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

切换为opencv官方包

Browse Source
devA
yuxue 5 years ago
parent 5aab00224e
commit afb9562ffa
1 changed files with 212 additions and 151 deletions
Show Stats Download Patch File Download Diff File

361
src/main/java/com/yuxue/util/ImageUtil.java
Unescape View File

@ -1,25 +1,40 @@
package com.yuxue.util; package com.yuxue.util;
import java.util.Arrays; import java.util.Arrays;
import java.util.List;
import java.util.Map; import java.util.Map;
import java.util.Set; import java.util.Set;
import java.util.Vector; import java.util.Vector;
import org.opencv.core.Core;
import org.opencv.core.CvType;
import org.opencv.core.Mat;
import org.opencv.core.MatOfPoint;
import org.opencv.core.MatOfPoint2f;
import org.opencv.core.RotatedRect;
import org.opencv.core.Scalar;
import org.opencv.core.Size;
import org.opencv.imgcodecs.Imgcodecs;
import org.opencv.imgproc.Imgproc;
import org.opencv.ml.ANN_MLP;
import org.opencv.ml.SVM;
import com.google.common.collect.Lists;
/*
import org.bytedeco.javacpp.BytePointer; import org.bytedeco.javacpp.BytePointer;
import org.bytedeco.javacpp.IntPointer;
import org.bytedeco.javacpp.opencv_core; import org.bytedeco.javacpp.opencv_core;
import org.bytedeco.javacpp.opencv_core.Mat; import org.bytedeco.javacpp.Core.Mat;
import org.bytedeco.javacpp.opencv_core.MatVector; import org.bytedeco.javacpp.Core.MatVector;
import org.bytedeco.javacpp.opencv_core.Point2d; import org.bytedeco.javacpp.Core.Point2d;
import org.bytedeco.javacpp.opencv_core.Point2f; import org.bytedeco.javacpp.Core.Point2f;
import org.bytedeco.javacpp.opencv_core.RotatedRect; import org.bytedeco.javacpp.Core.RotatedRect;
import org.bytedeco.javacpp.opencv_core.Scalar; import org.bytedeco.javacpp.Core.Scalar;
import org.bytedeco.javacpp.opencv_core.Size; import org.bytedeco.javacpp.Core.Size;
import org.bytedeco.javacpp.opencv_ml.ANN_MLP; import org.bytedeco.javacpp.opencv_ml.ANN_MLP;
import org.bytedeco.javacpp.opencv_ml.SVM; import org.bytedeco.javacpp.opencv_ml.SVM;
import org.bytedeco.javacpp.opencv_imgcodecs; import org.bytedeco.javacpp.opencv_imgcodecs;
import org.bytedeco.javacpp.opencv_imgproc; import org.bytedeco.javacpp.opencv_imgproc;
*/
import com.google.common.collect.Maps; import com.google.common.collect.Maps;
@ -31,12 +46,11 @@ import com.google.common.collect.Maps;
*/ */
public class ImageUtil { public class ImageUtil {
private static SVM svm = SVM.create(); private static String DEFAULT_BASE_TEST_PATH = "D:/PlateDetect/temp/";
private static ANN_MLP ann=ANN_MLP.create();
/*private static SVM svm = SVM.create();
private static String DEFAULT_BASE_TEST_PATH = "D:/PlateDetect/temp/"; private static ANN_MLP ann=ANN_MLP.create();
public static void loadSvmModel(String path) { public static void loadSvmModel(String path) {
svm.clear(); svm.clear();
@ -47,6 +61,10 @@ public class ImageUtil {
public static void loadAnnModel(String path) { public static void loadAnnModel(String path) {
ann.clear(); ann.clear();
ann = ANN_MLP.load(path); ann = ANN_MLP.load(path);
}*/
static {
System.loadLibrary(Core.NATIVE_LIBRARY_NAME);
} }
// 车牌定位处理步骤，该map用于表示步骤图片的顺序 // 车牌定位处理步骤，该map用于表示步骤图片的顺序
@ -80,9 +98,10 @@ public class ImageUtil {
String tempPath = DEFAULT_BASE_TEST_PATH + "test/"; String tempPath = DEFAULT_BASE_TEST_PATH + "test/";
String filename = tempPath + "/100_yuantu.jpg"; String filename = tempPath + "/100_yuantu.jpg";
filename = tempPath + "/100_yuantu1.jpg"; //filename = tempPath + "/100_yuantu1.jpg";
//filename = tempPath + "/109_crop_0.png";
Mat src = opencv_imgcodecs.imread(filename); Mat src = Imgcodecs.imread(filename);
Boolean debug = true; Boolean debug = true;
@ -91,18 +110,18 @@ public class ImageUtil {
Mat grey = ImageUtil.grey(gsMat, debug, tempPath); Mat grey = ImageUtil.grey(gsMat, debug, tempPath);
Mat sobel = ImageUtil.sobel(grey, debug, tempPath); Mat sobel = ImageUtil.sobel(grey, debug, tempPath);
// Mat sobel = ImageUtil.scharr(grey, debug, tempPath); // Mat sobel = ImageUtil.scharr(grey, debug, tempPath);
// Mat threshold = ImageUtil.threshold(sobel, debug, tempPath); Mat threshold = ImageUtil.threshold(sobel, debug, tempPath);
Mat morphology = ImageUtil.morphology(ImageUtil.threshold(sobel, debug, tempPath), debug, tempPath); Mat morphology = ImageUtil.morphology(threshold, debug, tempPath);
MatVector contours = ImageUtil.contours(src, morphology, debug, tempPath); List<MatOfPoint> contours = ImageUtil.contours(src, morphology, debug, tempPath);
Vector<Mat> rects = ImageUtil.screenBlock(src, contours, debug, tempPath); Vector<Mat> rects = ImageUtil.screenBlock(src, contours, debug, tempPath);
// ImageUtil.rgb2Hsv(inMat, debug, tempPath); // ImageUtil.rgb2Hsv(src, debug, tempPath);
// ImageUtil.getHSVValue(src, debug, tempPath);
System.err.println("done!!!"); System.err.println("done!!!");
@ -119,9 +138,9 @@ public class ImageUtil {
public static final int DEFAULT_GAUSSIANBLUR_SIZE = 5; public static final int DEFAULT_GAUSSIANBLUR_SIZE = 5;
public static Mat gaussianBlur(Mat inMat, Boolean debug, String tempPath) { public static Mat gaussianBlur(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat(); Mat dst = new Mat();
opencv_imgproc.GaussianBlur(inMat, dst, new Size(DEFAULT_GAUSSIANBLUR_SIZE, DEFAULT_GAUSSIANBLUR_SIZE), 0, 0, opencv_core.BORDER_DEFAULT); Imgproc.GaussianBlur(inMat, dst, new Size(DEFAULT_GAUSSIANBLUR_SIZE, DEFAULT_GAUSSIANBLUR_SIZE), 0, 0, Core.BORDER_DEFAULT);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("gaussianBlur") + 100) + "_gaussianBlur.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("gaussianBlur") + 100) + "_gaussianBlur.jpg", dst);
} }
return dst; return dst;
} }
@ -136,9 +155,9 @@ public class ImageUtil {
*/ */
public static Mat grey(Mat inMat, Boolean debug, String tempPath) { public static Mat grey(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat(); Mat dst = new Mat();
opencv_imgproc.cvtColor(inMat, dst, opencv_imgproc.CV_RGB2GRAY); Imgproc.cvtColor(inMat, dst, Imgproc.COLOR_BGR2GRAY);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("gray") + 100) + "_gray.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("gray") + 100) + "_gray.jpg", dst);
} }
inMat.release(); inMat.release();
return dst; return dst;
@ -165,20 +184,20 @@ public class ImageUtil {
Mat abs_grad_x = new Mat(); Mat abs_grad_x = new Mat();
Mat abs_grad_y = new Mat(); Mat abs_grad_y = new Mat();
opencv_imgproc.Sobel(inMat, grad_x, opencv_core.CV_16S, 1, 0, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT); Imgproc.Sobel(inMat, grad_x, CvType.CV_16S, 1, 0, 3, SOBEL_SCALE, SOBEL_DELTA, Core.BORDER_DEFAULT);
opencv_core.convertScaleAbs(grad_x, abs_grad_x); Core.convertScaleAbs(grad_x, abs_grad_x);
opencv_imgproc.Sobel(inMat, grad_y, opencv_core.CV_16S, 0, 1, 3, SOBEL_SCALE, SOBEL_DELTA, opencv_core.BORDER_DEFAULT); Imgproc.Sobel(inMat, grad_y, CvType.CV_16S, 0, 1, 3, SOBEL_SCALE, SOBEL_DELTA, Core.BORDER_DEFAULT);
opencv_core.convertScaleAbs(grad_y, abs_grad_y); Core.convertScaleAbs(grad_y, abs_grad_y);
grad_x.release(); grad_x.release();
grad_y.release(); grad_y.release();
opencv_core.addWeighted(abs_grad_x, SOBEL_X_WEIGHT, abs_grad_y, SOBEL_Y_WEIGHT, 0, dst); Core.addWeighted(abs_grad_x, SOBEL_X_WEIGHT, abs_grad_y, SOBEL_Y_WEIGHT, 0, dst);
abs_grad_x.release(); abs_grad_x.release();
abs_grad_y.release(); abs_grad_y.release();
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst);
} }
return dst; return dst;
} }
@ -202,20 +221,20 @@ public class ImageUtil {
//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 //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 //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); Imgproc.Scharr(inMat, grad_x, CvType.CV_32F, 1, 0);
opencv_imgproc.Scharr(inMat, grad_y, opencv_core.CV_32F, 0, 1); Imgproc.Scharr(inMat, grad_y, CvType.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); Core.convertScaleAbs(grad_x, abs_grad_x);
opencv_core.convertScaleAbs(grad_y, abs_grad_y); 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 //openCV中使用release()释放Mat类图像，使用recycle()释放BitMap类图像
grad_x.release(); grad_x.release();
grad_y.release(); grad_y.release();
opencv_core.addWeighted(abs_grad_x, 0.5, abs_grad_y, 0.5, 0, dst); Core.addWeighted(abs_grad_x, 0.5, abs_grad_y, 0.5, 0, dst);
abs_grad_x.release(); abs_grad_x.release();
abs_grad_y.release(); abs_grad_y.release();
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("sobel") + 100) + "_sobel.jpg", dst);
} }
return dst; return dst;
} }
@ -231,9 +250,15 @@ public class ImageUtil {
*/ */
public static Mat threshold(Mat inMat, Boolean debug, String tempPath) { public static Mat threshold(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat(); Mat dst = new Mat();
opencv_imgproc.threshold(inMat, dst, 0, 255, opencv_imgproc.CV_THRESH_OTSU + opencv_imgproc.CV_THRESH_BINARY); Imgproc.threshold(inMat, dst, 100, 255, Imgproc.THRESH_OTSU + Imgproc.THRESH_BINARY);
/*for (int i = 0; i < dst.rows(); i++) {
for (int j = 0; j < dst.cols(); j++) {
System.err.println((int)dst.get(i, j)[0]);
}
}*/
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("threshold") + 100) + "_threshold.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("threshold") + 100) + "_threshold.jpg", dst);
} }
inMat.release(); inMat.release();
return dst; return dst;
@ -252,22 +277,21 @@ public class ImageUtil {
public static final int DEFAULT_MORPH_SIZE_WIDTH = 9; public static final int DEFAULT_MORPH_SIZE_WIDTH = 9;
public static final int DEFAULT_MORPH_SIZE_HEIGHT = 3; public static final int DEFAULT_MORPH_SIZE_HEIGHT = 3;
public static Mat morphology(Mat inMat, Boolean debug, String tempPath) { public static Mat morphology(Mat inMat, Boolean debug, String tempPath) {
Mat dst = new Mat(inMat.size(), opencv_core.CV_8UC1); Mat dst = new Mat(inMat.size(), CvType.CV_8UC1);
Size size = new Size(DEFAULT_MORPH_SIZE_WIDTH, DEFAULT_MORPH_SIZE_HEIGHT); Size size = new Size(DEFAULT_MORPH_SIZE_WIDTH, DEFAULT_MORPH_SIZE_HEIGHT);
Mat element = opencv_imgproc.getStructuringElement(opencv_imgproc.MORPH_RECT, size); Mat element = Imgproc.getStructuringElement(Imgproc.MORPH_RECT, size);
opencv_imgproc.morphologyEx(inMat, dst, opencv_imgproc.MORPH_CLOSE, element); Imgproc.morphologyEx(inMat, dst, Imgproc.MORPH_CLOSE, element);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology0.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology0.jpg", dst);
} }
// 去除小连通区域 // 去除小连通区域
// removeSmallRegion(dst, dst, 100, 1, 1, debug, tempPath); Mat a = clearSmallConnArea(dst, 3, 8, debug, tempPath);
Mat b = clearSmallConnArea(a, 8, 3, debug, tempPath);
// 去除孔洞 // 去除孔洞
// removeSmallRegion(dst, dst, 100, 0, 0, debug, tempPath); Mat c = clearHole(b, 3, 8, debug, tempPath);
Mat d = clearHole(c, 3, 8, debug, tempPath);
clearHole(dst, 136/10, 36/10, debug, tempPath); return d;
return dst;
} }
@ -280,29 +304,24 @@ public class ImageUtil {
* @param tempPath * @param tempPath
* @return * @return
*/ */
public static MatVector contours(Mat src, Mat inMat, Boolean debug, String tempPath) { public static List<MatOfPoint> contours(Mat src, Mat inMat, Boolean debug, String tempPath) {
MatVector contours = new MatVector(); List<MatOfPoint> contours = Lists.newArrayList();
Mat hierarchy = new Mat();
// 提取外部轮廓 // 提取外部轮廓
// CV_RETR_EXTERNAL只检测最外围轮廓， // CV_RETR_EXTERNAL只检测最外围轮廓，
// CV_RETR_LIST 检测所有的轮廓 // 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 // 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); Imgproc.findContours(inMat, contours, hierarchy, Imgproc.RETR_EXTERNAL, Imgproc.CHAIN_APPROX_NONE);
// 在小连接处分割轮廓 // 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) { if (debug) {
Mat result = new Mat(); 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 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 // 将轮廓描绘到原图
opencv_imgproc.drawContours(result, contours, -1, new Scalar(0, 0, 255, 255)); Imgproc.drawContours(result, contours, -1, new Scalar(0, 0, 255, 255));
// 输出带轮廓的原图 // 输出带轮廓的原图
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("contours") + 100) + "_contours.jpg", result); Imgcodecs.imwrite(tempPath + (debugMap.get("contours") + 100) + "_contours.jpg", result);
} }
// return retContour;
return contours; return contours;
} }
@ -318,44 +337,45 @@ public class ImageUtil {
public static final int DEFAULT_ANGLE = 30; // 角度判断所用常量 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 WIDTH = 136;
public static final int HEIGHT = 36; public static final int HEIGHT = 36;
public static final int TYPE = opencv_core.CV_8UC3; public static final int TYPE = CvType.CV_8UC3;
@SuppressWarnings("resource") public static Vector<Mat> screenBlock(Mat src, List<MatOfPoint> contours, Boolean debug, String tempPath){
public static Vector<Mat> screenBlock(Mat src, MatVector contours, Boolean debug, String tempPath){
Vector<Mat> dst = new Vector<Mat>(); Vector<Mat> dst = new Vector<Mat>();
MatVector mv = new MatVector(); // 用于在原图上描绘筛选后的结果 List<MatOfPoint> mv = Lists.newArrayList(); // 用于在原图上描绘筛选后的结果
for (int i = 0, j = 0; i < contours.size(); i++) { for (int i = 0, j = 0; i < contours.size(); i++) {
MatOfPoint m1 = contours.get(i);
MatOfPoint2f m2 = new MatOfPoint2f();
m1.convertTo(m2, CvType.CV_32F);
// 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 // 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 // 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)); RotatedRect mr = Imgproc.minAreaRect(m2);
float angle = Math.abs(mr.angle()); double angle = Math.abs(mr.angle);
if (checkPlateSize(mr) && angle <= DEFAULT_ANGLE) { // 判断尺寸及旋转角度 ±30°，排除不合法的图块 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)); mv.add(contours.get(i));
Size rect_size = new Size((int) mr.size().width(), (int) mr.size().height()); 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 if (mr.size.width / mr.size.height < 1) { // 宽度小于高度
angle = 90 + angle; // 旋转90° angle = 90 + angle; // 旋转90°
rect_size = new Size(rect_size.height(), rect_size.width()); rect_size = new Size(rect_size.height, rect_size.width);
} }
// 旋转角度，根据需要是否进行角度旋转 // 旋转角度，根据需要是否进行角度旋转
Mat img_rotated = new Mat(); Mat img_rotated = new Mat();
Mat rotmat = opencv_imgproc.getRotationMatrix2D(mr.center(), angle, 1); // 旋转 Mat rotmat = Imgproc.getRotationMatrix2D(mr.center, angle, 1); // 旋转
opencv_imgproc.warpAffine(src, img_rotated, rotmat, src.size()); // 仿射变换 Imgproc.warpAffine(src, img_rotated, rotmat, src.size()); // 仿射变换
// 切图 // 切图
Mat img_crop = new Mat(); Mat img_crop = new Mat();
opencv_imgproc.getRectSubPix(src, rect_size, mr.center(), img_crop); Imgproc.getRectSubPix(src, rect_size, mr.center, img_crop);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("crop") + 100) + "_crop_" + j + ".png", img_crop); Imgcodecs.imwrite(tempPath + (debugMap.get("crop") + 100) + "_crop_" + j + ".png", img_crop);
} }
// 处理切图，调整为指定大小 // 处理切图，调整为指定大小
Mat resized = new Mat(HEIGHT, WIDTH, TYPE); Mat resized = new Mat(HEIGHT, WIDTH, TYPE);
opencv_imgproc.resize(img_crop, resized, resized.size(), 0, 0, opencv_imgproc.INTER_CUBIC); Imgproc.resize(img_crop, resized, resized.size(), 0, 0, Imgproc.INTER_CUBIC);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("resize") + 100) + "_resize_" + j + ".png", resized); Imgcodecs.imwrite(tempPath + (debugMap.get("resize") + 100) + "_resize_" + j + ".png", resized);
j++; j++;
} }
dst.add(resized); dst.add(resized);
@ -365,9 +385,9 @@ public class ImageUtil {
Mat result = new Mat(); 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 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 // 将轮廓描绘到原图
opencv_imgproc.drawContours(result, mv, -1, new Scalar(0, 0, 255, 255)); Imgproc.drawContours(result, mv, -1, new Scalar(0, 0, 255, 255));
// 输出带轮廓的原图 // 输出带轮廓的原图
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("screenblock") + 100) + "_screenblock.jpg", result); Imgcodecs.imwrite(tempPath + (debugMap.get("screenblock") + 100) + "_screenblock.jpg", result);
} }
return dst; return dst;
} }
@ -399,15 +419,13 @@ public class ImageUtil {
float rmax = DEFAULT_ASPECT + DEFAULT_ASPECT * DEFAULT_ERROR; float rmax = DEFAULT_ASPECT + DEFAULT_ASPECT * DEFAULT_ERROR;
// 切图计算面积 // 切图计算面积
int area = (int) (mr.size().height() * mr.size().width()); int area = (int) (mr.size.height * mr.size.width);
// 切图宽高比 // 切图宽高比
float r = mr.size().width() / mr.size().height(); double r = mr.size.width / mr.size.height;
if (r < 1) { if (r < 1) {
r = mr.size().height() / mr.size().width(); r = mr.size.height / mr.size.width;
} }
return min <= area && area <= max && rmin <= r && r <= rmax; return min <= area && area <= max && rmin <= r && r <= rmax;
} }
@ -429,15 +447,15 @@ public class ImageUtil {
public static Mat rgb2Hsv(Mat inMat, Boolean debug, String tempPath) { public static Mat rgb2Hsv(Mat inMat, Boolean debug, String tempPath) {
// 转到HSV空间进行处理 // 转到HSV空间进行处理
Mat dst = new Mat(); Mat dst = new Mat();
opencv_imgproc.cvtColor(inMat, dst, opencv_imgproc.CV_BGR2HSV); Imgproc.cvtColor(inMat, dst, Imgproc.COLOR_BGR2HSV);
MatVector hsvSplit = new MatVector(); List<Mat> hsvSplit = Lists.newArrayList();
opencv_core.split(dst, hsvSplit); Core.split(dst, hsvSplit);
// 直方图均衡化是一种常见的增强图像对比度的方法，使用该方法可以增强局部图像的对比度，尤其在数据较为相似的图像中作用更加明显 // 直方图均衡化是一种常见的增强图像对比度的方法，使用该方法可以增强局部图像的对比度，尤其在数据较为相似的图像中作用更加明显
opencv_imgproc.equalizeHist(hsvSplit.get(2), hsvSplit.get(2)); Imgproc.equalizeHist(hsvSplit.get(2), hsvSplit.get(2));
opencv_core.merge(hsvSplit, dst); Core.merge(hsvSplit, dst);
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + "hsvMat_"+System.currentTimeMillis()+".jpg", dst); Imgcodecs.imwrite(tempPath + "hsvMat_"+System.currentTimeMillis()+".jpg", dst);
} }
return dst; return dst;
} }
@ -453,25 +471,14 @@ public class ImageUtil {
* @param debug * @param debug
*/ */
public static void getHSVValue(Mat inMat, Boolean debug, String tempPath) { public static void getHSVValue(Mat inMat, Boolean debug, String tempPath) {
int channels = inMat.channels();
int nRows = inMat.rows(); int nRows = inMat.rows();
// 图像数据列需要考虑通道数的影响； int nCols = inMat.cols();
int nCols = inMat.cols() * channels;
// 连续存储的数据，按一行处理
if (inMat.isContinuous()) {
nCols *= nRows;
nRows = 1;
}
Map<Integer, Integer> map = Maps.newHashMap(); Map<Integer, Integer> map = Maps.newHashMap();
for (int i = 0; i < nRows; ++i) { for (int i = 0; i < nRows; ++i) {
BytePointer p = inMat.ptr(i);
for (int j = 0; j < nCols; j += 3) { for (int j = 0; j < nCols; j += 3) {
int H = p.get(j) & 0xFF; int H = (int)inMat.get(i, j)[0];
int S = p.get(j + 1) & 0xFF; int S = (int)inMat.get(i, j)[1];
int V = p.get(j + 2) & 0xFF; int V = (int)inMat.get(i, j)[2];
if(map.containsKey(H)) { if(map.containsKey(H)) {
int count = map.get(H); int count = map.get(H);
map.put(H, count+1); map.put(H, count+1);
@ -480,14 +487,12 @@ public class ImageUtil {
} }
} }
} }
Set set = map.keySet(); Set set = map.keySet();
Object[] arr = set.toArray(); Object[] arr = set.toArray();
Arrays.sort(arr); Arrays.sort(arr);
for (Object key : arr) { for (Object key : arr) {
System.out.println(key + ": " + map.get(key)); System.out.println(key + ": " + map.get(key));
} }
return; return;
} }
@ -499,7 +504,7 @@ public class ImageUtil {
* @param inMat * @param inMat
* @return * @return
*/ */
public static RotatedRect maxAreaRect(Mat threshold, Point2f point2f) { /*public static RotatedRect maxAreaRect(Mat threshold, Point2f point2f) {
int edge[] = new int[4]; int edge[] = new int[4];
edge[0] = (int) point2f.x() + 1;//top edge[0] = (int) point2f.x() + 1;//top
edge[1] = (int) point2f.x() + 1;//right edge[1] = (int) point2f.x() + 1;//right
@ -515,12 +520,12 @@ public class ImageUtil {
} }
//[3] //[3]
//qDebug() << edge[0] << edge[1] << edge[2] << edge[3]; //qDebug() << edge[0] << edge[1] << edge[2] << edge[3];
/*Point tl = Point(edge[3], edge[0]); Point tl = Point(edge[3], edge[0]);
Point br = Point(edge[1], edge[2]); Point br = Point(edge[1], edge[2]);
return new Rect(tl, br);*/ return new Rect(tl, br);
return null; return null;
} }*/
/** /**
@ -535,12 +540,12 @@ public class ImageUtil {
int nr = img.rows(); int nr = img.rows();
switch (edgeID) { switch (edgeID) {
case 0: /*case 0:
if (edge[0] > nr) { if (edge[0] > nr) {
return false; return false;
} }
for (int i = edge[3]; i <= edge[1]; ++i) { 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 if (img.get(edge[0], i)[0]== 255) {// 遇见255像素表明碰到边缘线
return false; return false;
} }
} }
@ -575,7 +580,7 @@ public class ImageUtil {
return false; return false;
} }
edge[3]--; edge[3]--;
return true; return true;*/
default: default:
return false; return false;
} }
@ -587,6 +592,7 @@ public class ImageUtil {
* 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 * 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(dst, erzhi,100, 1, 1);
* removeSmallRegion(erzhi, erzhi,100, 0, 0); * removeSmallRegion(erzhi, erzhi,100, 0, 0);
* https://blog.csdn.net/dajiyi1998/article/details/60601410#
* @param Src 二值图 * @param Src 二值图
* @param Dst 返回值 * @param Dst 返回值
* @param AreaLimit 100 * @param AreaLimit 100
@ -596,10 +602,10 @@ public class ImageUtil {
public static void removeSmallRegion(Mat Src, Mat Dst, int AreaLimit, int checkMode, int mode, Boolean debug, String tempPath) { 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 // 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 // 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); Mat PointLabel = new Mat(Src.size(), CvType.CV_8UC1);
// opencv_imgcodecs.imwrite(tempPath + "99_remove.jpg", PointLabel); // Imgcodecs.imwrite(tempPath + "99_remove.jpg", PointLabel);
if (checkMode == 1) {// 去除小连通区域的白色点 /*if (checkMode == 1) {// 去除小连通区域的白色点
for (int i = 0; i < Src.rows(); i++) { for (int i = 0; i < Src.rows(); i++) {
for (int j = 0; j < Src.cols(); j++) { for (int j = 0; j < Src.cols(); j++) {
if (Src.ptr(i, j).getInt() < 10) { if (Src.ptr(i, j).getInt() < 10) {
@ -679,7 +685,7 @@ public class ImageUtil {
Dst.ptr(i, j).put(Src.ptr(i, j)); Dst.ptr(i, j).put(Src.ptr(i, j));
} }
} }
} }*/
} }
@ -687,74 +693,129 @@ public class ImageUtil {
* 清除二值图像的黑洞 * 清除二值图像的黑洞
* 按矩形清理 * 按矩形清理
* @param inMat 二值图像 * @param inMat 二值图像
* @param rowLimit 宽度半径限制 * @param rowLimit 像素值
* @param colsLimit 高度半径限制 * @param colsLimit 像素值
* @param debug * @param debug
* @param tempPath * @param tempPath
*/ */
public static void clearHole(Mat inMat, int rowLimit, int colsLimit, Boolean debug, String tempPath) { public static Mat clearHole(Mat inMat, int rowLimit, int colsLimit, Boolean debug, String tempPath) {
int uncheck = 0, checking = 1, black = 2, white = 3; int uncheck = 0, black = 1, white = 2;
Mat dst = new Mat(inMat.size(), opencv_core.CV_8UC1); Mat dst = new Mat(inMat.size(), CvType.CV_8UC1);
inMat.copyTo(dst); inMat.copyTo(dst);
rowLimit = 2;
colsLimit = 2;
// 初始化的图像全部为0，未检查; 全黑图像 // 初始化的图像全部为0，未检查; 全黑图像
Mat label = new Mat(inMat.size(), opencv_core.CV_8UC1); Mat label = new Mat(inMat.size(), CvType.CV_8UC1);
// 标记所有的白色区域 // 标记所有的白色区域
for (int i = 0; i < inMat.rows(); i++) { for (int i = 0; i < inMat.rows(); i++) {
for (int j = 0; j < inMat.cols(); j++) { for (int j = 0; j < inMat.cols(); j++) {
if (inMat.ptr(i, j).getInt() != 0) { // 对于二值图，0代表黑色，255代表白色 if (inMat.get(i, j)[0] > 10) { // 对于二值图，0代表黑色，255代表白色
label.ptr(i, j).putInt(white); label.put(i, j, white); // 中心点
int x1 = i - rowLimit < 0 ? 0 : i - rowLimit; int x1 = i - rowLimit < 0 ? 0 : i - rowLimit;
int x2 = i + rowLimit > inMat.rows() ? inMat.rows() : i + rowLimit; int x2 = i + rowLimit >= inMat.rows() ? inMat.rows()-1 : i + rowLimit;
int y1 = j - colsLimit < 0 ? 0 : j - colsLimit ; int y1 = j - colsLimit < 0 ? 0 : j - colsLimit ;
int y2 = j + colsLimit > inMat.cols() ? inMat.cols() : j + colsLimit ; int y2 = j + colsLimit >= inMat.cols() ? inMat.cols()-1 : j + colsLimit ;
/*IntPointer p1 = new IntPointer(x1, y1); // 左上角
IntPointer p2 = new IntPointer(x1, y2); // 左下角
IntPointer p3 = new IntPointer(x2, y1); // 右上角
IntPointer p4 = new IntPointer(x2, y2); // 右下角
*/
// System.out.println(x1 + "," + x2 + "\t" + y1 + "," + y2 + "\n");
// System.out.println(inMat.ptr(x1, y1).getInt() + "," + inMat.ptr(x1, y2).getInt() + "," + inMat.ptr(x2, y1).getInt() + "," + inMat.ptr(x2, y2).getInt());
//if(inMat.ptr(x1, y1).getInt() > 10 && inMat.ptr(x1, y2).getInt() > 10 && inMat.ptr(x2, y1).getInt() > 10 && inMat.ptr(x2, y2).getInt() > 10 ) { // 根据中心点+limit，定位四个角生成一个矩形，
// 将四个角都是白色的矩形，内部的黑点标记为 要被替换的对象
if(inMat.get(x1, y1)[0] > 10 && inMat.get(x1, y2)[0] > 10 && inMat.get(x2, y1)[0] > 10 && inMat.get(x2, y2)[0] > 10 ) {
for (int n = x1; n < x2; n++) { for (int n = x1; n < x2; n++) {
for (int m = y1; m < y2; m++) { for (int m = y1; m < y2; m++) {
//System.err.println(n + "," + m); if (inMat.get(n, m)[0] < 10 && label.get(n, m)[0] == uncheck) {
if (/*inMat.ptr(n, m).getInt() < 10 && */label.ptr(n, m).getInt() == uncheck) { label.put(n, m, black);
// System.err.println(n + "," + m);
label.ptr(n, m).putInt(black);
} }
} }
} }
// }
} }
} }
} }
// 1184 //1550 }
int count = 0;
// 替换颜色 // 黑色替换成白色
for (int i = 0; i < inMat.rows(); i++) {
for (int j = 0; j < inMat.cols(); j++) {
if(label.get(i, j)[0] == black) {
dst.put(i, j, 255);
}
}
}
if (debug) {
Imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology2.jpg", dst);
}
return dst;
}
public static Mat clearSmallConnArea(Mat inMat, int rowLimit, int colsLimit, Boolean debug, String tempPath) {
int uncheck = 0, black = 1, white = 2;
Mat dst = new Mat(inMat.size(), CvType.CV_8UC1);
inMat.copyTo(dst);
// 初始化的图像全部为0，未检查; 全黑图像
Mat label = new Mat(inMat.size(), CvType.CV_8UC1);
// 标记所有的白色区域
for (int i = 0; i < inMat.rows(); i++) { for (int i = 0; i < inMat.rows(); i++) {
for (int j = 0; j < inMat.cols(); j++) { for (int j = 0; j < inMat.cols(); j++) {
if(label.ptr(i, j).getInt() == black) { if (inMat.get(i, j)[0] < 10) { // 对于二值图，0代表黑色，255代表白色
dst.ptr(i, j).putInt(255); label.put(i, j, black); // 中心点
count ++;
int x1 = i - rowLimit < 0 ? 0 : i - rowLimit;
int x2 = i + rowLimit >= inMat.rows() ? inMat.rows()-1 : i + rowLimit;
int y1 = j - colsLimit < 0 ? 0 : j - colsLimit ;
int y2 = j + colsLimit >= inMat.cols() ? inMat.cols()-1 : j + colsLimit ;
int count = 0;
if(inMat.get(x1, y1)[0] < 10) {// 左上角
count++;
}
if(inMat.get(x1, y2)[0] < 10) { // 左下角
count++;
} }
if(inMat.get(x2, y1)[0] < 10) { // 右上角
count++;
} }
if(inMat.get(x2, y2)[0] < 10) { // 右下角
count++;
} }
System.err.println(count);
// 根据中心点+limit，定位四个角生成一个矩形，
// 将四个角都是白色的矩形，内部的黑点标记为 要被替换的对象
if(count >= 4) {
for (int n = x1; n < x2; n++) {
for (int m = y1; m < y2; m++) {
if (inMat.get(n, m)[0] > 10 && label.get(n, m)[0] == uncheck) {
label.put(n, m, white);
}
}
}
}
}
}
}
// 黑色替换成白色
for (int i = 0; i < inMat.rows(); i++) {
for (int j = 0; j < inMat.cols(); j++) {
if(label.get(i, j)[0] == white) {
dst.put(i, j, 0);
}
}
}
if (debug) { if (debug) {
opencv_imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology1.jpg", dst); Imgcodecs.imwrite(tempPath + (debugMap.get("morphology") + 100) + "_morphology1.jpg", dst);
} }
return dst;
} }
} }

Write Preview
Loading…
Cancel
Save