(转)Opencv卷积操作

转自：http://www.2cto.com/kf/201312/267308.html

Mask Operation filter2D函数 Last Edit 2013/12/24 所谓的Mask Operation就是滤波。 第一步：建立Mask:

?

1 2 3

Mat kern = (Mat_<char>(3,3) <<  0, -1,  0,                                -1,  5, -1,                                 0, -1,  0);</char>

Mat_是一个模板，建立了一个3*3的矩阵，矩阵的值在-128~127. 
第二步：使用filter2D. 函数原型：

?

1 2 3 4 5 6 7 8

void filter2D(InputArray src, //要进行滤波的图像               OutputArray dst,//滤波后的图像               int ddepth,     //原图像的深度  src.depth()               InputArray kernel, //第一步建立的Mask               Point anchor=Point(-1,-1),//Mask的中心点               double delta=0, //Optional value added to the filtered pixels before storing them in dst               int borderType=BORDER_DEFAULT                )

?

1	filter2D(I, K, I.depth(), kern );

以下是OpenCV2.0提供的sample:

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

#include <opencv2 core="" core.hpp=""> #include <opencv2 highgui="" highgui.hpp=""> #include <opencv2 imgproc="" imgproc.hpp=""> #include <iostream>   using namespace std; using namespace cv;   void help(char* progName) {     cout << endl         <<  "This program shows how to filter images with mask: the write it yourself and the"         << "filter2d way. " << endl         <<  "Usage:"                                                                        << endl         << progName << " [image_name -- default lena.jpg] [G -- grayscale] "        << endl << endl; }     void Sharpen(const Mat& myImage,Mat& Result);   int main( int argc, char* argv[]) {     help(argv[0]);     const char* filename = argc >=2 ? argv[1] : "lena.jpg";       Mat I, J, K;       if (argc >= 3 && !strcmp("G", argv[2]))         I = imread( filename, CV_LOAD_IMAGE_GRAYSCALE);     else         I = imread( filename, CV_LOAD_IMAGE_COLOR);       namedWindow("Input", CV_WINDOW_AUTOSIZE);     namedWindow("Output", CV_WINDOW_AUTOSIZE);       imshow("Input", I);     double t = (double)getTickCount();           Sharpen(I, J);           t = ((double)getTickCount() - t)/getTickFrequency();     cout << "Hand written function times passed in seconds: " << t << endl;       imshow("Output", J);     cvWaitKey(0);       Mat kern = (Mat_<char>(3,3) <<  0, -1,  0,                                    -1,  5, -1,                                     0, -1,  0);     t = (double)getTickCount();     filter2D(I, K, I.depth(), kern );     t = ((double)getTickCount() - t)/getTickFrequency();     cout << "Built-in filter2D time passed in seconds:      " << t << endl;       imshow("Output", K);       cvWaitKey(0);     return 0; } void Sharpen(const Mat& myImage,Mat& Result) {     CV_Assert(myImage.depth() == CV_8U);  // accept only uchar images       const int nChannels = myImage.channels();     Result.create(myImage.size(),myImage.type());           for(int j = 1 ; j < myImage.rows-1; ++j)     {         const uchar* previous = myImage.ptr<uchar>(j - 1);         const uchar* current  = myImage.ptr<uchar>(j    );         const uchar* next     = myImage.ptr<uchar>(j + 1);           uchar* output = Result.ptr<uchar>(j);           for(int i= nChannels;i < nChannels*(myImage.cols-1); ++i)         {             *output++ = saturate_cast<uchar>(5*current[i]                          -current[i-nChannels] - current[i+nChannels] - previous[i] - next[i]);         }     }       Result.row(0).setTo(Scalar(0));     Result.row(Result.rows-1).setTo(Scalar(0));     Result.col(0).setTo(Scalar(0));     Result.col(Result.cols-1).setTo(Scalar(0)); }</uchar></uchar></uchar></uchar></uchar></char></iostream></opencv2></o