.close(); // 归还失败才销毁
|
||||
} catch (Exception ex) {
|
||||
log.error("关闭Predictor失败", ex);
|
||||
}
|
||||
}
|
||||
}
|
||||
imageAlignList.forEach(ImageUtils::releaseOpenCVMat);
|
||||
imageAlignList.clear();
|
||||
}
|
||||
}
|
||||
|
||||
@@ -525,4 +578,24 @@ public class OcrCommonRecModelImpl implements OcrCommonRecModel {
|
||||
public boolean isFromFactory() {
|
||||
return fromFactory;
|
||||
}
|
||||
|
||||
private long elapsedMillis(long startNanos) {
|
||||
return (System.nanoTime() - startNanos) / 1_000_000;
|
||||
}
|
||||
|
||||
private void releaseTemporaryMats(List<Image> imageList, List<Mat> matList) {
|
||||
int size = Math.min(imageList.size(), matList.size());
|
||||
for (int i = 0; i < size; i++) {
|
||||
releaseTemporaryMat(imageList.get(i), matList.get(i));
|
||||
}
|
||||
}
|
||||
|
||||
private void releaseTemporaryMat(Image image, Mat mat) {
|
||||
if (mat == null || image == null) {
|
||||
return;
|
||||
}
|
||||
if (!(image.getWrappedImage() instanceof Mat)) {
|
||||
mat.release();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
278
ocr/src/main/java/cn/smartjavaai/ocr/utils/BoxUtils.java
Normal file
278
ocr/src/main/java/cn/smartjavaai/ocr/utils/BoxUtils.java
Normal file
@@ -0,0 +1,278 @@
|
||||
package cn.smartjavaai.ocr.utils;
|
||||
|
||||
import cn.smartjavaai.common.entity.Point;
|
||||
import cn.smartjavaai.ocr.entity.OcrBox;
|
||||
|
||||
import java.util.ArrayList;
|
||||
import java.util.Comparator;
|
||||
import java.util.List;
|
||||
|
||||
/**
|
||||
* @author dwj
|
||||
* @date 2026/1/11
|
||||
*/
|
||||
public class BoxUtils {
|
||||
|
||||
|
||||
public enum Direction {
|
||||
UP, DOWN, LEFT, RIGHT
|
||||
}
|
||||
|
||||
/**
|
||||
* 寻找指定方向上距离最近的框
|
||||
*
|
||||
* @param anchor 锚点框，作为搜索的起始参考框
|
||||
* @param boxList 候选框列表，在其中搜索目标框
|
||||
* @param direction 搜索方向，可选 UP（上）、DOWN（下）、LEFT（左）、RIGHT（右）
|
||||
* @return 找到的最近的 OcrBox，如果找不到符合条件的框则返回 null
|
||||
*
|
||||
* @apiNote
|
||||
* - 该方法会在指定方向上寻找与锚点框在同一行或同一列的最近邻框
|
||||
* - 水平方向（LEFT/RIGHT）：要求候选框与锚点框在 Y 轴上有重叠（同一行）
|
||||
* - 垂直方向（UP/DOWN）：要求候选框与锚点框在 X 轴上有重叠（同一列）
|
||||
* - 使用欧几里得距离计算中心点之间的距离
|
||||
*/
|
||||
public static OcrBox findNearestBox(OcrBox anchor, List<OcrBox> boxList, Direction direction) {
|
||||
OcrBox nearest = null;
|
||||
double bestScore = Double.MAX_VALUE;
|
||||
AxisSystem axisSystem = buildAxisSystem(anchor);
|
||||
|
||||
for (OcrBox target : boxList) {
|
||||
if (target == anchor) {
|
||||
continue;
|
||||
}
|
||||
|
||||
CandidateMetrics metrics = evaluateCandidate(anchor, target, direction, axisSystem);
|
||||
if (metrics == null) {
|
||||
continue;
|
||||
}
|
||||
if (metrics.score < bestScore) {
|
||||
bestScore = metrics.score;
|
||||
nearest = target;
|
||||
}
|
||||
}
|
||||
|
||||
return nearest;
|
||||
}
|
||||
|
||||
/**
|
||||
* 寻找指定方向上距离最近的多个框（按距离升序返回）
|
||||
*
|
||||
* @param anchor 锚点框
|
||||
* @param boxList 候选框列表
|
||||
* @param direction 搜索方向
|
||||
* @param limit 返回的最大数量（<=0 时返回空列表）
|
||||
* @return 按距离由近到远排序的 OcrBox 列表
|
||||
*/
|
||||
public static List<OcrBox> findNearestBoxes(OcrBox anchor, List<OcrBox> boxList, Direction direction, int limit) {
|
||||
if (anchor == null || boxList == null || boxList.isEmpty() || limit <= 0) {
|
||||
return new ArrayList<>();
|
||||
}
|
||||
|
||||
AxisSystem axisSystem = buildAxisSystem(anchor);
|
||||
List<Neighbor> candidates = new ArrayList<>();
|
||||
|
||||
for (OcrBox target : boxList) {
|
||||
if (target == anchor) {
|
||||
continue;
|
||||
}
|
||||
|
||||
CandidateMetrics metrics = evaluateCandidate(anchor, target, direction, axisSystem);
|
||||
if (metrics == null) {
|
||||
continue;
|
||||
}
|
||||
candidates.add(new Neighbor(target, metrics.score));
|
||||
}
|
||||
|
||||
candidates.sort(Comparator.comparingDouble(n -> n.distance));
|
||||
|
||||
List<OcrBox> result = new ArrayList<>();
|
||||
int size = Math.min(limit, candidates.size());
|
||||
for (int i = 0; i < size; i++) {
|
||||
result.add(candidates.get(i).box);
|
||||
}
|
||||
return result;
|
||||
}
|
||||
|
||||
/**
|
||||
* 内部使用的邻居结构体，存储框和距离
|
||||
*/
|
||||
private static class Neighbor {
|
||||
private final OcrBox box;
|
||||
private final double distance;
|
||||
|
||||
private Neighbor(OcrBox box, double distance) {
|
||||
this.box = box;
|
||||
this.distance = distance;
|
||||
}
|
||||
}
|
||||
|
||||
private static CandidateMetrics evaluateCandidate(OcrBox anchor, OcrBox target, Direction direction, AxisSystem axisSystem) {
|
||||
double mainAxisX = isHorizontalDirection(direction) ? axisSystem.horizontalAxisX : axisSystem.verticalAxisX;
|
||||
double mainAxisY = isHorizontalDirection(direction) ? axisSystem.horizontalAxisY : axisSystem.verticalAxisY;
|
||||
double crossAxisX = isHorizontalDirection(direction) ? axisSystem.verticalAxisX : axisSystem.horizontalAxisX;
|
||||
double crossAxisY = isHorizontalDirection(direction) ? axisSystem.verticalAxisY : axisSystem.horizontalAxisY;
|
||||
|
||||
Projection anchorMain = projectBox(anchor, mainAxisX, mainAxisY);
|
||||
Projection anchorCross = projectBox(anchor, crossAxisX, crossAxisY);
|
||||
Projection targetMain = projectBox(target, mainAxisX, mainAxisY);
|
||||
Projection targetCross = projectBox(target, crossAxisX, crossAxisY);
|
||||
Point anchorCenter = getCenter(anchor);
|
||||
Point targetCenter = getCenter(target);
|
||||
double centerMainDelta = projectPointDelta(anchorCenter, targetCenter, mainAxisX, mainAxisY);
|
||||
|
||||
double mainGap;
|
||||
switch (direction) {
|
||||
case RIGHT:
|
||||
case DOWN:
|
||||
mainGap = targetMain.min - anchorMain.max;
|
||||
if (centerMainDelta <= 0) {
|
||||
return null;
|
||||
}
|
||||
break;
|
||||
case LEFT:
|
||||
case UP:
|
||||
mainGap = anchorMain.min - targetMain.max;
|
||||
if (centerMainDelta >= 0) {
|
||||
return null;
|
||||
}
|
||||
break;
|
||||
default:
|
||||
return null;
|
||||
}
|
||||
|
||||
double anchorMainSize = Math.max(1.0, anchorMain.max - anchorMain.min);
|
||||
double targetMainSize = Math.max(1.0, targetMain.max - targetMain.min);
|
||||
double allowedBacktrack = Math.min(anchorMainSize, targetMainSize) * 0.35;
|
||||
if (mainGap < -allowedBacktrack) {
|
||||
return null;
|
||||
}
|
||||
|
||||
double overlap = Math.max(0.0, Math.min(anchorCross.max, targetCross.max) - Math.max(anchorCross.min, targetCross.min));
|
||||
double minCrossSize = Math.max(1.0, Math.min(anchorCross.max - anchorCross.min, targetCross.max - targetCross.min));
|
||||
double overlapRatio = overlap / minCrossSize;
|
||||
|
||||
double anchorCrossCenter = (anchorCross.min + anchorCross.max) / 2.0;
|
||||
double targetCrossCenter = (targetCross.min + targetCross.max) / 2.0;
|
||||
double crossCenterDistance = Math.abs(anchorCrossCenter - targetCrossCenter);
|
||||
double crossTolerance = Math.max(anchorCross.max - anchorCross.min, targetCross.max - targetCross.min) * 0.6;
|
||||
|
||||
if (overlapRatio < 0.2 && crossCenterDistance > crossTolerance) {
|
||||
return null;
|
||||
}
|
||||
|
||||
double score = Math.max(0.0, mainGap) * 10.0 + crossCenterDistance + Math.abs(centerMainDelta) * 0.01;
|
||||
if (overlapRatio < 0.2) {
|
||||
score += (0.2 - overlapRatio) * 100.0;
|
||||
}
|
||||
return new CandidateMetrics(score);
|
||||
}
|
||||
|
||||
private static boolean isHorizontalDirection(Direction direction) {
|
||||
return direction == Direction.LEFT || direction == Direction.RIGHT;
|
||||
}
|
||||
|
||||
private static Point getCenter(OcrBox box) {
|
||||
float cx = (float) (box.getTopLeft().getX() + box.getTopRight().getX() + box.getBottomRight().getX() + box.getBottomLeft().getX()) / 4;
|
||||
float cy = (float) (box.getTopLeft().getY() + box.getTopRight().getY() + box.getBottomRight().getY() + box.getBottomLeft().getY()) / 4;
|
||||
return new Point(cx, cy);
|
||||
}
|
||||
|
||||
private static AxisSystem buildAxisSystem(OcrBox anchor) {
|
||||
Point topLeft = anchor.getTopLeft();
|
||||
Point topRight = anchor.getTopRight();
|
||||
Point bottomLeft = anchor.getBottomLeft();
|
||||
|
||||
double horizontalX = topRight.getX() - topLeft.getX();
|
||||
double horizontalY = topRight.getY() - topLeft.getY();
|
||||
double verticalX = bottomLeft.getX() - topLeft.getX();
|
||||
double verticalY = bottomLeft.getY() - topLeft.getY();
|
||||
|
||||
double horizontalNorm = Math.hypot(horizontalX, horizontalY);
|
||||
double verticalNorm = Math.hypot(verticalX, verticalY);
|
||||
|
||||
if (horizontalNorm < 1e-6) {
|
||||
horizontalX = 1.0;
|
||||
horizontalY = 0.0;
|
||||
horizontalNorm = 1.0;
|
||||
}
|
||||
if (verticalNorm < 1e-6) {
|
||||
verticalX = 0.0;
|
||||
verticalY = 1.0;
|
||||
verticalNorm = 1.0;
|
||||
}
|
||||
|
||||
return new AxisSystem(
|
||||
horizontalX / horizontalNorm,
|
||||
horizontalY / horizontalNorm,
|
||||
verticalX / verticalNorm,
|
||||
verticalY / verticalNorm
|
||||
);
|
||||
}
|
||||
|
||||
private static Projection projectBox(OcrBox box, double axisX, double axisY) {
|
||||
double[] values = new double[]{
|
||||
dot(box.getTopLeft(), axisX, axisY),
|
||||
dot(box.getTopRight(), axisX, axisY),
|
||||
dot(box.getBottomRight(), axisX, axisY),
|
||||
dot(box.getBottomLeft(), axisX, axisY)
|
||||
};
|
||||
double min = values[0];
|
||||
double max = values[0];
|
||||
for (int i = 1; i < values.length; i++) {
|
||||
min = Math.min(min, values[i]);
|
||||
max = Math.max(max, values[i]);
|
||||
}
|
||||
return new Projection(min, max);
|
||||
}
|
||||
|
||||
private static double dot(Point point, double axisX, double axisY) {
|
||||
return point.getX() * axisX + point.getY() * axisY;
|
||||
}
|
||||
|
||||
private static double projectPointDelta(Point from, Point to, double axisX, double axisY) {
|
||||
return (to.getX() - from.getX()) * axisX + (to.getY() - from.getY()) * axisY;
|
||||
}
|
||||
|
||||
private static double min(double a, double b, double c, double d) {
|
||||
return Math.min(Math.min(a, b), Math.min(c, d));
|
||||
}
|
||||
|
||||
private static double max(double a, double b, double c, double d) {
|
||||
return Math.max(Math.max(a, b), Math.max(c, d));
|
||||
}
|
||||
|
||||
private static class AxisSystem {
|
||||
private final double horizontalAxisX;
|
||||
private final double horizontalAxisY;
|
||||
private final double verticalAxisX;
|
||||
private final double verticalAxisY;
|
||||
|
||||
private AxisSystem(double horizontalAxisX, double horizontalAxisY, double verticalAxisX, double verticalAxisY) {
|
||||
this.horizontalAxisX = horizontalAxisX;
|
||||
this.horizontalAxisY = horizontalAxisY;
|
||||
this.verticalAxisX = verticalAxisX;
|
||||
this.verticalAxisY = verticalAxisY;
|
||||
}
|
||||
}
|
||||
|
||||
private static class Projection {
|
||||
private final double min;
|
||||
private final double max;
|
||||
|
||||
private Projection(double min, double max) {
|
||||
this.min = min;
|
||||
this.max = max;
|
||||
}
|
||||
}
|
||||
|
||||
private static class CandidateMetrics {
|
||||
private final double score;
|
||||
|
||||
private CandidateMetrics(double score) {
|
||||
this.score = score;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
}
|
||||
@@ -8,6 +8,7 @@ import ai.djl.modality.cv.util.NDImageUtils;
|
||||
import ai.djl.ndarray.NDArray;
|
||||
import ai.djl.ndarray.NDList;
|
||||
import ai.djl.ndarray.NDManager;
|
||||
import ai.djl.ndarray.types.Shape;
|
||||
import ai.djl.opencv.OpenCVImageFactory;
|
||||
import cn.smartjavaai.common.cv.SmartImageFactory;
|
||||
import cn.smartjavaai.common.entity.*;
|
||||
@@ -412,4 +413,42 @@ public class OcrUtils {
|
||||
}
|
||||
|
||||
|
||||
public static boolean isSameRow(NDArray box1, NDArray box2) {
|
||||
// 1. 确保 box 是 (4, 2) 的形状
|
||||
NDArray b1 = formatBox(box1);
|
||||
NDArray b2 = formatBox(box2);
|
||||
|
||||
// 2. 获取 Y 坐标列（索引为 1 的列）
|
||||
NDArray y1 = b1.get(":, 1");
|
||||
NDArray y2 = b2.get(":, 1");
|
||||
|
||||
float yMin1 = y1.min().getFloat();
|
||||
float yMax1 = y1.max().getFloat();
|
||||
float yMin2 = y2.min().getFloat();
|
||||
float yMax2 = y2.max().getFloat();
|
||||
|
||||
// 3. 计算重叠高度
|
||||
float overlapHeight = Math.min(yMax1, yMax2) - Math.max(yMin1, yMin2);
|
||||
|
||||
if (overlapHeight <= 0) return false;
|
||||
|
||||
// 4. 计算各自高度
|
||||
float h1 = yMax1 - yMin1;
|
||||
float h2 = yMax2 - yMin2;
|
||||
|
||||
// 5. 判定标准
|
||||
return overlapHeight > (Math.min(h1, h2) * 0.5f);
|
||||
}
|
||||
|
||||
/**
|
||||
* 辅助方法：将 1D 的 8个元素 转换为 2D 的 (4, 2)
|
||||
*/
|
||||
private static NDArray formatBox(NDArray box) {
|
||||
if (box.getShape().dimension() == 1) {
|
||||
// 如果是 [x0, y0, x1, y1...] 这种 8 个元素的 1D 阵，转为 (4, 2)
|
||||
return box.reshape(new Shape(4, 2));
|
||||
}
|
||||
return box;
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
3
pom.xml
3
pom.xml
@@ -7,7 +7,7 @@
|
||||
<name>SmartJavaAI</name>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<packaging>pom</packaging>
|
||||
<description>SmartJavaAI</description>
|
||||
<modules>
|
||||
@@ -19,7 +19,6 @@
|
||||
<module>ocr</module>
|
||||
<module>bom</module>
|
||||
<module>speech</module>
|
||||
<!-- <module>face-pro</module>-->
|
||||
</modules>
|
||||
|
||||
<properties>
|
||||
|
||||
@@ -6,7 +6,7 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>speech</artifactId>
|
||||
@@ -57,7 +57,7 @@
|
||||
</dependencies>
|
||||
|
||||
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>speech</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
@@ -25,7 +25,7 @@
|
||||
</dependency>
|
||||
</dependencies>
|
||||
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>translate</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
@@ -6,11 +6,11 @@
|
||||
<parent>
|
||||
<groupId>cn.smartjavaai</groupId>
|
||||
<artifactId>smartjavaai-parent</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
</parent>
|
||||
|
||||
<artifactId>vision</artifactId>
|
||||
<version>1.1.1</version>
|
||||
<version>1.1.2</version>
|
||||
<name>vision</name>
|
||||
<description>SmartJavaAI</description>
|
||||
<url>https://github.com/geekwenjie/SmartJavaAI</url>
|
||||
|
||||
Reference in New Issue
Block a user