p86743915
/
pp6hrvtlm
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

no commit message

Browse Source
devA
yuxue 5 years ago
parent c5d6723405
commit 8b49a033d9
1 changed files with 321 additions and 44 deletions
Show Stats Download Patch File Download Diff File

365
src/main/java/com/yuxue/util/ImageUtil.java
Unescape View File

@ -9,6 +9,8 @@ import org.bytedeco.javacpp.BytePointer;
import org.bytedeco.javacpp.opencv_core;
import org.bytedeco.javacpp.opencv_core.Mat;
import org.bytedeco.javacpp.opencv_core.MatVector;
import org.bytedeco.javacpp.opencv_core.Point2d;
import org.bytedeco.javacpp.opencv_core.Point2f;
import org.bytedeco.javacpp.opencv_core.RotatedRect;
import org.bytedeco.javacpp.opencv_core.Scalar;
import org.bytedeco.javacpp.opencv_core.Size;
@ -22,6 +24,7 @@ import com.google.common.collect.Maps;
/**
* 车牌图片处理工具类
* 开发测试中...
* @author yuxue
* @date 2020-05-18 12:07
*/
@ -48,7 +51,6 @@ public class ImageUtil {
// 车牌定位处理步骤，该map用于表示步骤图片的顺序
private static Map<String, Integer> debugMap = Maps.newLinkedHashMap();
static {
// debugMap.put("result", 99);
debugMap.put("yuantu", 0); // 高斯模糊
debugMap.put("gaussianBlur", 1); // 高斯模糊
debugMap.put("gray", 2); // 图像灰度化
@ -61,6 +63,7 @@ public class ImageUtil {
debugMap.put("crop", 9); // 切图
debugMap.put("resize", 10); // 切图resize
debugMap.put("char_threshold", 11); //
// debugMap.put("char_clearLiuDing", 10); // 去除柳钉
// debugMap.put("specMat", 11);
// debugMap.put("chineseMat", 12);
@ -76,6 +79,7 @@ public class ImageUtil {
String tempPath = DEFAULT_BASE_TEST_PATH + "test/";
String filename = tempPath + "/100_yuantu.jpg";
// filename = tempPath + "/100_yuantu1.jpg";
Mat src = opencv_imgcodecs.imread(filename);
@ -86,6 +90,8 @@ public class ImageUtil {
Mat grey = ImageUtil.grey(gsMat, debug, tempPath);
Mat sobel = ImageUtil.sobel(grey, debug, tempPath);
// Mat sobel = ImageUtil.scharr(grey, debug, tempPath);
Mat threshold = ImageUtil.threshold(sobel, debug, tempPath);
@ -146,7 +152,6 @@ public class ImageUtil {
*/
public static final int SOBEL_SCALE = 1;
public static final int SOBEL_DELTA = 0;
public static final int SOBEL_DDEPTH = opencv_core.CV_16S;
public static final int SOBEL_X_WEIGHT = 1;
public static final int SOBEL_Y_WEIGHT = 0;
public static Mat sobel(Mat inMat, Boolean debug, String tempPath) {
@ -158,23 +163,65 @@ public class ImageUtil {
Mat abs_grad_x = new Mat();
Mat abs_grad_y = new Mat();
opencv_imgproc.Sobel(inMat, grad_x, SOBEL_DDEPTH, 1, 0, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT);
opencv_imgproc.Sobel(inMat, grad_x, opencv_core.CV_16S, 1, 0, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT);
opencv_core.convertScaleAbs(grad_x, abs_grad_x);
opencv_imgproc.Sobel(inMat, grad_y, SOBEL_DDEPTH, 0, 1, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT);
opencv_imgproc.Sobel(inMat, grad_y, opencv_core.CV_16S, 0, 1, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT);
opencv_core.convertScaleAbs(grad_y, abs_grad_y);
grad_x.release();
grad_y.release();
opencv_core.addWeighted(abs_grad_x, SOBEL_X_WEIGHT, abs_grad_y, SOBEL_Y_WEIGHT, 0, dst);
abs_grad_x.release();
abs_grad_y.release();
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst);
}
return dst;
}
/**
* 对图像进行scharr 运算，得到图像的一阶水平方向导数
* @param inMat
* @param debug
* @param tempPath
* @return
*/
public static Mat scharr(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat();
Mat grad_x = new Mat();
Mat grad_y = new Mat();
Mat abs_grad_x = new Mat();
Mat abs_grad_y = new Mat();
//注意求梯度的时候我们使用的是Scharr算法，sofia算法容易收到图像细节的干扰
//所谓梯度运算就是对图像中的像素点进行就导数运算，从而得到相邻两个像素点的差异值 by:Tantuo
opencv_imgproc.Scharr(inMat, grad_x, opencv_core.CV_32F, 1, 0);
opencv_imgproc.Scharr(inMat, grad_y, opencv_core.CV_32F, 0, 1);
//openCV中有32位浮点数的CvType用于保存可能是负值的像素数据值
opencv_core.convertScaleAbs(grad_x, abs_grad_x);
opencv_core.convertScaleAbs(grad_y, abs_grad_y);
//openCV中使用release()释放Mat类图像，使用recycle()释放BitMap类图像
grad_x.release();
grad_y.release();
opencv_core.addWeighted(abs_grad_x, 0.5, abs_grad_y, 0.5, 0, dst);
abs_grad_x.release();
abs_grad_y.release();
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst);
}
return dst;
}
/**
* 对图像进行二值化。将灰度图像（每个像素点有256 个取值可能）转化为二值图像（每个像素点仅有1 和0 两个取值可能）
* 对图像进行二值化。将灰度图像（每个像素点有256 个取值可能）
* 转化为二值图像（每个像素点仅有1 和0 两个取值可能）
* @param inMat
* @param debug
* @param tempPath
@ -183,6 +230,15 @@ public class ImageUtil {
public static Mat threshold(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat();
opencv_imgproc.threshold(inMat, dst, 0, 255, opencv_imgproc.CV_THRESH_OTSU + opencv_imgproc.CV_THRESH_BINARY);
/*for (int i = 0; i < dst.rows(); i++) {
for (int j = 0; j < dst.cols(); j++) {
if(dst.ptr(i, j).getInt() !=0 ) {
System.err.println(i + "\t" + j + "\t" +dst.ptr(i, j).getInt());
}
}
}*/
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("threshold") + 100) + "_threshold.jpg", dst);
}
@ -197,7 +253,9 @@ public class ImageUtil {
* @param tempPath
* @return
*/
public static final int DEFAULT_MORPH_SIZE_WIDTH = 17;
//public static final int DEFAULT_MORPH_SIZE_WIDTH = 15;
// public static final int DEFAULT_MORPH_SIZE_HEIGHT = 3;
public static final int DEFAULT_MORPH_SIZE_WIDTH = 9;
public static final int DEFAULT_MORPH_SIZE_HEIGHT = 3;
public static Mat morphology(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat();
@ -207,8 +265,18 @@ public class ImageUtil {
opencv_imgproc.morphologyEx(inMat, dst, opencv_imgproc.MORPH_CLOSE, element);
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology.jpg", dst);
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology0.jpg", dst);
}
// 去除小连通区域
removeSmallRegion(dst, dst, 100, 1, 1, debug, tempPath);
// 去除孔洞
removeSmallRegion(dst, dst, 100, 0, 0, debug, tempPath);
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology1.jpg", dst);
}
return dst;
}
@ -225,8 +293,17 @@ public class ImageUtil {
public static MatVector contours(Mat src, Mat inMat, Boolean debug, String tempPath) {
MatVector contours = new MatVector();
// 提取外部轮廓
// CV_RETR_EXTERNAL只检测最外围轮廓，
// CV_RETR_LIST 检测所有的轮廓
// CV_CHAIN_APPROX_NONE 保存物体边界上所有连续的轮廓点到contours向量内
opencv_imgproc.findContours(inMat, contours, opencv_imgproc.CV_RETR_EXTERNAL, opencv_imgproc.CV_CHAIN_APPROX_NONE);
// 在小连接处分割轮廓
/*MatVector retContour = new MatVector();
for (int i = 0; i < contours.size(); i++) {
retContour.put(contours.get(i));
}*/
if (debug) {
Mat result = new Mat();
src.copyTo(result); // 复制一张图，不在原图上进行操作，防止后续需要使用原图
@ -235,6 +312,7 @@ public class ImageUtil {
// 输出带轮廓的原图
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("contours") + 100) + "_contours.jpg", result);
}
// return retContour;
return contours;
}
@ -247,10 +325,6 @@ public class ImageUtil {
* @param tempPath
* @return
*/
final static float DEFAULT_ERROR = 0.6f;
final static float DEFAULT_ASPECT = 3.75f;
public static final int DEFAULT_VERIFY_MIN = 3;
public static final int DEFAULT_VERIFY_MAX = 20;
public static final int DEFAULT_ANGLE = 30; // 角度判断所用常量
public static final int WIDTH = 136;
public static final int HEIGHT = 36;
@ -259,78 +333,89 @@ public class ImageUtil {
public static Vector<Mat> screenBlock(Mat src, MatVector contours, Boolean debug, String tempPath){
Vector<Mat> dst = new Vector<Mat>();
MatVector mv = new MatVector();
for (int i = 0; i < contours.size(); ++i) {
MatVector mv = new MatVector(); // 用于在原图上描绘筛选后的结果
for (int i = 0, j = 0; i < contours.size(); i++) {
// RotatedRect 该类表示平面上的旋转矩形，有三个属性： 矩形中心点(质心); 边长(长和宽); 旋转角度
// boundingRect()得到包覆此轮廓的最小正矩形， minAreaRect()得到包覆轮廓的最小斜矩形
RotatedRect mr = opencv_imgproc.minAreaRect(contours.get(i));
float angle = Math.abs(mr.angle());
if (verifySizes(mr) && angle <= DEFAULT_ANGLE) { // 判断尺寸及旋转角度 ±30°，排除不合法的图块
if (debug) { // 描绘出筛选后的轮廓
mv.put(contours.get(i));
Mat result = new Mat();
src.copyTo(result); // 复制一张图，不在原图上进行操作，防止后续需要使用原图
// 将轮廓描绘到原图
opencv_imgproc.drawContours(result, mv, -1, new Scalar(0, 0, 255, 255));
// 输出带轮廓的原图
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("screenblock") + 100) + "_screenblock.jpg", result);
}
// 旋转角度，根据需要是否进行角度旋转
if (checkPlateSize(mr) && angle <= DEFAULT_ANGLE) { // 判断尺寸及旋转角度 ±30°，排除不合法的图块
mv.put(contours.get(i));
Size rect_size = new Size((int) mr.size().width(), (int) mr.size().height());
if (mr.size().width() / mr.size().height() < 1) { // 宽度小于高度
angle = 90 + angle; // 旋转90°
rect_size = new Size(rect_size.height(), rect_size.width());
}
Mat rotmat = opencv_imgproc.getRotationMatrix2D(mr.center(), angle, 1);
// 旋转角度，根据需要是否进行角度旋转
Mat img_rotated = new Mat();
opencv_imgproc.warpAffine(src, img_rotated, rotmat, src.size()); // CV_INTER_CUBIC
Mat rotmat = opencv_imgproc.getRotationMatrix2D(mr.center(), angle, 1); // 旋转
opencv_imgproc.warpAffine(src, img_rotated, rotmat, src.size()); // 仿射变换
// 切图
Mat img_crop = new Mat();
opencv_imgproc.getRectSubPix(src, rect_size, mr.center(), img_crop);
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("crop") + 100) + "_crop_" + i + ".jpg", img_crop);
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("crop") + 100) + "_crop_" + j + ".png", img_crop);
}
// 处理切图，调整为指定大小
Mat resized = new Mat(HEIGHT, WIDTH, TYPE);
opencv_imgproc.resize(img_crop, resized, resized.size(), 0, 0, opencv_imgproc.INTER_CUBIC);
if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("resize") + 100) + "_resize_" + i + ".jpg", resized);
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("resize") + 100) + "_resize_" + j + ".png", resized);
j++;
}
dst.add(resized);
}
}
if (debug) {
Mat result = new Mat();
src.copyTo(result); // 复制一张图，不在原图上进行操作，防止后续需要使用原图
// 将轮廓描绘到原图
opencv_imgproc.drawContours(result, mv, -1, new Scalar(0, 0, 255, 255));
// 输出带轮廓的原图
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("screenblock") + 100) + "_screenblock.jpg", result);
}
return dst;
}
/**
* 对minAreaRect获得的最小外接矩形，用纵横比进行判断
* 对minAreaRect获得的最小外接矩形
* 判断面积以及宽高比是否在制定的范围内
* 黄牌、蓝牌
* @param mr
* @return
*/
private static boolean verifySizes(RotatedRect mr) {
// China car plate size: 440mm*140mm，aspect 3.142857
final static float DEFAULT_ERROR = 0.6f;
final static float DEFAULT_ASPECT = 3.75f;
public static final int DEFAULT_VERIFY_MIN = 3;
public static final int DEFAULT_VERIFY_MAX = 20;
/*final static float DEFAULT_ERROR = 0.9f;
final static float DEFAULT_ASPECT = 4f;
public static final int DEFAULT_VERIFY_MIN = 1;
public static final int DEFAULT_VERIFY_MAX = 30;*/
private static boolean checkPlateSize(RotatedRect mr) {
// 国内车牌大小: 440mm*140mm，宽高比 3.142857
// 切图面积取值范围
int min = 44 * 14 * DEFAULT_VERIFY_MIN;
int max = 44 * 14 * DEFAULT_VERIFY_MAX;
// Get only patchs that match to a respect ratio.
// 切图横纵比取值范围
float rmin = DEFAULT_ASPECT - DEFAULT_ASPECT * DEFAULT_ERROR;
float rmax = DEFAULT_ASPECT + DEFAULT_ASPECT * DEFAULT_ERROR;
// 计算面积
// 切图计算面积
int area = (int) (mr.size().height() * mr.size().width());
// 计算纵横比
// 切图宽高比
float r = mr.size().width() / mr.size().height();
if (r < 1) {
r = mr.size().height() / mr.size().width();
}
return min <= area && area <= max && rmin <= r && r <= rmax;
}
@ -341,6 +426,8 @@ public class ImageUtil {
/**
* rgb图像转换为hsv图像
* @param inMat
@ -412,6 +499,196 @@ public class ImageUtil {
return;
}
/**
* 计算最大内接矩形
* https://blog.csdn.net/cfqcfqcfqcfqcfq/article/details/53084090
* @param inMat
* @return
*/
public static RotatedRect maxAreaRect(Mat threshold, Point2f point2f) {
int edge[] = new int[4];
edge[0] = (int) point2f.x() + 1;//top
edge[1] = (int) point2f.x() + 1;//right
edge[2] = (int) point2f.y() - 1;//bottom
edge[3] = (int) point2f.x() - 1;//left
boolean[] expand = { true, true, true, true};//扩展标记位
int n = 0;
while (expand[0] || expand[1] || expand[2] || expand[3]){
int edgeID = n % 4;
expand[edgeID] = expandEdge(threshold, edge, edgeID);
n++;
}
//[3]
//qDebug() << edge[0] << edge[1] << edge[2] << edge[3];
/*Point tl = Point(edge[3], edge[0]);
Point br = Point(edge[1], edge[2]);
return new Rect(tl, br);*/
return null;
}
/**
* @brief expandEdge 扩展边界函数
* @param img:输入图像，单通道二值图，深度为8
* @param edge 边界数组，存放4条边界值
* @param edgeID 当前边界号
* @return 布尔值 确定当前边界是否可以扩展
*/
public static boolean expandEdge(Mat img, int edge[], int edgeID) {
int nc = img.cols();
int nr = img.rows();
switch (edgeID) {
case 0:
if (edge[0] > nr) {
return false;
}
for (int i = edge[3]; i <= edge[1]; ++i) {
if (img.ptr(edge[0], i).getInt() == 255) {// 遇见255像素表明碰到边缘线
return false;
}
}
edge[0]++;
return true;
case 1:
if (edge[1] > nc) {
return false;
}
for (int i = edge[2]; i <= edge[0]; ++i) {
if (img.ptr(i, edge[1]).getInt() == 255)
return false;
}
edge[1]++;
return true;
case 2:
if (edge[2] < 0) {
return false;
}
for (int i = edge[3]; i <= edge[1]; ++i) {
if (img.ptr(edge[2], i).getInt() == 255)
return false;
}
edge[2]--;
return true;
case 3:
if (edge[3] < 0) {
return false;
}
for (int i = edge[2]; i <= edge[0]; ++i) {
if (img.ptr(i, edge[3]).getInt() == 255)
return false;
}
edge[3]--;
return true;
default:
return false;
}
}
/**
* 对于二值图，0代表黑色，255代表白色。
* 去除小连通区域与孔洞，小连通区域用8邻域，孔洞用4邻域
* removeSmallRegion(dst, erzhi,100, 1, 1);
* removeSmallRegion(erzhi, erzhi,100, 0, 0);
* @param Src 二值图
* @param Dst 返回值
* @param AreaLimit 100
* @param checkMode 0代表去除黑区域，1代表去除白区域
* @param mode 0代表4邻域，1代表8邻域;
*/
public static void removeSmallRegion(Mat Src, Mat Dst, int AreaLimit, int checkMode, int mode, Boolean debug, String tempPath) {
// 新建一幅标签图像初始化为0像素点，为了记录每个像素点检验状态的标签，0代表未检查，1代表正在检查,2代表检查不合格（需要反转颜色），3代表检查合格或不需检查
// 初始化的图像全部为0，未检查; 全黑图像
Mat PointLabel = new Mat(Src.size(), opencv_core.CV_8UC1);
// opencv_imgcodecs.imwrite(tempPath + "99_remove.jpg", PointLabel);
if (checkMode == 1) {// 去除小连通区域的白色点
for (int i = 0; i < Src.rows(); i++) {
for (int j = 0; j < Src.cols(); j++) {
if (Src.ptr(i, j).getInt() < 10) {
PointLabel.ptr(i, j).putInt(3); // 将背景黑色点标记为合格，像素为3
}
}
}
} else {// 去除孔洞，黑色点像素
for (int i = 0; i < Src.rows(); i++) {
for (int j = 0; j < Src.cols(); j++) {
if (Src.ptr(i, j).getInt() > 10) {
PointLabel.ptr(i, j).putInt(3);// 如果原图是白色区域，标记为合格，像素为3
}
}
}
}
Vector<Point2d> neihbor = new Vector<Point2d>();// 将邻域压进容器
neihbor.add(new Point2d(-1, 0));
neihbor.add(new Point2d(1, 0));
neihbor.add(new Point2d(0, -1));
neihbor.add(new Point2d(0, 1));
if (mode == 1) { // 8邻域
neihbor.add(new Point2d(-1, -1));
neihbor.add(new Point2d(-1, 1));
neihbor.add(new Point2d(1, -1));
neihbor.add(new Point2d(1, 1));
}
int neihborCount = 4 + 4 * mode;
int CurrX = 0, CurrY = 0;
// 开始检测
for (int i = 0; i < Src.rows(); i++) {
for (int j = 0; j < Src.cols(); j++) {
if (PointLabel.ptr(i, j).getInt() == 0) {// 标签图像像素点为0，表示还未检查的不合格点
Vector<Point2d> GrowBuffer = new Vector<Point2d>(); // 记录检查像素点的个数
GrowBuffer.add(new Point2d(j, i));
PointLabel.ptr(i, j).putInt(1);// 标记为正在检查
int CheckResult = 0;
for (int z = 0; z < GrowBuffer.size(); z++) {
for (int q = 0; q < neihborCount; q++) {
CurrX = (int) (GrowBuffer.get(z).x() + neihbor.get(q).x());
CurrY = (int) (GrowBuffer.get(z).y() + neihbor.get(q).y());
if (CurrX >= 0 && CurrX < Src.cols() && CurrY >= 0 && CurrY < Src.rows()) { // 防止越界
if (PointLabel.ptr(CurrY, CurrX).getInt() == 0) {
GrowBuffer.add(new Point2d(CurrX, CurrY)); // 邻域点加入buffer
PointLabel.ptr(CurrY, CurrX).putInt(1); // 更新邻域点的检查标签，避免重复检查
}
}
}
}
if (GrowBuffer.size() > AreaLimit) { // 判断结果（是否超出限定的大小），1为未超出，2为超出
CheckResult = 2;
} else {
CheckResult = 1;
}
for (int z = 0; z < GrowBuffer.size(); z++) {
CurrX = (int) GrowBuffer.get(z).x();
CurrY = (int) GrowBuffer.get(z).y();
PointLabel.ptr(CurrY, CurrX).putInt(CheckResult);// 标记不合格的像素点，像素值为2
}
}
}
}
// 开始反转面积过小的区域
checkMode = 255 * (1 - checkMode);
for (int i = 0; i < Src.rows(); ++i) {
for (int j = 0; j < Src.cols(); ++j) {
if (PointLabel.ptr(i, j).getInt() == 2) {
Dst.ptr(i, j).putInt(checkMode);
} else if (PointLabel.ptr(i, j).getInt() == 3) {
Dst.ptr(i, j).put(Src.ptr(i, j));
}
}
}
}
}

Write Preview
Loading…
Cancel
Save