OCR-tess4j,Tesseract-OCR识别图片文字-Java

前提:

1,下载Tesseract-OCR,https://tesseract-ocr.github.io/并安装;

2,下载训练字库tessdata_best,当然也可以不用下载,安装好之后,自带字库,识别成功效果当然也不太好;

所有字库都可以从官方网站下载到,你如果愿意自己做字库训练也是可以的;

下载的字库可以和安装的字库放统一目录,

例如我安装的位置:D:\Program Files\Tesseract-OCR,本身这个目录下有了tessdata字库,下载的tessdata_best字库可以和tessdata平级;

3, 追加环境变量:

Path=D:\Program Files\Tesseract-OCR

TESSDATA_PREFIX=D:\Program Files\Tesseract-OCR\tessdata

打开命令终端，输入：tesseract -v 可以看到版本信息;

4,建立java工程,添加maven依赖,版本号用最新的就好了,还用到了hutool-all工具,自行添加就好

<dependency>
<groupId>net.sourceforge.tess4j</groupId>
<artifactId>tess4j</artifactId><version>4.5.5</version>
</dependency>

5,上代码

package com.boot.test;

import java.awt.Image;
import java.awt.Rectangle;
import java.awt.image.BufferedImage;
import java.io.File;
import java.util.List;

import javax.imageio.ImageIO;

import org.junit.Test;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;

import cn.hutool.core.img.ImgUtil;
import net.sourceforge.tess4j.ITessAPI.TessPageIteratorLevel;
import net.sourceforge.tess4j.ITesseract;
import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.util.ImageHelper;

public class OcrTest {
	
	private static Logger logger = LoggerFactory.getLogger(OcrTest.class);
	
	@Test
	public void testOcr() throws Exception {
		process("D:\\card1.jpg");
		logger.info("-------------------------------------");
		process("D:\\card2.jpg");
		logger.info("-------------------------------------");
		process("D:\\card3.jpg");
	}
	
	public void process(String path) throws Exception {
		double MINIMUM_DESKEW_THRESHOLD = 5.0d;
        File file = new File(path);
        BufferedImage image = ImageIO.read(file);
        
        ImgUtil.read(file);
        
//        ImageDeskew id = new ImageDeskew(image);
//        double imageSkewAngle = id.getSkewAngle(); // determine skew angle
//        logger.info("imageSkewAngle:{}",imageSkewAngle);
//        if ((imageSkewAngle > MINIMUM_DESKEW_THRESHOLD || imageSkewAngle < -(MINIMUM_DESKEW_THRESHOLD))) {
//        	image = ImageHelper.rotateImage(image, -imageSkewAngle); // deskew image
//        }
        
        //灰色图片
        Image gray = ImgUtil.gray(ImgUtil.read(file));
        BufferedImage imageGray = ImageHelper.convertImageToGrayscale(image);
        
        //二值化处理
        BufferedImage image2D =ImageHelper.convertImageToBinary(imageGray);
        Image finalImg = ImgUtil.cut(gray, new Rectangle(5,372,248,26));
        
        ITesseract instance = new Tesseract();
        //获得Tesseract的文字库，设置语言库位置
        instance.setDatapath("D:\\Program Files\\Tesseract-OCR\\tessdata_best");
        instance.setTessVariable("tessedit_char_whitelist", "1234567890");//设置识别范围
        //chi_sim ：简体中文， eng：英文    根据需求选择语言库
//        instance.setLanguage("chi_sim");
//        instance.setOcrEngineMode(TessOcrEngineMode.OEM_TESSERACT_ONLY);
        
        //把要识别的图片按文字行拆成多个区域块
        List<Rectangle> resultList = instance.getSegmentedRegions(imageGray,TessPageIteratorLevel.RIL_TEXTLINE);
        Rectangle rect = null;
        for (int i = 0; i < resultList.size(); i++) {
            rect = resultList.get(i);
//            String result =  instance.doOCR(ImgUtil.toBufferedImage(gray),rect);
//            logger.info(result);
//            logger.info(String.format("Box[%d]: x=%d, y=%d, w=%d, h=%d", i, rect.x, rect.y, rect.width, rect.height));
        }
        
//        ImgUtil.cut(gray, new File("D:\\card1-c.jpg"), rect);
        
        String result = "";
//        long startTime = System.currentTimeMillis();
        //识别灰色图片的最后一个区域块
        result =  instance.doOCR(ImgUtil.toBufferedImage(gray),rect);
        
        //识别二值化图片的最后一个区域块
//        result =  instance.doOCR(image2D,rect);
//        long endTime = System.currentTimeMillis();
//        logger.info("Time is：" + (endTime - startTime) + " 毫秒");
       logger.info("result: \n {}",result);
       
	}

}

原图片: