12306验证码识别

   

12306验证码识别

           

猪圈 发布于 2013年01月18日 5时,
            64评/19465阅

           

   
   

分享到： 

收藏 +165

踩
顶
4

干扰线去除判断比较挫是导致识别率低的原因，希望高手指点吧~后面的特征提取和训练识别就交给大家了~ （很不幸地告诉大家，上班前12306的验证码干扰又加强了，主要还是干扰线部分）

实现原理很简单：

1.图像灰度化与
二值化

2.去除干扰线（二值化在一定程度上已经消弱部分）



在5点以前，12306二值化后的图像干扰线是离散的，所以很容易地使用纵向扫描就能擦除干扰点提取字模，代码如下：

		for (int x = 0; x < w; ++x) {
			for (int y = 0; y < h; ++y) {
				if (isBlack(gray[x][y])) {
					if (x > 0 && x < (w - 1) && isWhite(gray[x - 1][y]) && isWhite(gray[x + 1][y])) {
						gray[x][y] = 65535;
					}
				}
			}
		}
		
		for (int x = 0; x < w; ++x) {
			for (int y = 0; y < h; ++y) {
				if (isBlack(gray[x][y])) {
					if (y > 0 && y < (h - 1) && isWhite(gray[x][y - 1]) && isWhite(gray[x][y + 1])) {
						gray[x][y] = 65535;
					}
				}
				binaryBufferedImage.setRGB(x, y, gray[x][y]);
			}
		}

注：目前该方法已经不奏效，已经删除~（16:18更新调整亮度提高识别率）

   

    标签：
    <无>
   

   

   

       

       

       

代码片段(1)
[全屏查看所有代码]

                       

                       

1. [代码][Java]代码    


跳至
                    [1]
   
[全屏预览]

           

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

package org.chinasb.client; import java.awt.Color; import java.awt.image.BufferedImage; import java.io.File; import java.io.IOException; import javax.imageio.ImageIO; public class BinaryTest {     public static void main(String[] args) throws IOException {         BufferedImage bufferedImage = ImageIO.read(new File("D:/passCodeAction.jpg"));         int h = bufferedImage.getHeight();         int w = bufferedImage.getWidth();                   // 灰度化         int[][] gray = new int[w][h];         for (int x = 0; x < w; x++) {             for (int y = 0; y < h; y++) {                 int argb = bufferedImage.getRGB(x, y);                 // 图像加亮（调整亮度识别率非常高）                 int r = (int)(((argb >> 16) & 0xFF) * 1.1 + 30);                 int g = (int)(((argb >> 8) & 0xFF) * 1.1 + 30);                 int b = (int)(((argb >> 0) & 0xFF) * 1.1 + 30);                 if (r >= 255) {                     r = 255;                 }                 if (g >= 255) {                     g = 255;                 }                 if (b >= 255) {                     b = 255;                 }                 gray[x][y] = (int) Math.pow((Math.pow(r, 2.2) * 0.2973 + Math.pow(g, 2.2) * 0.6274 + Math.pow(b, 2.2) * 0.0753), 1/2.2);             }         }         // 二值化         int threshold = ostu(gray, w, h);         BufferedImage binaryBufferedImage = new BufferedImage(w, h, BufferedImage.TYPE_BYTE_BINARY);         for (int x = 0; x < w; x++) {             for (int y = 0; y < h; y++) {                 if (gray[x][y] > threshold) {                     gray[x][y] |= 0x00FFFF;                 } else {                     gray[x][y] &= 0xFF0000;                 }                 binaryBufferedImage.setRGB(x, y, gray[x][y]);             }         }         // 矩阵打印         for (int y = 0; y < h; y++) {             for (int x = 0; x < w; x++) {                 if (isBlack(binaryBufferedImage.getRGB(x, y))) {                     System.out.print("*");                 } else {                     System.out.print(" ");                 }             }             System.out.println();         }         ImageIO.write(binaryBufferedImage, "jpg", new File("D:/code.jpg"));     }     public static boolean isBlack(int colorInt) {         Color color = new Color(colorInt);         if (color.getRed() + color.getGreen() + color.getBlue() <= 300) {             return true;         }         return false;     }     public static boolean isWhite(int colorInt) {         Color color = new Color(colorInt);         if (color.getRed() + color.getGreen() + color.getBlue() > 300) {             return true;         }         return false;     }     public static int isBlackOrWhite(int colorInt) {         if (getColorBright(colorInt) < 30 || getColorBright(colorInt) > 730) {             return 1;         }         return 0;     }     public static int getColorBright(int colorInt) {         Color color = new Color(colorInt);         return color.getRed() + color.getGreen() + color.getBlue();     }     public static int ostu(int[][] gray, int w, int h) {         int[] histData = new int[w * h];         // Calculate histogram         for (int x = 0; x < w; x++) {             for (int y = 0; y < h; y++) {                 int red = 0xFF & gray[x][y];                 histData[red]++;             }         }         // Total number of pixels         int total = w * h;                   float sum = 0;         for (int t = 0; t < 256; t++)             sum += t * histData[t];         float sumB = 0;         int wB = 0;         int wF = 0;         float varMax = 0;         int threshold = 0;         for (int t = 0; t < 256; t++) {             wB += histData[t]; // Weight Background             if (wB == 0)                 continue;             wF = total - wB; // Weight Foreground             if (wF == 0)                 break;             sumB += (float) (t * histData[t]);             float mB = sumB / wB; // Mean Background             float mF = (sum - sumB) / wF; // Mean Foreground             // Calculate Between Class Variance             float varBetween = (float) wB * (float) wF * (mB - mF) * (mB - mF);             // Check if new maximum found             if (varBetween > varMax) {                 varMax = varBetween;                 threshold = t;             }         }         return threshold;     } }