Java之Tesseract實現OCR

1、實現邏輯

package com.vue.demo.service.serviceimpl;

import com.vue.demo.service.OCRService;
import net.sourceforge.tess4j.Tesseract;
import net.sourceforge.tess4j.TesseractException;
import net.sourceforge.tess4j.util.ImageHelper;
import org.apache.commons.io.FileUtils;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.stereotype.Service;
import org.springframework.web.multipart.MultipartFile;

import javax.imageio.ImageIO;
import java.awt.image.BufferedImage;
import java.io.File;
import java.io.IOException;

/**
 * @author yangwj
 * @date 2020/4/1 9:29
 */
@Service
public class OCRServiceImpl implements OCRService {
    private  static  final Logger ocrServiceImplLog = LoggerFactory.getLogger(OCRServiceImpl.class);
    String language = "";

    /**
     * 方法一
     * @param file
     * @return
     */
    @Override
    public String getCharacterFromPic(MultipartFile file) {
//        String modelPath = "D:\\software\\ocr-tesseract\\tessdata";
        String modelPath = "/root/project/java/tesseract_model";

        Tesseract tessreact = new Tesseract();
        //需要指定訓練集 訓練集到 https://github.com/tesseract-ocr/tessdata 下載。
        tessreact.setDatapath(modelPath);

        if(language.equals("ch")) {
            //注意  默認是英文識別，如果做中文識別，需要單獨設置。
            tessreact.setLanguage("chi_sim");
        }
        try {
            File imageFile = new File(file.getOriginalFilename());
            FileUtils.copyInputStreamToFile(file.getInputStream(), imageFile);
            String result = tessreact.doOCR(imageFile);
            ocrServiceImplLog.info(result);
            System.out.println("----------------");
            String handleResult  =  this.ocr(imageFile,modelPath);
            ocrServiceImplLog.info(handleResult);
            return result+"----------------------------------\n\r"+handleResult;
        } catch (TesseractException e) {
            System.err.println(e.getMessage());
        } catch (IOException e) {
            e.printStackTrace();
        }
        return null;
    }

    @Override
    public String getLanguage(String language) {
        if(language == null || language == "" ) {
            return null;
        }
        this.language = language;
        return "success";
    }

    /**
     * 方法二
     * @param file
     * @param modelPath
     * @return
     */
    private  String ocr(File file,String modelPath) {
        String result = null;
        try {
            double start = System.currentTimeMillis();
            BufferedImage textImage = ImageIO.read(file);
            // 這里對圖片黑白處理,增強識別率.這里先通過截圖,截取圖片中需要識別的部分
            textImage = ImageHelper.convertImageToGrayscale(textImage);
            // 圖片銳化
            textImage = ImageHelper.convertImageToBinary(textImage);
            // 圖片放大倍數,增強識別率(很多圖片本身無法識別,放大5倍時就可以輕易識,但是考濾到客戶電腦配置低,針式打印機打印不連貫的問題,這里就放大5倍)
            textImage = ImageHelper.getScaledInstance(textImage, textImage.getWidth() * 1, textImage.getHeight() * 1);

            textImage = ImageHelper.convertImageToBinary(textImage);
            String saveImgPath = "/root/project/java/tesseract_model/temp_img";
//            String saveImgPath = "D:\\software\\ocr-tesseract\\img_tem\\temp.img";
            ImageIO.write(textImage, "png", new File(saveImgPath));

            Tesseract instance = new Tesseract();
            //設置訓練庫的位置
//            String modelPath = "/root/project/java/tesseract_model";


            instance.setDatapath(modelPath);
            //中文識別
            instance.setLanguage("chi_sim");
            result = instance.doOCR(textImage);
            double end = System.currentTimeMillis();
            System.out.println("耗時" + (end - start) / 1000 + " s");
        } catch (Exception e) {
            e.printStackTrace();
        }
        return result;
    }

}

2、部署到centos,遇到的問題，可以看這篇

https://www.cnblogs.com/ywjfx/p/12629697.html