http://blog.csdn.net/problc/article/details/5797507
Java验证码识别--1
http://blog.csdn.net/problc/article/details/5794460
java验证码识别--2
http://blog.csdn.net/problc/article/details/5797507
java验证码识别--3
http://blog.csdn.net/problc/article/details/5800093
java验证码识别--4
http://blog.csdn.net/problc/article/details/5846614
java验证码识别--5
http://blog.csdn.net/problc/article/details/5983276
(本文仅用于学习研究图像匹配识别原理,不得用于其他用途。)
完整eclipse工程http://download.csdn.net/detail/problc/3829004
github地址 https://github.com/isee15/captcha-ocr
换一个字体固定,大小固定,位置不固定的验证码
还是四步。
1。图像预处理
这验证码还是很厚道的,都没有任何干扰。不用处理
2。分割
先纵向扫描,很容易分成四部分
再对每一部分横向扫描
3。训练就容易了
把分割的结果对应存成5.jpg,9.jpg,3.jpg,a.jpg 就可以了
4。识别
因为固定大小,识别跟 验证码识别--1 里面一样,像素比较就可以了。
识别结果如下,识别率100%:
源码:
- public class ImagePreProcess2 {
- private static Map<BufferedImage, String> trainMap = null;
- private static int index = 0;
- public static int isBlack(int colorInt) {
- Color color = new Color(colorInt);
- if (color.getRed() + color.getGreen() + color.getBlue() <= 100) {
- return 1;
- }
- return 0;
- }
- public static int isWhite(int colorInt) {
- Color color = new Color(colorInt);
- if (color.getRed() + color.getGreen() + color.getBlue() > 100) {
- return 1;
- }
- return 0;
- }
- public static BufferedImage removeBackgroud(String picFile)
- throws Exception {
- BufferedImage img = ImageIO.read(new File(picFile));
- return img;
- }
- public static BufferedImage removeBlank(BufferedImage img) throws Exception {
- int width = img.getWidth();
- int height = img.getHeight();
- int start = 0;
- int end = 0;
- Label1: for (int y = 0; y < height; ++y) {
- int count = 0;
- for (int x = 0; x < width; ++x) {
- if (isWhite(img.getRGB(x, y)) == 1) {
- count++;
- }
- if (count >= 1) {
- start = y;
- break Label1;
- }
- }
- }
- Label2: for (int y = height - 1; y >= 0; --y) {
- int count = 0;
- for (int x = 0; x < width; ++x) {
- if (isWhite(img.getRGB(x, y)) == 1) {
- count++;
- }
- if (count >= 1) {
- end = y;
- break Label2;
- }
- }
- }
- return img.getSubimage(0, start, width, end - start + 1);
- }
- public static List<BufferedImage> splitImage(BufferedImage img)
- throws Exception {
- List<BufferedImage> subImgs = new ArrayList<BufferedImage>();
- int width = img.getWidth();
- int height = img.getHeight();
- List<Integer> weightlist = new ArrayList<Integer>();
- for (int x = 0; x < width; ++x) {
- int count = 0;
- for (int y = 0; y < height; ++y) {
- if (isWhite(img.getRGB(x, y)) == 1) {
- count++;
- }
- }
- weightlist.add(count);
- }
- for (int i = 0; i < weightlist.size();) {
- int length = 0;
- while (weightlist.get(i++) > 1) {
- length++;
- }
- if (length > 12) {
- subImgs.add(removeBlank(img.getSubimage(i - length - 1, 0,
- length / 2, height)));
- subImgs.add(removeBlank(img.getSubimage(i - length / 2 - 1, 0,
- length / 2, height)));
- } else if (length > 3) {
- subImgs.add(removeBlank(img.getSubimage(i - length - 1, 0,
- length, height)));
- }
- }
- return subImgs;
- }
- public static Map<BufferedImage, String> loadTrainData() throws Exception {
- if (trainMap == null) {
- Map<BufferedImage, String> map = new HashMap<BufferedImage, String>();
- File dir = new File("train2");
- File[] files = dir.listFiles();
- for (File file : files) {
- map.put(ImageIO.read(file), file.getName().charAt(0) + "");
- }
- trainMap = map;
- }
- return trainMap;
- }
- public static String getSingleCharOcr(BufferedImage img,
- Map<BufferedImage, String> map) {
- String result = "";
- int width = img.getWidth();
- int height = img.getHeight();
- int min = width * height;
- for (BufferedImage bi : map.keySet()) {
- int count = 0;
- int widthmin = width < bi.getWidth() ? width : bi.getWidth();
- int heightmin = height < bi.getHeight() ? height : bi.getHeight();
- Label1: for (int x = 0; x < widthmin; ++x) {
- for (int y = 0; y < heightmin; ++y) {
- if (isWhite(img.getRGB(x, y)) != isWhite(bi.getRGB(x, y))) {
- count++;
- if (count >= min)
- break Label1;
- }
- }
- }
- if (count < min) {
- min = count;
- result = map.get(bi);
- }
- }
- return result;
- }
- public static String getAllOcr(String file) throws Exception {
- BufferedImage img = removeBackgroud(file);
- List<BufferedImage> listImg = splitImage(img);
- Map<BufferedImage, String> map = loadTrainData();
- String result = "";
- for (BufferedImage bi : listImg) {
- result += getSingleCharOcr(bi, map);
- }
- ImageIO.write(img, "JPG", new File("result2//" + result + ".jpg"));
- return result;
- }
- public static void downloadImage() {
- HttpClient httpClient = new HttpClient();
- GetMethod getMethod = null;
- for (int i = 0; i < 30; i++) {
- getMethod = new GetMethod("http://www.pkland<a href="http://lib.csdn.net/base/dotnet" class='replace_word' title=".NET知识库" target='_blank' style='color:#df3434; font-weight:bold;'>.NET</a>/img.<a href="http://lib.csdn.net/base/php" class='replace_word' title="PHP知识库" target='_blank' style='color:#df3434; font-weight:bold;'>PHP</a>?key="
- + (2000 + i));
- try {
- // 执行getMethod
- int statusCode = httpClient.executeMethod(getMethod);
- if (statusCode != HttpStatus.SC_OK) {
- System.err.println("Method failed: "
- + getMethod.getStatusLine());
- }
- // 读取内容
- String picName = "img2//" + i + ".jpg";
- InputStream inputStream = getMethod.getResponseBodyAsStream();
- OutputStream outStream = new FileOutputStream(picName);
- IOUtils.copy(inputStream, outStream);
- outStream.close();
- System.out.println(i + "OK!");
- } catch (Exception e) {
- e.printStackTrace();
- } finally {
- // 释放连接
- getMethod.releaseConnection();
- }
- }
- }
- public static void trainData() throws Exception {
- File dir = new File("temp");
- File[] files = dir.listFiles();
- for (File file : files) {
- BufferedImage img = removeBackgroud("temp//" + file.getName());
- List<BufferedImage> listImg = splitImage(img);
- if (listImg.size() == 4) {
- for (int j = 0; j < listImg.size(); ++j) {
- ImageIO.write(listImg.get(j), "JPG", new File("train2//"
- + file.getName().charAt(j) + "-" + (index++)
- + ".jpg"));
- }
- }
- }
- }
- /**
- * @param args
- * @throws Exception
- */
- public static void main(String[] args) throws Exception {
- // downloadImage();
- for (int i = 0; i < 30; ++i) {
- String text = getAllOcr("img2//" + i + ".jpg");
- System.out.println(i + ".jpg = " + text);
- }
- }
- }