java识别图片上的文字

布满荆棘的人生 2022-03-08 14:56 301阅读 0赞

一、需要用到[Tesseract-OCR][]所以先下载，并安装

二、配置环境变量，在系统环境变量中，Path添加C:\\Program Files (x86)\\Tesseract-OCR；新建变量TESSDATA\_PREFIX，变量值为C:\\Program Files (x86)\\Tesseract-OCR\\tessdata

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70][]

![20190315142613547.png][]

三、开始识别

（1）例如C:\\Users\\dell\\Desktop\\test\\test.jpg的图片

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 1][]

（2）测试识别效果

识别图片命令

tesseract 图片名称 生成的结果文件的名称 字库

cmd命令行进入图片所在文件夹位置，输入命令（其中chi\_sim是中文的字库），result是在当前文件夹下生产一个result.txt文件并写入识别的内容

tesseract test.jpg result -l chi_sim

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 2][]

（3）打开result.txt文件，识别的不是很准确，可以对字库进行训练

![20190315144527748.png][]

四、训练字库

（1）需要下载[jTessBoxEditor][]工具，jTessBoxEditor是java写的，所以需要安装jdk

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 3][]

（2）修改图片名字，不能随意命名，有固定的格式，这里我把图片名字改成cs.roy.exp0.jpg

[lang].[fontname].exp[num]

（3）利用windows自带的画图工具另存为.tif文件

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 4][]

![20190315152248593.png][]

（4）生成.box文件，cmd输入命令

tesseract cs.roy.exp0.tif cs.roy.exp0 -l chi_sim batch.nochop makebox

（5）打开jTessBoxEditor的train.bat，选择tif文件，可以点击merge按钮进行合并文字，如果没有中文可以在Settings->Font里设置宋体

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 5][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 6][]

（6）优化后点保存

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 7][]

（7）输入命令

tesseract  cs.roy.exp0.jpg cs.roy.exp0  nobatch box.train

（8）输入命令

unicharset_extractor cs.roy.exp0.box

（9）新建font\_properties文件，内容写入roy 0 0 0 0 0

注：roy对应的是命名规则的\[fontname\],font\_properties文件不是font\_properties.txt

![20190315152758337.png][]

（10）输入命令

shapeclustering -F font_properties -U unicharset cs.roy.exp0.tr

（11）输入命令

mftraining -F font_properties -U unicharset -O unicharset cs.roy.exp0.tr

（12）输入命令

cntraining cs.roy.exp0.tr

(13)把新生成的五个文件加上前缀roy.

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 8][]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 9][]

(14)合并相关文件，输入命令，生成.traineddata文件

combine_tessdata roy.

![20190315160258901.png][]

（15）把roy.traineddata文件复制到C:\\Program Files (x86)\\Tesseract-OCR\\tessdata文件夹下

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 10][]

（16）输入命令进行识别测试，-l 后面输入对应的字库

tesseract cs.roy.exp0.jpg cs.roy.exp0 -l roy

![20190315160953845.png][]

五、java代码调用（其中getImg()方法是把图片二值化处理增加识别效果）

package springboot;
    
    import javax.imageio.ImageIO;
    import java.awt.*;
    import java.awt.image.BufferedImage;
    import java.io.*;
    import java.util.ArrayList;
    import java.util.List;
    
    public class OCRHelper {
    	private final String LANG_OPTION = "-l";
    	private final String EOL = System.getProperty("line.separator");
    
    	public static void main(String[] args) {
    		try {
    			//需要被识别的图片
    			File file = new File("C:\\Users\\dell\\Desktop\\test\\cs.roy.exp0.jpg");
    			String recognizeText = new OCRHelper().recognizeText(file);
    			System.out.print(recognizeText + "\t");
    		} catch (IOException e) {
    			e.printStackTrace();
    		} catch (Exception e) {
    			e.printStackTrace();
    		}
    	}
    
    	private static void getImg(File read,File write) throws IOException {
    		BufferedImage image = ImageIO.read(read);
    		int w = image.getWidth();
    		int h = image.getHeight();
    		float[] rgb = new float[3];
    		double[][] zuobiao = new double[w][h];
    		int R = 0;
    		float red = 0;
    		float green = 0;
    		float blue = 0;
    		BufferedImage bi= new BufferedImage(w, h, BufferedImage.TYPE_BYTE_BINARY);
    		;
    		for (int x = 0; x < w; x++) {
    			for (int y = 0; y < h; y++) {
    				int pixel = image.getRGB(x, y);
    				rgb[0] = (pixel & 0xff0000) >> 16;
    				rgb[1] = (pixel & 0xff00) >> 8;
    				rgb[2] = (pixel & 0xff);
    				red += rgb[0];
    				green += rgb[1];
    				blue += rgb[2];
    				R = (x+1) *(y+1);
    				float avg = (rgb[0]+rgb[1]+rgb[2])/3;
    				zuobiao[x][y] = avg;
    
    			}
    		}
    		double SW = 170;
    		for (int x = 0; x < w; x++) {
    			for (int y = 0; y < h; y++) {
    				if (zuobiao[x][y] <= SW) {
    					int max = new Color(0, 0, 0).getRGB();
    					bi.setRGB(x, y, max);
    				}else{
    					int min = new Color(255, 255, 255).getRGB();
    					bi.setRGB(x, y, min);
    				}
    			}
    		}
    
    		ImageIO.write(bi, "jpg", write);
    	}
    
    	/**
    	 * Tesseract-OCR的安装路径
    	 */
    	private String tessPath = "C://Program Files (x86)//Tesseract-OCR";
    	//private String tessPath = new File("tesseract").getAbsolutePath();
    
    	/**
    	 * @param imageFile 传入的图像文件
    	 * @return 识别后的字符串
    	 */
    	public String recognizeText(File imageFile) throws Exception {
    		/**
    		 * 设置输出文件的保存的文件目录
    		 */
    		File outputFile = new File(imageFile.getParentFile(), "output");
    
    		StringBuffer strB = new StringBuffer();
    		List<String> cmd = new ArrayList<String>();
    
    		cmd.add(tessPath + "\\tesseract");
    
    		cmd.add("");
    		cmd.add(outputFile.getName());
    		cmd.add(LANG_OPTION);
    //		cmd.add("chi_sim");
    		cmd.add("roy");
    
    		ProcessBuilder pb = new ProcessBuilder();
    		pb.directory(imageFile.getParentFile());
    		cmd.set(1, imageFile.getName());
    		pb.command(cmd);
    		pb.redirectErrorStream(true);
    		long startTime = System.currentTimeMillis();
    		System.out.println("开始时间：" + startTime);
    		Process process = pb.start();
    		int w = process.waitFor();
    		if (w == 0) {// 0代表正常退出
    			BufferedReader in = new BufferedReader(new InputStreamReader(new FileInputStream(outputFile.getAbsolutePath() + ".txt"), "UTF-8"));
    			String str;
    			while ((str = in.readLine()) != null) {
    				strB.append(str).append(EOL);
    			}
    			in.close();
    			long endTime = System.currentTimeMillis();
    			System.out.println("结束时间：" + endTime);
    			System.out.println("耗时：" + (endTime - startTime) + "毫秒");
    		} else {
    			String msg;
    			switch (w) {
    				case 1:
    					msg = "Errors accessing files. There may be spaces in your image's filename.";
    					break;
    				case 29:
    					msg = "Cannot recognize the image or its selected region.";
    					break;
    				case 31:
    					msg = "Unsupported image format.";
    					break;
    				default:
    					msg = "Errors occurred.";
    			}
    			throw new RuntimeException(msg);
    		}
    		new File(outputFile.getAbsolutePath() + ".txt").delete();
    		return strB.toString().replaceAll("\\s*", "");
    	}
    }

[Tesseract-OCR]: https://download.csdn.net/download/qq_15260315/11032139
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70]: /images/20220308/f271b0909e3a4b6f899f19b1158d1f9a.png
[20190315142613547.png]: /images/20220308/3a1338676a2c4115a1a59cfa47ea6585.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 1]: /images/20220308/a5aa87f16ef74a71b8d8734b58d31499.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 2]: /images/20220308/b830dfe3998049dd84595efe6ff2ffbd.png
[20190315144527748.png]: /images/20220308/9788467ad6a449ea8ad464c9dd9a06e9.png
[jTessBoxEditor]: https://sourceforge.net/projects/vietocr/files/jTessBoxEditor/
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 3]: /images/20220308/ce45eef59c174e0ea3e4ad19778a00ae.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 4]: /images/20220308/bab1ade56b6a4739839cfa9c9dc1db79.png
[20190315152248593.png]: /images/20220308/1c0dca4fbbed4717bd049d8d188ebd31.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 5]: /images/20220308/18659776893f460988db2781d0c68f52.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 6]: /images/20220308/2edc2b7dc34c49808d5af6f7eb999e2a.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 7]: /images/20220308/88b62082142b460e865da01dab056528.png
[20190315152758337.png]: /images/20220308/c195e812b3704cfea851d04ad216c555.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 8]: /images/20220308/f9f84c42c9ad4102b13241f5997254ff.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 9]: /images/20220308/47ad23476a5143adab88e5ed548df784.png
[20190315160258901.png]: /images/20220308/150ec2de336c498d9c4b620aec8fc481.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzE1MjYwMzE1_size_16_color_FFFFFF_t_70 10]: /images/20220308/077475105e82467dad5ea8c4111d19b6.png
[20190315160953845.png]: /images/20220308/0a21276154964bc9b8e429b9da67f1c8.png