package com.jackie.crawler.doubanmovie.ocr;
import java.awt.Color;
import java.awt.image.BufferedImage;
import java.io.File;
import java.net.URL;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import javax.imageio.ImageIO;
import net.sourceforge.tess4j.Tesseract;
public class ImageUtil {
// public int getPiexPostion(int x, int y){
//
// }
public static void main(String[] args){
try {
boolean load = true;
load = false;
// BufferedImage image = ImageIO.read(new URL("http://www.miitbeian.gov.cn/captcha.jpg")) ;
// if(load){
// ImageIO.write(image, "jpg", new File("E:/captcha.jpg") );
// }else{
// image = ImageIO.read(new File("D:\\爬虫测试\\yzm\\111.png")) ;
// }
BufferedImage image = ImageIO.read(new File("D:\\爬虫测试\\yzm\\11.jpg")) ;
// image = ImageUtil.grayFilter(image);
image = ImageUtil.binaryFilter(image);
image = ImageUtil.lineFilter(image);
// image = ImageUtil.lineFilter(image);
// image = ImageUtil.line2Filter(image);
// image = ImageUtil.point2Filter(image);
// image = ImageUtil.lineFilter(image);
image = ImageUtil.meanFilter(image);
// image = ImageUtil.lineFilter(image);
// image = ImageUtil.binaryFilter(image);
File imageFile = new File("E:/captcha5.jpg");
// imageFile = new File("E:/test/test.jpg");
ImageIO.write(image, "jpg", imageFile);
Tesseract tesseract = Tesseract.getInstance();
tesseract.setLanguage("eng");
String code = tesseract.doOCR(imageFile);
System.out.println(code);
} catch (Exception e) {
e.printStackTrace();
}
}
/**
* 获取RGB的灰度
*/
public static int getGray(int rgb){
//or 直接new个color对象
Color cc = new Color(rgb);
int r = cc.getRed();
int g = cc.getGreen();
int b = cc.getBlue();
int gray = (int)(0.299 * (double)r + 0.587 * (double)g + 0.114 * (double)b);
return gray;
}
/**
* 处理图片二值化
* @param image
* @return
*/
public static BufferedImage binaryFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
BufferedImage outImage = new BufferedImage(width, height, image.getType());
double redSum = 0, greenSum = 0, blueSum = 0;
double total = height * width;
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
redSum += color.getRed();
greenSum += color.getGreen();
blueSum += color.getBlue();
}
}
int means = (int)(redSum / total);
for (int x = minx; x < width; x++) {
int alpha = 0, red = 0, green = 0, blue = 0;
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
alpha = color.getAlpha();
if(color.getRed() >= 20){
red = green = blue =255;
}else{
red = green = blue = 0;
}
color = new Color(red, green, blue);
outImage.setRGB(x, y, color.getRGB());
}
}
return outImage;
}
/**
* 图片灰度处理
* @param image
* @return
*/
public static BufferedImage grayFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
int alpha = color.getAlpha();
int gray = (int)(0.299 * (double)color.getRed() + 0.587 * (double)color.getGreen() + 0.114 * (double)color.getBlue());
color = new Color(gray, gray, gray, alpha);
outImage.setRGB(x, y, color.getRGB());
}
}
return outImage;
}
public static BufferedImage medianFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
int[] order = new int[9];
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
//处理西北位置像素
if(x-1 >= width && y-1 >= height){
order[0] = new Color(image.getRGB(x-1, y-1)).getRed();
}
//处理正北位置像素
if(x-1 >= width && y-1 >= height){
order[1] = new Color(image.getRGB(x-1, y)).getRed();
}
//处理东北位置像素
if(x-1 >= width && y+1 <= height){
order[2] = new Color(image.getRGB(x-1, y+1)).getRed();
}
//处理正东位置像素
if(x+1 <= width-1){
order[3] = new Color(image.getRGB(x+1, y)).getRed();
}
//处理东南位置像素
if(x+1 <= width-1 && y+1 <= height-1){
order[4] = new Color(image.getRGB(x+1, y+1)).getRed();
}
//处理正南位置像素
if(y+1 <= height-1){
order[5] = new Color(image.getRGB(x, y+1)).getRed();
}
//处理西南位置像素
if(x-1 >= width && y+1 <= height-1){
order[6] = new Color(image.getRGB(x-1, y+1)).getRed();
}
//处理正西位置像素
if(x-1 >= width){
order[7] = new Color(image.getRGB(x-1, y)).getRed();
}
//处理自己位置像素
order[8] = new Color(image.getRGB(x, y)).getRed();
Arrays.sort(order);
int rgb = order[order.length/2];
color = new Color(rgb, rgb, rgb, color.getAlpha());
//
outImage.setRGB(x, y, color.getRGB());
}
}
return outImage;
}
/**
* 均值滤波
* @param image
* @return
*/
public static BufferedImage meanFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
int meanSum = 0;
int count = 0;
//处理西北位置像素
if(x-1 >= width && y-1 >= height){
meanSum += new Color(image.getRGB(x-1, y-1)).getRed();
count++;
}
//处理正北位置像素
if(x-1 >= width && y-1 >= height){
meanSum += new Color(image.getRGB(x-1, y)).getRed();
count++;
}
//处理东北位置像素
if(x-1 >= width && y+1 <= height){
meanSum += new Color(image.getRGB(x-1, y+1)).getRed();
count++;
}
//处理正东位置像素
if(x+1 <= width-1){
meanSum += new Color(image.getRGB(x+1, y)).getRed();
count++;
}
//处理东南位置像素
if(x+1 <= width-1 && y+1 <= height-1){
meanSum += new Color(image.getRGB(x+1, y+1)).getRed();
count++;
}
//处理正南位置像素
if(y+1 <= height-1){
meanSum += new Color(image.getRGB(x, y+1)).getRed();
count++;
}
//处理西南位置像素
if(x-1 >= width && y+1 <= height-1){
meanSum += new Color(image.getRGB(x-1, y+1)).getRed();
}
//处理正西位置像素
if(x-1 >= width){
meanSum += new Color(image.getRGB(x-1, y)).getRed();
count++;
}
//处理自己位置像素
// meanSum += new Color(image.getRGB(x, y)).getRed();
if(count != 0){
int rgb = meanSum / count;
color = new Color(rgb, rgb, rgb, color.getAlpha());
}
//
outImage.setRGB(x, y, color.getRGB());
}
}
return outImage;
}
/**
* 去除杂点、杂线
* @param image
* @param length
* @return
*/
public static BufferedImage point2Filter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
Color black = new Color(0, 0, 0);
Color white = new Color(255, 255, 255);
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
if(color.getRed() == white.getRed()){
outImage.setRGB(x, y, color.getRGB());
continue;
}
//先判断左右
int self = color.getRed();
int left_1 = ImageUtil.getColor(image, x-1, y).getRed();
int left_2 = ImageUtil.getColor(image, x-1, y+1).getRed();
int right_1 = ImageUtil.getColor(image, x+1, y).getRed();
int right_2 = ImageUtil.getColor(image, x+1, y+1).getRed();
int up_1 = ImageUtil.getColor(image, x, y-1).getRed();
int up_2 = ImageUtil.getColor(image, x+1, y-1).getRed();
int down_1 = ImageUtil.getColor(image, x, y+1).getRed();
int down_2 = ImageUtil.getColor(image, x+1, y+1).getRed();
int left = left_1 == 0 && left_2 == 0 ? black.getRed() : white.getRed();
int right = right_1 == 0 && right_2 == 0 ? black.getRed() : white.getRed();
int up = up_1 == 0 && up_2 == 0 ? black.getRed() : white.getRed();
int down = down_1 == 0 && down_2 == 0 ? black.getRed() : white.getRed();
// if(left == right && left == 0 && up == down && up != 0 && self == 0){
// //左右为黑 上下为白
// setRGB(outImage, x, y, white.getRGB());//自己
// setRGB(outImage, x-1, y, white.getRGB());//左
// setRGB(outImage, x-1, y+1, white.getRGB());//左
// setRGB(outImage, x+1, y, white.getRGB());//右
// setRGB(outImage, x+1, y+1, white.getRGB());//右
// }else if(left == right && left != 0 && up == down && up == 0 && self == 0){
// //上下为黑 左右为白
// setRGB(outImage, x, y, white.getRGB());//自己
// setRGB(outImage, x, y-1, white.getRGB());//上
// setRGB(outImage, x+1, y-1, white.getRGB());//上
// setRGB(outImage, x, y+1, white.getRGB());//下
// setRGB(outImage, x+1, y+1, white.getRGB());//下
// }else if(left == right && up == down && left == up && left != 0 && self == 0){
// //上下左右均为白色
// setRGB(outImage, x, y, white.getRGB());//自己
// setRGB(outImage, x, y-1, white.getRGB());//上
// setRGB(outImage, x+1, y-1, white.getRGB());//上
// setRGB(outImage, x, y+1, white.getRGB());//下
// setRGB(outImage, x+1, y+1, white.getRGB());//下
// setRGB(outImage, x-1, y, white.getRGB());//左
// setRGB(outImage, x-1, y+1, white.getRGB());//左
// setRGB(outImage, x+1, y, white.getRGB());//右
// setRGB(outImage, x+1, y+1, white.getRGB());//右
// }else{
int count = 0;
count = left != 0 ? count+1 : count;
count = right != 0 ? count+1 : count;
count = up != 0 ? count+1 : count;
count = down != 0 ? count+1 : count;
if(count > 2){
setRGB(outImage, x, y, white.getRGB());//自己
}else{
setRGB(outImage, x, y, black.getRGB());
}
// }
}
}
return outImage;
}
/**
* 去除杂点
* @param image
* @param sideLength
* @return
*/
public static BufferedImage pointFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
Color black = new Color(0, 0, 0);
Color white = new Color(255, 255, 255);
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
if(color.getRed() == white.getRed()){
outImage.setRGB(x, y, color.getRGB());
continue;
}
//先判断左右
int self = color.getRed();
int left = ImageUtil.getColor(image, x-1, y).getRed();
int right = ImageUtil.getColor(image, x+1, y).getRed();
int up = ImageUtil.getColor(image, x, y-1).getRed();
int down = ImageUtil.getColor(image, x, y+1).getRed();
if(left == right && right == up && up == down && down != 0 && self == 0){
setRGB(outImage, x, y, white.getRGB());
}
}
}
return outImage;
}
/**
* 去除杂点、杂线
* @param image
* @param length
* @return
*/
public static BufferedImage line2Filter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
Color black = new Color(0, 0, 0);
Color white = new Color(255, 255, 255);
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
if(color.getRed() == white.getRed()){
outImage.setRGB(x, y, color.getRGB());
continue;
}
//先判断左右
int self_1 = color.getRed();
int self_2 = ImageUtil.getColor(image, x+1, y).getRed();
int self_3 = ImageUtil.getColor(image, x, y+1).getRed();
int self_4 = ImageUtil.getColor(image, x+1, y+1).getRed();
if(self_1 == self_2 && self_2 == self_3 && self_3 == self_4 && self_1 != 0){
continue;
}
int self = 0;
int left_1 = ImageUtil.getColor(image, x-1, y).getRed();
int left_2 = ImageUtil.getColor(image, x-1, y+1).getRed();
int right_1 = ImageUtil.getColor(image, x+2, y).getRed();
int right_2 = ImageUtil.getColor(image, x+2, y+1).getRed();
int up_1 = ImageUtil.getColor(image, x, y-1).getRed();
int up_2 = ImageUtil.getColor(image, x+1, y-1).getRed();
int down_1 = ImageUtil.getColor(image, x, y+2).getRed();
int down_2 = ImageUtil.getColor(image, x+1, y+2).getRed();
int left = left_1 == 0 && left_2 == 0 ? black.getRed() : white.getRed();
int right = right_1 == 0 && right_2 == 0 ? black.getRed() : white.getRed();
int up = up_1 == 0 && up_2 == 0 ? black.getRed() : white.getRed();
int down = down_1 == 0 && down_2 == 0 ? black.getRed() : white.getRed();
if(left == right && left == 0 && up == down && up != 0 && self == 0){
//左右为黑 上下为白
setRGB(outImage, x, y, white.getRGB());//自己
}else if(left == right && left != 0 && up == down && up == 0 && self == 0){
//上下为黑 左右为白
setRGB(outImage, x, y, white.getRGB());//自己
}
else if(left == right && up == down && left == up && left != 0 && self == 0){
// //上下左右均为白色
setRGB(outImage, x, y, white.getRGB());//自己
}
else{
int count = 0;
count = left != 0 ? count+1 : count;
count = right != 0 ? count+1 : count;
count = up != 0 ? count+1 : count;
count = down != 0 ? count+1 : count;
//
if(count > 2){
setRGB(outImage, x, y, white.getRGB());//自己
}else{
setRGB(outImage, x, y, black.getRGB());
}
}
}
}
return outImage;
}
/**
* 去除杂点、杂线
* @param image
* @param length
* @return
*/
public static BufferedImage lineFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
int minx = image.getMinX();
int miny = image.getMinY();
Color black = new Color(0, 0, 0);
Color white = new Color(255, 255, 255);
BufferedImage outImage = new BufferedImage(width, height, image.getType());
for (int x = minx; x < width; x++) {
for (int y = miny; y < height; y++) {
Color color = new Color(image.getRGB(x, y));
if(color.getRed() == white.getRed()){
outImage.setRGB(x, y, color.getRGB());
continue;
}
//先判断左右
int self = color.getRed();
int left = ImageUtil.getColor(image, x-1, y).getRed();
int right = ImageUtil.getColor(image, x+1, y).getRed();
int up = ImageUtil.getColor(image, x, y-1).getRed();
int down = ImageUtil.getColor(image, x, y+1).getRed();
if(left == right && left == 0 && up == down && up != 0 && self == 0){
//左右为黑 上下为白
setRGB(outImage, x, y, white.getRGB());
}else if(left == right && left != 0 && up == down && up == 0 && self == 0){
setRGB(outImage, x, y, white.getRGB());
}else if(left == right && up == down && left == up && left != 0 && self == 0){
setRGB(outImage, x, y, white.getRGB());
}else{
int count = 0;
count = left != 0 ? count+1 : count;
count = right != 0 ? count+1 : count;
count = up != 0 ? count+1 : count;
count = down != 0 ? count+1 : count;
//
if(count > 2){
setRGB(outImage, x, y, white.getRGB());
}else{
// setRGB(outImage, x, y, black.getRGB());
}
}
}
}
return outImage;
}
/**
* 根据坐标获取像素点的Color对象，如果像素位置位于图片外，则默认为白色
* @param image
* @param x
* @param y
* @return
*/
public static Color getColor(BufferedImage image, int x, int y){
if(x < image.getMinX() || x > image.getWidth() - 1){
return new Color(255, 255, 255);
}
if(y < image.getMinY() || y > image.getHeight() - 1){
return new Color(255, 255, 255);
}
return new Color(image.getRGB(x, y));
}
/**
* 使用近似一维Means方法寻找二值化阈值进行二值化
* @param image
* @return
*/
public static BufferedImage thresholdBinaryFilter(BufferedImage image){
int width = image.getWidth();
int height = image.getHeight();
BufferedImage outImage = new BufferedImage(width, height, image.getType());
int[] inPixels = new int[width * height];
int[] outPixels = new int[width * height];
inPixels = getPixels(image, 0, 0, width, height);
int index = 0;
int means = getThreshold(inPixels, height, width);
for (int x = 0; x < width; x++) {
int ta = 0, tr = 0, tg = 0, tb = 0;
for (int y = 0; y < height; y++) {
int rgb = image.getRGB(x, y);
Color color = new Color(rgb);
ta = color.getAlpha();
tr = color.getRed();
tg = color.getGreen();
tb = color.getBlue();
if (tr > means) {
tr = tg = tb = 255; // white
} else {
tr = tg = tb = 0; // black
}
outPixels[index] = (ta << 24) | (tr << 16) | (tg << 8) | tb;
outImage.setRGB(x, y, outPixels[index]);
}
}
return outImage;
}
//获取阈值
private static int getThreshold(int[] inPixels, int height, int width) {
// maybe this value can reduce the calculation consume;
int inithreshold = 127;
int finalthreshold = 0;
int temp[] = new int[inPixels.length];
for(int index=0; index<inPixels.length; index++) {
temp[index] = (inPixels[index] >> 16) & 0xff;
}
List<Integer> sub1 = new ArrayList<Integer>();
List<Integer> sub2 = new ArrayList<Integer>();
int means1 = 0, means2 = 0;
while(finalthreshold != inithreshold) {
finalthreshold = inithreshold;
for(int i=0; i<temp.length; i++) {
if(temp[i] <= inithreshold) {
sub1.add(temp[i]);
} else {
sub2.add(temp[i]);
}
}
means1 = getMeans(sub1);
means2 = getMeans(sub2);
sub1.clear();
sub2.clear();
inithreshold = (means1 + means2) / 2;
}
return finalthreshold;
}
private static int getMeans(List<Integer> data) {
if(data.size() == 0){
return 0;
}
int result = 0;
int size = data.size();
for (Integer i : data) {
result += i;
}
return (result / size);
}
/**
* 获取像素
* @param image
* @param minX
* @param minY
* @param width
* @param height
* @return
*/
public static int[] getPixels(BufferedImage image, int minX, int minY, int width, int height){
int[] pixels = new int[(width-minX) * (height - minY)];
int i = 0;
for (int x = minX; x < width; x++) {
for (int y = minY; y < height; y++) {
int rgb = image.getRGB(x, y);
pixels[i] = rgb;
i++;
}
}
return pixels;
}
public static void setRGB(BufferedImage image, int x, int y, int rgb){
if(x < image.getMinX() || x > image.getWidth() - 1){
return;
}
if(y < image.getMinY() || y > image.getHeight() - 1){
return;
}
image.setRGB(x, y, rgb);
}
private static Map<String, Integer> getRoundPiexls(BufferedImage image, int x, int y, int sideLength){
Map<String, Integer> piexlMap = new HashMap<String, Integer>();
if(sideLength == 2){
getRoundPiexls(image, x, y, piexlMap);
getRoundPiexls(image, x + 1, y, piexlMap);
getRoundPiexls(image, x + 1, y + 1, piexlMap);
getRoundPiexls(image, x, y + 1, piexlMap);
}else{
getRoundPiexls(image, x, y, piexlMap);
}
// int[] piexls = new int[piexlMap.size()];
//
// int i = 0;
// for (Map.Entry<String, Integer> entry: piexlMap.entrySet()) {
// piexls[i] = entry.getValue();
// i++;
// }
return piexlMap;
}
private static void getRoundPiexls(BufferedImage image, int x, int y, Map<String, Integer> piexlMap){
int tempX = x;
int tempY = y;
piexlMap.put("xy"+x+","+y, getColor(image, x, y).getRGB());
tempX = x;
tempY = y - 1;
piexlMap.put(tempX+","+tempY, getColor(image, tempX, tempY).getRGB());
tempX = x + 1;
tempY = y;
piexlMap.put(tempX+","+tempY, getColor(image, tempX, tempY).getRGB());
tempX = x;
tempY = y + 1;
piexlMap.put(tempX+","+tempY, getColor(image, tempX, tempY).getRGB());
tempX = x - 1;
tempY = y;
piexlMap.put(tempX+","+tempY, getColor(image, tempX, tempY).getRGB());
}
}
