1概述
针对当前互联网流行的滑块验证码，这里做了一个工具，适用于前后端分离的项目。工具将抠图和抠图后的原图以字节码的形式输出。
2工具源码
package com.liutao.util;
import javax.imageio.ImageIO;
import javax.imageio.ImageReadParam;
import javax.imageio.ImageReader;
import javax.imageio.stream.ImageInputStream;
import java.awt.*;
import java.awt.image.BufferedImage;
import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.FileInputStream;
import java.io.InputStream;
import java.text.NumberFormat;
import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;
import java.util.Random;
/**
* 滑块验证工具类
*
* @author: LIUTAO
* @Description:
* @Date: Created in 10:57 2018/6/25
* @Modified By:
*/
public class VerifyImageUtil {
private static int ORI_WIDTH = 590; //源文件宽度
private static int ORI_HEIGHT = 360; //源文件高度
private static int X; //抠图坐标x
private static int Y; //抠图坐标y
private static int WIDTH; //模板图宽度
private static int HEIGHT; //模板图高度
private static float xPercent; //X位置移动百分比
private static float yPercent; //Y位置移动百分比
public static int getX() {
return X;
}
public static int getY() {
return Y;
}
public static float getxPercent() {
return xPercent;
}
public static float getyPercent() {
return yPercent;
}
/**
* 根据模板切图
* newImage：抠图直字节码
* oriCopyImage：原图字节码
*
* @param templateFile
* @param targetFile
* @param templateType
* @param targetType
* @return
* @throws Exception
*/
public static Map<String, byte[]> pictureTemplatesCut(File templateFile, File targetFile, String templateType, String targetType) throws Exception {
Map<String, byte[]> pictureMap = new HashMap<>();
// 文件类型
String templateFiletype = templateType;
String oriFiletype = targetType;
if (StringUtils.isEmpty(templateFiletype) || StringUtils.isEmpty(oriFiletype)) {
throw new RuntimeException("file type is empty");
}
// 源文件流
File Orifile = targetFile;
InputStream oriis = new FileInputStream(Orifile);
// 模板图
BufferedImage imageTemplate = ImageIO.read(templateFile);
WIDTH = imageTemplate.getWidth();
HEIGHT = imageTemplate.getHeight();
generateCutoutCoordinates();
// 最终图像
BufferedImage newImage = new BufferedImage(WIDTH, HEIGHT, imageTemplate.getType());
Graphics2D graphics = newImage.createGraphics();
graphics.setBackground(Color.white);
int bold = 5;
// 获取感兴趣的目标区域
BufferedImage targetImageNoDeal = getTargetArea(X, Y, WIDTH, HEIGHT, oriis, oriFiletype);
// 根据模板图片抠图
newImage = DealCutPictureByTemplate(targetImageNoDeal, imageTemplate, newImage);
// 设置“抗锯齿”的属性
graphics.setRenderingHint(RenderingHints.KEY_ANTIALIASING, RenderingHints.VALUE_ANTIALIAS_ON);
graphics.setStroke(new BasicStroke(bold, BasicStroke.CAP_BUTT, BasicStroke.JOIN_BEVEL));
graphics.drawImage(newImage, 0, 0, null);
graphics.dispose();
ByteArrayOutputStream os = new ByteArrayOutputStream();//新建流。
ImageIO.write(newImage, "png", os);//利用ImageIO类提供的write方法，将bi以png图片的数据模式写入流。
byte[] newImages = os.toByteArray();
pictureMap.put("newImage", newImages);
// 源图生成遮罩
BufferedImage oriImage = ImageIO.read(Orifile);
byte[] oriCopyImages = DealOriPictureByTemplate(oriImage, imageTemplate, X, Y);
pictureMap.put("oriCopyImage", oriCopyImages);
return pictureMap;
}
/**
* 抠图后原图生成
*
* @param oriImage
* @param templateImage
* @param x
* @param y
* @return
* @throws Exception
*/
private static byte[] DealOriPictureByTemplate(BufferedImage oriImage, BufferedImage templateImage, int x,
int y) throws Exception {
// 源文件备份图像矩阵 支持alpha通道的rgb图像
BufferedImage ori_copy_image = new BufferedImage(oriImage.getWidth(), oriImage.getHeight(), BufferedImage.TYPE_4BYTE_ABGR);
// 源文件图像矩阵
int[][] oriImageData = getData(oriImage);
// 模板图像矩阵
int[][] templateImageData = getData(templateImage);
//copy 源图做不透明处理
for (int i = 0; i < oriImageData.length; i++) {
for (int j = 0; j < oriImageData[0].length; j++) {
int rgb = oriImage.getRGB(i, j);
int r = (0xff & rgb);
int g = (0xff & (rgb >> 8));
int b = (0xff & (rgb >> 16));
//无透明处理
rgb = r + (g << 8) + (b << 16) + (255 << 24);
ori_copy_image.setRGB(i, j, rgb);
}
}
for (int i = 0; i < templateImageData.length; i++) {
for (int j = 0; j < templateImageData[0].length - 5; j++) {
int rgb = templateImage.getRGB(i, j);
//对源文件备份图像(x+i,y+j)坐标点进行透明处理
if (rgb != 16777215 && rgb <= 0) {
int rgb_ori = ori_copy_image.getRGB(x + i, y + j);
int r = (0xff & rgb_ori);
int g = (0xff & (rgb_ori >> 8));
int b = (0xff & (rgb_ori >> 16));
rgb_ori = r + (g << 8) + (b << 16) + (150 << 24);
ori_copy_image.setRGB(x + i, y + j, rgb_ori);
} else {
//do nothing
}
}
}
ByteArrayOutputStream os = new ByteArrayOutputStream();//新建流。
ImageIO.write(ori_copy_image, "png", os);//利用ImageIO类提供的write方法，将bi以png图片的数据模式写入流。
byte b[] = os.toByteArray();//从流中获取数据数组。
return b;
}
/**
* 根据模板图片抠图
*
* @param oriImage
* @param templateImage
* @return
*/
private static BufferedImage DealCutPictureByTemplate(BufferedImage oriImage, BufferedImage templateImage,
BufferedImage targetImage) throws Exception {
// 源文件图像矩阵
int[][] oriImageData = getData(oriImage);
// 模板图像矩阵
int[][] templateImageData = getData(templateImage);
// 模板图像宽度
for (int i = 0; i < templateImageData.length; i++) {
// 模板图片高度
for (int j = 0; j < templateImageData[0].length; j++) {
// 如果模板图像当前像素点不是白色 copy源文件信息到目标图片中
int rgb = templateImageData[i][j];
if (rgb != 16777215 && rgb <= 0) {
targetImage.setRGB(i, j, oriImageData[i][j]);
}
}
}
return targetImage;
}
/**
* 获取目标区域
*
* @param x 随机切图坐标x轴位置
* @param y 随机切图坐标y轴位置
* @param targetWidth 切图后目标宽度
* @param targetHeight 切图后目标高度
* @param ois 源文件输入流
* @return
* @throws Exception
*/
private static BufferedImage getTargetArea(int x, int y, int targetWidth, int targetHeight, InputStream ois,
String filetype) throws Exception {
Iterator<ImageReader> imageReaderList = ImageIO.getImageReadersByFormatName(filetype);
ImageReader imageReader = imageReaderList.next();
// 获取图片流
ImageInputStream iis = ImageIO.createImageInputStream(ois);
// 输入源中的图像将只按顺序读取
imageReader.setInput(iis, true);
ImageReadParam param = imageReader.getDefaultReadParam();
Rectangle rec = new Rectangle(x, y, targetWidth, targetHeight);
param.setSourceRegion(rec);
BufferedImage targetImage = imageReader.read(0, param);
return targetImage;
}
/**
* 生成图像矩阵
*
* @param
* @return
* @throws Exception
*/
private static int[][] getData(BufferedImage bimg) throws Exception {
int[][] data = new int[bimg.getWidth()][bimg.getHeight()];
for (int i = 0; i < bimg.getWidth(); i++) {
for (int j = 0; j < bimg.getHeight(); j++) {
data[i][j] = bimg.getRGB(i, j);
}
}
return data;
}
/**
* 随机生成抠图坐标
*/
private static void generateCutoutCoordinates() {
Random random = new Random();
int widthDifference = ORI_WIDTH - WIDTH;
int heightDifference = ORI_HEIGHT - HEIGHT;
if (widthDifference <= 0) {
X = 5;
} else {
X = random.nextInt(ORI_WIDTH - WIDTH) + 5;
}
if (heightDifference <= 0) {
Y = 5;
} else {
Y = random.nextInt(ORI_HEIGHT - HEIGHT) + 5;
}
NumberFormat numberFormat = NumberFormat.getInstance();
numberFormat.setMaximumFractionDigits(2);
xPercent = Float.parseFloat(numberFormat.format((float) X / (float) ORI_WIDTH));
yPercent = Float.parseFloat(numberFormat.format((float) Y / (float) ORI_HEIGHT));
}
}3测试代码
public void testImg() throws Exception {
Map<String, byte[]> pictureMap;
File templateFile; //模板图片
File targetFile; //
Random random = new Random();
int templateNo = random.nextInt(4) + 1;
int targetNo = random.nextInt(20) + 1;
InputStream stream = getClass().getClassLoader().getResourceAsStream("static/templates/" + templateNo + ".png");
templateFile = new File(templateNo + ".png");
FileUtils.copyInputStreamToFile(stream, templateFile);
stream = getClass().getClassLoader().getResourceAsStream("static/targets/" + targetNo + ".jpg");
targetFile = new File(targetNo + ".jpg");
FileUtils.copyInputStreamToFile(stream, targetFile);
pictureMap = VerifyImageUtil.pictureTemplatesCut(templateFile, targetFile, "png", "jpg");
byte[] oriCopyImages = pictureMap.get("oriCopyImage");
byte[] newImages = pictureMap.get("newImage");
FileOutputStream fout = new FileOutputStream("C:/Users/Administrator/Desktop/oriCopyImage.png");
//将字节写入文件
try {
fout.write(oriCopyImages);
} catch (IOException e) {
e.printStackTrace();
}
fout.close();
FileOutputStream newImageFout = new FileOutputStream("C:/Users/Administrator/Desktop/newImage.png");
//将字节写入文件
newImageFout.write(newImages);
newImageFout.close();
}本文章为转载内容，我们尊重原作者对文章享有的著作权。如有内容错误或侵权问题，欢迎原作者联系我们进行内容更正或删除文章。
