基于C++传统图像处理方法实现导光板缺陷检测源码+项目说明+详细注释.zip

#include <opencv2\opencv.hpp> #include <iostream> #include <string> #include <stdio.h> //#include <math.h> //#include "rect_detector.h" //#include "artifacts_detector.h" //#include "mura_detector.h" //#include "curve_detector.h" #include <cv.hpp> using namespace cv; using namespace std; //定义灰度图像变量 IplImage *g_GrayImage = NULL; //定义二值化图片变量 IplImage *g_BinaryImage = NULL; //定义二值化窗口标题 const char *WindowBinaryTitle = "二值化图片"; //定义滑块响应函数 //创建源图像窗口标题变量 const char *WindowSrcTitle = "灰度图像"; //创建滑块标题变量 const char *TheSliderTitle = "二值化阀值"; //const char *SrcPath = "C:\\Users\\lenovo\\Desktop\\1.jpg"; ////定义图片路径 IplImage *g_pGrayImage_liantong = NULL; IplImage *g_pBinralyImage_liantong = NULL; int contour_num = 0; //数字编号 char number_buf[10]; ////数字编号存入数组，puttext #define num_col 11 ////二维数组的列，每一个点缺陷信息的详细信息 long int liantong_all_area = 0; ////连通区域总面积 long int Rect_all_area = 0; //// 保存最小外接矩形总的面积 ////===================================================================== struct my_struct1{ double scale; //// 定义显示图像的比例 const int threshold_value_binaryzation; ////定义第一次二值化阀值 const int threshold_value_second_binaryzation; ////定义第二次二值化阀值 }; my_struct1 picture = { 0.5, 50, 100 }; ////===================================================================== struct my_struct2{ int Model1_k1; ////图像膨胀腐蚀 int Model1_k2; ////图像膨胀腐蚀 int Model2_k1; ////图像膨胀腐蚀 int Model2_k2; ////图像膨胀腐蚀 }; my_struct2 value = { 5, 2, 3, 2 }; ////===================================================================== struct my_struct3{ double maxarea; ////最大缺陷面积 double minarea; ////最小显示保留的缺陷面积 double font_scale; ////字体大小 int font_thickness; ////字体粗细 const int Feature_value2_number; ////定义一个二维数组的列，即缺陷的个数 }; my_struct3 value2 = { 5000, 1000, 0.6, 0.8, 100 }; ////===================================================================== struct my_struct4{ const int hough_Canny_thresh1; const int hough_Canny_thresh2; const int hough_Canny_kernel; const int cvHoughLines2_thresh; ////像素值大于多少才显示，值越大，显示的线段越少 const int cvHoughLines2_param1; ////显示线段的最小长度 const int cvHoughLines2_param2; ////线段之间的 最小间隔 }; my_struct4 Hough = { 50, 100, 3, 50, 20, 10 }; void fft2(IplImage *src, IplImage *dst) { //实部、虚部 IplImage *image_Re = 0, *image_Im = 0, *Fourier = 0; // int i, j; image_Re = cvCreateImage(cvGetSize(src), IPL_DEPTH_64F, 1); //实部 //Imaginary part image_Im = cvCreateImage(cvGetSize(src), IPL_DEPTH_64F, 1); //虚部 //2 channels (image_Re, image_Im) Fourier = cvCreateImage(cvGetSize(src), IPL_DEPTH_64F, 2); // Real part conversion from u8 to 64f (double) cvConvertScale(src, image_Re); // Imaginary part (zeros) cvZero(image_Im); // Join real and imaginary parts and stock them in Fourier image cvMerge(image_Re, image_Im, 0, 0, Fourier); // Application of the forward Fourier transform cvDFT(Fourier, dst, CV_DXT_FORWARD); cvReleaseImage(&image_Re); cvReleaseImage(&image_Im); cvReleaseImage(&Fourier); } void fft2shift(IplImage *src, IplImage *dst) { IplImage *image_Re = 0, *image_Im = 0; int nRow, nCol, i, j, cy, cx; double scale, shift, tmp13, tmp24; image_Re = cvCreateImage(cvGetSize(src), IPL_DEPTH_64F, 1); //Imaginary part image_Im = cvCreateImage(cvGetSize(src), IPL_DEPTH_64F, 1); cvSplit(src, image_Re, image_Im, 0, 0); //具体原理见冈萨雷斯数字图像处理p123 // Compute the magnitude of the spectrum Mag = sqrt(Re^2 + Im^2) //计算傅里叶谱 cvPow(image_Re, image_Re, 2.0); cvPow(image_Im, image_Im, 2.0); cvAdd(image_Re, image_Im, image_Re); cvPow(image_Re, image_Re, 0.5); //对数变换以增强灰度级细节(这种变换使以窄带低灰度输入图像值映射 //一宽带输出值，具体可见冈萨雷斯数字图像处理p62) // Compute log(1 + Mag); cvAddS(image_Re, cvScalar(1.0), image_Re); // 1 + Mag cvLog(image_Re, image_Re); // log(1 + Mag) //Rearrange the quadrants of Fourier image so that the origin is at the image center nRow = src->height; nCol = src->width; cy = nRow / 2; // image center cx = nCol / 2; //CV_IMAGE_ELEM为OpenCV定义的宏，用来读取图像的像素值，这一部分就是进行中心变换 for (j = 0; j < cy; j++){ for (i = 0; i < cx; i++){ //中心化，将整体份成四块进行对角交换 tmp13 = CV_IMAGE_ELEM(image_Re, double, j, i); CV_IMAGE_ELEM(image_Re, double, j, i) = CV_IMAGE_ELEM( image_Re, double, j + cy, i + cx); CV_IMAGE_ELEM(image_Re, double, j + cy, i + cx) = tmp13; tmp24 = CV_IMAGE_ELEM(image_Re, double, j, i + cx); CV_IMAGE_ELEM(image_Re, double, j, i + cx) = CV_IMAGE_ELEM(image_Re, double, j + cy, i); CV_IMAGE_ELEM(image_Re, double, j + cy, i) = tmp24; } } //归一化处理将矩阵的元素值归一为[0,255] //[(f(x,y)-minVal)/(maxVal-minVal)]*255 double minVal = 0, maxVal = 0; // Localize minimum and maximum values cvMinMaxLoc(image_Re, &minVal, &maxVal); // Normalize image (0 - 255) to be observed as an u8 image scale = 255 / (maxVal - minVal); shift = -minVal * scale; cvConvertScale(image_Re, dst, scale, shift); cvReleaseImage(&image_Re); cvReleaseImage(&image_Im); } ////===================================================================== //自适应中值滤波 uchar adaptiveProcess(const Mat &im, int row, int col, int kernelSize, int maxSize) { vector<uchar> pixels; for (int a = -kernelSize / 2; a <= kernelSize / 2; a++) for (int b = -kernelSize / 2; b <= kernelSize / 2; b++) { pixels.push_back(im.at<uchar>(row + a, col + b)); } sort(pixels.begin(), pixels.end()); auto min = pixels[0]; auto max = pixels[kernelSize * kernelSize - 1]; auto med = pixels[kernelSize * kernelSize / 2]; auto zxy = im.at<uchar>(row, col); if (med > min && med < max) { // to B if (zxy > min && zxy < max) return zxy; else return med; } else { kernelSize += 2; if (kernelSize <= maxSize) return adaptiveProcess(im, row, col, kernelSize, maxSize); // 增大窗口尺寸，继续A过程。 else return med; } } int** on_trackbar(const char *SrcPath = "1.BMP"){ CvSeq* contour = 0; CvSeq* _contour = contour; //定义存放数组的二维数组，返回指针数组 int** Feature_value2 = 0; Feature_value2 = new int*[value2.Feature_value2_number]; IplImage *SrcImage_or; CvSize src_sz; ////===============================================================================================预处理 //载入原图 printf("预处理\n"); IplImage *SrcImage_origin = cvLoadImage(SrcPath, CV_LOAD_IMAGE_UNCHANGED); //缩放 src_sz.width = SrcImage_origin->width* picture.scale; src_sz.height = SrcImage_origin->height* picture.scale; SrcImage_or = cvCreateImage(src_sz, SrcImage_origin->depth, SrcImage_origin->nChannels); cvResize(SrcImage_origin, SrcImage_or, CV_INTER_CUBIC); //cvNamedWindow("原图", 0); ////显示原图到原图窗口 //cvShowImage("原图", SrcImage_or); //单通道灰度化处理 if (SrcImage_or->nChannels > 1) { g_GrayImage = cvCreateImage(cvSize(SrcImage_or->width, SrcImage_or->height), IPL_DEPTH_8U, 1); cvCvtColor(SrcImage_or, g_GrayImage, CV_BGR2GRAY); } else g_GrayImage = SrcImage_or; //抑制曝光过度 //IplImage *src_threshold = cvCreateImage(cvGetSize(SrcImage_or), IPL_DEPTH_8U, 1); //cvThreshold(SrcImage_or, src_threshold, 100, 255, CV_THRESH_BINARY); for (int i = 0; i < src_sz.height; i++) { for (int j = 0; j < src_sz.width; j++) { if (cvGet2D(g_GrayImage, i, j).val[0]>100) cvSet2D(g_GrayImage, i, j, 100); } } //cvNamedWindow("抑制", 0); //////显示原图到原图窗口 //cvShowImage("抑制", SrcImage_or); /// 应用直方图均衡化 IplImage *src_his = cvCreateImage(src_sz, g_GrayImage->depth, g_GrayImage->nChannels); cvEqualizeHist(g_GrayImage, src_his); cvSaveImage("均衡化.jpg",src_his); //fft�