[OpenCV]代码整理

开发环境：Windows7, VS2010, OpenCV2.4.10

1.图像特征匹配

 // AxFeatureExtract.cpp : 定义控制台应用程序的入口点。
 //

 #include "stdafx.h"
 #include "opencv2/opencv.hpp"
 #include <opencv2/features2d/features2d.hpp>
 #include <opencv2/nonfree/nonfree.hpp>
 #include <opencv2/calib3d/calib3d.hpp>
 #include "vector"

 using namespace std;

 int _tmain(int argc, _TCHAR* argv[])
 {
     cv::Mat rgb1 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\rgb_png\\1.png");
     cv::Mat rgb2 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\rgb_png\\2.png");
     cv::Mat depth1 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\depth_png\\1.png", -);
     cv::Mat depth2 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\depth_png\\2.png", -);

     // 声明特征提取器与描述子提取器
     cv::Ptr<cv::FeatureDetector> _detector;
     cv::Ptr<cv::DescriptorExtractor> _descriptor;

     // 构建提取器，默认两者都为sift
     // 构建sift, surf之前要初始化nonfree模块
     cv::initModule_nonfree();
     _detector = cv::FeatureDetector::create( "GridSIFT" );
     _descriptor = cv::DescriptorExtractor::create( "SIFT" );

     vector< cv::KeyPoint > kp1, kp2; //关键点
     _detector->detect( rgb1, kp1 );  //提取关键点
     _detector->detect( rgb2, kp2 );

     cout<<"Key points of two images: "<<kp1.size()<<", "<<kp2.size()<<endl;

     // 可视化， 显示关键点
     cv::Mat imgShow;
     cv::drawKeypoints( rgb1, kp1, imgShow, cv::Scalar::all(-), cv::DrawMatchesFlags::DRAW_RICH_KEYPOINTS );
     cv::imshow( "keypoints", imgShow );
     cv::imwrite( "data\\keypoints.png", imgShow );
     cv::waitKey(); //暂停等待一个按键

     // 计算描述子
     cv::Mat desp1, desp2;
     _descriptor->compute( rgb1, kp1, desp1 );
     _descriptor->compute( rgb2, kp2, desp2 );

     // 匹配描述子
     vector< cv::DMatch > matches;
     cv::FlannBasedMatcher matcher;
     matcher.match( desp1, desp2, matches );
     cout<<"Find total "<<matches.size()<<" matches."<<endl;

     // 可视化：显示匹配的特征
     cv::Mat imgMatches;
     cv::drawMatches( rgb1, kp1, rgb2, kp2, matches, imgMatches );
     cv::imshow( "matches", imgMatches );
     cv::imwrite( "data\\matches.png", imgMatches );
     cv::waitKey(  );

     // 筛选匹配，把距离太大的去掉
     // 这里使用的准则是去掉大于四倍最小距离的匹配
     vector< cv::DMatch > goodMatches;
     double minDis = ;
     for ( size_t i=; i<matches.size(); i++ )
     {
             if ( matches[i].distance < minDis )
                 minDis = matches[i].distance;
     }

     for ( size_t i=; i<matches.size(); i++ )
     {
         if (matches[i].distance < *minDis)
             goodMatches.push_back( matches[i] );
     }

         // 显示 good matches
     cout<<"good matches="<<goodMatches.size()<<endl;
     cv::drawMatches( rgb1, kp1, rgb2, kp2, goodMatches, imgMatches );
     cv::imshow( "good matches", imgMatches );
     cv::imwrite( "data\\good_matches.png", imgMatches );
     system("pause");
     return ;
 }

2.膨胀和侵蚀

形态学方法：读取图像，实现图像的膨胀和侵蚀操作。

 // AxFeatureExtract.cpp : 定义控制台应用程序的入口点。
 //

 #include "stdafx.h"
 #include "opencv2/opencv.hpp"
 #include <opencv2/core/core.hpp>
 #include <opencv2/features2d/features2d.hpp>
 #include <opencv2/nonfree/nonfree.hpp>
 #include <opencv2/calib3d/calib3d.hpp>
 #include <opencv2/imgproc/imgproc.hpp>
 #include "vector"

 using namespace std;
 using namespace cv; 

 int _tmain(int argc, _TCHAR* argv[])
 {
     cv::Mat rgb1 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\rgb_png\\1.png");
     cv::Mat depth1 = cv::imread( "G:\\SLAMDatasets\\一起做RGB-DSLAM数据\\depth_png\\1.png", -);

     cv::imshow( "matches", rgb1 );
     cv::Mat elementdilate = getStructuringElement(MORPH_RECT, Size(, ));
     cv::Mat outdilate;
     //进行膨胀操作
     dilate(rgb1, outdilate, elementdilate);
     cv::imshow( "dilate", outdilate );
     cv::imwrite( "data\\dilate.png", outdilate );
     cv::waitKey(  );

     cv::Mat elementerode = getStructuringElement(MORPH_RECT, Size(, ));
     cv::Mat outerode;  

     //进行腐蚀操作
     erode(depth1,outerode, elementerode);
     cv::imshow( "erode", outerode );
     cv::imwrite( "data\\outerode.png", outerode );
     system("pause");
     return ;
 }

3.图像二值化

4.图像边缘提取

 // AxFeatureExtract.cpp : 定义控制台应用程序的入口点。
 //

 #include "stdafx.h"
 #include "opencv2/opencv.hpp"
 #include <opencv2/core/core.hpp>
 #include <opencv2/features2d/features2d.hpp>
 #include <opencv2/nonfree/nonfree.hpp>
 #include <opencv2/calib3d/calib3d.hpp>
 #include <opencv2/imgproc/imgproc.hpp>
 #include "vector"

 using namespace std;
 using namespace cv;
 void RemoveSmallRegion(Mat& Src, Mat& Dst, int AreaLimit=, int CheckMode=, int NeihborMode=);

 int _tmain(int argc, _TCHAR* argv[])
 {
     cv::Mat rgb1 = cv::imread( "data\\dump-00050.png");
     cv::imshow( "dump", rgb1 );

     cv::waitKey(  );
     //提取边缘
     cv::Mat src=rgb1.clone();
     cv::Mat dstSobel;
     Sobel(src,dstSobel,src.depth(),,);
     cv::imshow( "Sobel", dstSobel );
     imwrite("sobel.jpg",dstSobel);

     cv::Mat dstLaplacian;
     Laplacian(src,dstLaplacian,src.depth());
     cv::imshow("Laplacian", dstLaplacian );
     imwrite("laplacian.jpg",dstLaplacian);

     cv::Mat g_grayImage;
     cv::Mat dstCanny;
     cvtColor(src,g_grayImage,CV_BGR2GRAY);  //canny只处理灰度图
     Canny(g_grayImage,dstCanny,,,);
     cv::imshow("Canny", dstCanny );
     imwrite("canny.jpg",dstCanny);

     cv::Mat g_dstImage;
     threshold(g_grayImage,g_dstImage,,,CV_THRESH_BINARY_INV);  //大津阈值
     imshow("threshold",g_dstImage);
     imwrite("threshold.jpg",g_dstImage);

     cv::Mat elementdilate = getStructuringElement(MORPH_RECT, Size(, ));
     cv::Mat outdilate;
     //进行膨胀操作
     dilate(g_dstImage, outdilate, elementdilate);
     cv::imshow( "dilate", outdilate );
     //cv::imwrite( "data\\dilate.png", outdilate );

     cv::Mat outresult;
     cv::Mat image=dstLaplacian.clone();
     cv::subtract(dstCanny,  outdilate,  outresult);//注意保证两个图像的格式一致
     cv::imshow( "frontier", outresult );

     Mat rgb2 = Mat::zeros(outresult.size(), CV_8UC1);
     /*blur(outresult,rgb2,Size(3,3),Point(-1,-1));
     cv::imshow( "blur", rgb2 );*/
     RemoveSmallRegion(outresult,rgb2,,, );
     cv::imshow( "blur", rgb2 );
     cv::waitKey(  );
     system("pause");
     return ;
 }

 void RemoveSmallRegion(Mat& Src, Mat& Dst, int AreaLimit, int CheckMode, int NeihborMode)
 {
     int RemoveCount=;       //记录除去的个数
     //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
     Mat Pointlabel = Mat::zeros( Src.size(), CV_8UC1 );  

     if(CheckMode==)
     {
         cout<<"Mode: 去除小区域. ";
         for(int i = ; i < Src.rows; ++i)
         {
             uchar* iData = Src.ptr<uchar>(i);
             uchar* iLabel = Pointlabel.ptr<uchar>(i);
             for(int j = ; j < Src.cols; ++j)
             {
                 if (iData[j] < )
                 {
                     iLabel[j] = ;
                 }
             }
         }
     }
     else
     {
         cout<<"Mode: 去除孔洞. ";
         for(int i = ; i < Src.rows; ++i)
         {
             uchar* iData = Src.ptr<uchar>(i);
             uchar* iLabel = Pointlabel.ptr<uchar>(i);
             for(int j = ; j < Src.cols; ++j)
             {
                 if (iData[j] > )
                 {
                     iLabel[j] = ;
                 }
             }
         }
     }  

     vector<Point2i> NeihborPos;  //记录邻域点位置
     NeihborPos.push_back(Point2i(-, ));
     NeihborPos.push_back(Point2i(, ));
     NeihborPos.push_back(Point2i(, -));
     NeihborPos.push_back(Point2i(, ));
     if (NeihborMode==)
     {
         cout<<"Neighbor mode: 8邻域."<<endl;
         NeihborPos.push_back(Point2i(-, -));
         NeihborPos.push_back(Point2i(-, ));
         NeihborPos.push_back(Point2i(, -));
         NeihborPos.push_back(Point2i(, ));
     }
     else cout<<"Neighbor mode: 4邻域."<<endl;
     int NeihborCount=+*NeihborMode;
     int CurrX=, CurrY=;
     //开始检测
     for(int i = ; i < Src.rows; ++i)
     {
         uchar* iLabel = Pointlabel.ptr<uchar>(i);
         for(int j = ; j < Src.cols; ++j)
         {
             if (iLabel[j] == )
             {
                 //********开始该点处的检查**********
                 vector<Point2i> GrowBuffer;                                      //堆栈，用于存储生长点
                 GrowBuffer.push_back( Point2i(j, i) );
                 Pointlabel.at<uchar>(i, j)=;
                 int CheckResult=;                                               //用于判断结果（是否超出大小），0为未超出，1为超出  

                 for ( int z=; z<GrowBuffer.size(); z++ )
                 {  

                     for (int q=; q<NeihborCount; q++)                                      //检查四个邻域点
                     {
                         CurrX=GrowBuffer.at(z).x+NeihborPos.at(q).x;
                         CurrY=GrowBuffer.at(z).y+NeihborPos.at(q).y;
                         if (CurrX>=&&CurrX<Src.cols&&CurrY>=&&CurrY<Src.rows)  //防止越界
                         {
                             if ( Pointlabel.at<uchar>(CurrY, CurrX)== )
                             {
                                 GrowBuffer.push_back( Point2i(CurrX, CurrY) );  //邻域点加入buffer
                                 Pointlabel.at<uchar>(CurrY, CurrX)=;           //更新邻域点的检查标签，避免重复检查
                             }
                         }
                     }  

                 }
                 if (GrowBuffer.size()>AreaLimit) CheckResult=;                 //判断结果（是否超出限定的大小），1为未超出，2为超出
                 else {CheckResult=;   RemoveCount++;}
                 for (int z=; z<GrowBuffer.size(); z++)                         //更新Label记录
                 {
                     CurrX=GrowBuffer.at(z).x;
                     CurrY=GrowBuffer.at(z).y;
                     Pointlabel.at<uchar>(CurrY, CurrX) += CheckResult;
                 }
                 //********结束该点处的检查**********  

             }
         }
     }    

     CheckMode=*(-CheckMode);
     //开始反转面积过小的区域
     for(int i = ; i < Src.rows; ++i)
     {
         uchar* iData = Src.ptr<uchar>(i);
         uchar* iDstData = Dst.ptr<uchar>(i);
         uchar* iLabel = Pointlabel.ptr<uchar>(i);
         for(int j = ; j < Src.cols; ++j)
         {
             if (iLabel[j] == )
             {
                 iDstData[j] = CheckMode;
             }
             else if(iLabel[j] == )
             {
                 iDstData[j] = iData[j];
             }
         }
     }   

     cout<<RemoveCount<<" objects removed."<<endl;
 }