使用 Java 语言，通过Tesseract-OCR对图片进行识别。

1.Tesseract-OCR

下载windows版本并安装。

2.程序如下：

a.Image io Helper类

2. package OCR;
4. import java.awt.image.BufferedImage;
5. import java.io. File ;
6. import java.io.IOException;
7. import java.util.Iterator;
8. import java.util.Locale;
10. import javax.imageio.IIOImage;
11. import javax.imageio.ImageIO;
12. import javax.imageio.Image reader ;
13. import javax.imageio.ImageWriteParam;
14. import javax.imageio.ImageWriter;
15. import javax.imageio.metadata.IIOMetadata;
16. import javax.imageio.stream.ImageInputStream;
17. import javax.imageio.stream.ImageOutputStream;
19. import com.sun.media.imageio.plugins. tiff .TIFFImageWriteParam;
21. public class ImageIOHelper {
22. /**
23. * 图片文件转换为tif格式
24. * @param imageFile 文件路径
25. * @param imageFormat 文件扩展名
26. * @return
27. */
28. public static File createImage(File imageFile, String imageFormat) {
29. File tempFile = null;
30. try {
31. Iterator<ImageReader> readers = ImageIO.getImageReadersByFormatName(imageFormat);
32. ImageReader reader = readers.next();
34. ImageInputStream iis = ImageIO.createImageInputStream(imageFile);
35. reader.setInput(iis);
36. //Read the stream metadata
37. IIOMetadata streamMetadata = reader.getStreamMetadata();
39. //Set up the writeParam
40. TIFFImageWriteParam tiffWriteParam = new TIFFImageWriteParam(Locale.CHINESE);
41. tiffWriteParam.setCompressionMode(ImageWriteParam.MODE_DISABLED);
43. //Get tif writer and set output to file
44. Iterator<ImageWriter> writers = ImageIO.getImageWritersByFormatName(“tiff”);
45. ImageWriter writer = writers.next();
47. BufferedImage bi = reader.read(0);
48. IIOImage image = new IIOImage(bi,null,reader.getImageMetadata(0));
49. tempFile = tempImageFile(imageFile);
50. ImageOutputStream ios = ImageIO.createImageOutputStream(tempFile);
51. writer.setOutput(ios);
52. writer.write(streamMetadata, image, tiffWriteParam);
53. ios. close ();
55. writer.dispose();
56. reader.dispose();
58. } catch (IOException e) {
59. e.printStackTrace();
60. }
61. return tempFile;
62. }
64. private static File tempImageFile(File imageFile) {
65. String path = imageFile.getPath();
66. StringBuffer strB = new StringBuffer(path);
67. strB.insert(path.lastIndexOf(‘.’),0);
68. return new File(strB.toString().replaceFirst(“(?<=//.)(//w+)$”, “tif”));
69. }
71. }

b. OCR 核心类

2. package OCR;
4. import java.io.BufferedReader;
5. import java.io.File;
6. import java.io.FileInputStream;
7. import java.io.InputStreamReader;
8. import java.util.ArrayList;
9. import java.util.List;
11. import org.jdesktop.swingx.util.OS;
13. public class OCR {
14. private final String LANG_OPTION = “-l”; //英文字母小写l，并非数字1
15. private final String EOL = System.getProperty(“line.separator”);
16. private String tessPath = “C://Program Files//Tesseract-OCR”;
17. //private String tessPath = new File(“tesseract”).getAbsolutePath();
19. public String recognizeText(File imageFile,String imageFormat)throws Exception{
20. File tempImage = ImageIOHelper.createImage(imageFile,imageFormat);
21. File outputFile = new File(imageFile.getParentFile(),”output”);
22. StringBuffer strB = new StringBuffer();
23. List<String> cmd = new ArrayList<String>();
24. if(OS.isWindowsXP()){
25. cmd.add(tessPath+”//tesseract”);
26. }else if(OS.isLinux()){
27. cmd.add(“tesseract”);
28. }else{
29. cmd.add(tessPath+”//tesseract”);
30. }
31. cmd.add(“”);
32. cmd.add(outputFile.getName());
33. //cmd.add(LANG_OPTION);
34. //cmd.add(“chi_sim”);
35. //cmd.add(“eng”);
37. ProcessBuilder pb = new ProcessBuilder();
38. pb.directory(imageFile.getParentFile());
40. cmd.set(1, tempImage.getName());
41. pb.command(cmd);
42. pb.redirectErrorStream(true);
44. Process process = pb.start();
45. //tesseract.exe 1.jpg 1 -l chi_sim
46. int w = process.waitFor();
48. //删除临时正在工作文件
49. tempImage.delete();
51. if(w==0){
52. BufferedReader in = new BufferedReader(new InputStreamReader(new FileInputStream(outputFile.getAbsolutePath()+”.txt”),”UTF-8″));
54. String str;
55. while((str = in.readLine())!=null){
56. strB. append (str).append(EOL);
57. }
58. in.close();
59. }else{
60. String msg ;
61. switch(w){
62. case 1:
63. msg = “Errors accessing files.There may be spaces in your image’s filename.”;
64. break ;
65. case 29:
66. msg = “Cannot recongnize the image or its selected region.”;
67. break;
68. case 31:
69. msg = “Unsupported image format.”;
70. break;
71. default:
72. msg = “Errors occurred.”;
73. }
74. tempImage.delete();
75. //throw new RuntimeException(msg);
76. }
77. new File(outputFile.getAbsolutePath()+”.txt”).delete();
78. return strB.toString();
79. }
80. }

c.main

2. package OCR;
3. import java.io.File;
4. import java.io.IOException;
6. public class TestOcr {
8. /**
9. * @param args
10. */
11. public static void main(String[] args) {
12. //输入图片地址
13. String path = “d://test//test.bmp”;
14. try {
15. String valCode = new OCR().recognizeText(new File(path), “bmp”);
16. System.out.println(valCode);
17. } catch (IOException e) {
18. e.printStackTrace();
19. } catch (Exception e) {
20. e.printStackTrace();
21. }
22. }
24. }