學習OpenCV——BOW特征提取函數（特征點篇）

沒日沒夜的改論文生活終於要告一段落了，比起改論文，學OpenCV就是一件幸福的事情。OpenCV的發展越來越完善了，已經可以直接使用BOW函數來進行對象分類了。

 

簡單的通過特征點分類的方法：                                                                      

一、train

1.提取+/- sample的feature，每幅圖提取出的sift特征個數不定（假設每個feature有128維）

2.利用聚類方法（e.g K-means）將不定數量的feature聚類為固定數量的(比如10個)words即BOW（bag of word）

（本篇文章主要完成以上的工作！）

3.normalize，並作這10個類的直方圖e.g [0.1，0.2，0.7，0...0];

4.將each image的這10個word作為feature_instance 和 (手工標記的) label（+/-）進入SVM訓練

二、predict

1. 提取test_img的feature（如137個）

2. 分別求each feature與10個類的距離(e.g. 128維歐氏距離)，確定該feature屬於哪個類

3. normalize，並作這10個類的直方圖e.g [0，0.2，0.2，0.6，0...0];

4. 應用SVM_predict進行結果預測

 

 

通過OpenCV實現feature聚類 BOW                                                             

首先在此介紹一下OpenCV的特征描述符與BOW的通用函數。

主要的通用接口有：

 

1.特征點提取

Ptr<FeatureDetector> FeatureDetector::create(const string& detectorType)

    Ptr<FeatureDetector> FeatureDetector::create(const string& detectorType)
//  "FAST" – FastFeatureDetector
//  "STAR" – StarFeatureDetector
//  "SIFT" – SIFT (nonfree module)//必須使用 initModule_nonfree()初始化
//  "SURF" – SURF (nonfree module)//同上;
//  "ORB" – ORB
//  "MSER" – MSER
//  "GFTT" – GoodFeaturesToTrackDetector
//  "HARRIS" – GoodFeaturesToTrackDetector with Harris detector enabled
//  "Dense" – DenseFeatureDetector
//  "SimpleBlob" – SimpleBlobDetector

 

根據以上接口，測試不同的特征點：

對同一幅圖像進行水平翻轉前后的兩幅圖像檢測特征點檢測結果，

檢測到的特征點的坐標類型為：pt: int / float（與keyPoint的性質有關）

數量分別為num1, num2,

 

 "FAST" – FastFeatureDetector           pt:int (num1:615  num2:618)
 "STAR" – StarFeatureDetector           pt:int (num1:43   num2:42 )
 "SIFT" – SIFT (nonfree module)          pt:float(num1:155  num2:135)            //必須使用 initModule_nonfree()初始化
 "SURF" – SURF (nonfree module)     pt:float(num1:344  num2:342)           //同上; 
 "ORB" – ORB                                        pt:float(num1:496  num2:497)
 "MSER" – MSER                                 pt:float(num1:51   num2:45 )
 "GFTT" – GoodFeaturesToTrackDetector        pt:int (num1:744  num2:771)
 "HARRIS" – GoodFeaturesToTrackDetector with Harris detector enabled         pt:float(num1:162  num2:160)
 "Dense" – DenseFeatureDetector          pt:int (num1:3350 num2:3350)

 

2.特征描述符提取

Ptr<DescriptorExtractor> DescriptorExtractor::create(const string& descriptorExtractorType)

//  Ptr<DescriptorExtractor> DescriptorExtractor::create(const string& descriptorExtractorType)
//  "SIFT" – SIFT
//  "SURF" – SURF
//  "ORB" – ORB
//  "BRIEF" – BriefDescriptorExtractor

 

3.描述符匹配

Ptr<DescriptorMatcher> descriptorMatcher = DescriptorMatcher::create(const string& descriptorMatcherType)

//  descriptorMatcherType – Descriptor matcher type.
//  Now the following matcher types are supported:
//      BruteForce (it uses L2 )
//      BruteForce-L1
//      BruteForce-Hamming
//      BruteForce-Hamming(2)
//      FlannBased
    Ptr<DescriptorMatcher> descriptorMatcher = DescriptorMatcher::create( "BruteForce" );

 

4.class BOWTrainer

class BOWKmeansTrainer::public BOWTrainer:Kmeans算法訓練

BOWKMeansTrainer ::BOWKmeansTrainer(int clusterCount, const TermCriteria& termcrit=TermCriteria(), int attempts=3, int flags=KMEANS_PP_CENTERS)¶

parameter same as Kmeans

 

代碼實現：                                                                                                                    

1.畫特征點。

2.特征點Kmeans聚類，每一種顏色代表一個類別。

 

#include "opencv2/highgui/highgui.hpp"
#include "opencv2/calib3d/calib3d.hpp"
#include "opencv2/imgproc/imgproc.hpp"
#include "opencv2/features2d/features2d.hpp"
#include "opencv2/nonfree/nonfree.hpp"

#include <iostream>

using namespace cv;
using namespace std;

#define ClusterNum 10

void DrawAndMatchKeypoints(const Mat& Img1,const Mat& Img2,const vector<KeyPoint>& Keypoints1,
    const vector<KeyPoint>& Keypoints2,const Mat& Descriptors1,const Mat& Descriptors2)
{
    Mat keyP1,keyP2;
    drawKeypoints(Img1,Keypoints1,keyP1,Scalar::all(-1),0);
    drawKeypoints(Img2,Keypoints2,keyP2,Scalar::all(-1),0);
    putText(keyP1, "drawKeyPoints", cvPoint(10,30), FONT_HERSHEY_SIMPLEX, 1 ,Scalar :: all(-1));
    putText(keyP2, "drawKeyPoints", cvPoint(10,30), FONT_HERSHEY_SIMPLEX, 1 ,Scalar :: all(-1));
    imshow("img1 keyPoints",keyP1);
    imshow("img2 keyPoints",keyP2);

    Ptr<DescriptorMatcher> descriptorMatcher = DescriptorMatcher::create( "BruteForce" );
    vector<DMatch> matches;
    descriptorMatcher->match( Descriptors1, Descriptors2, matches );
    Mat show;
    drawMatches(Img1,Keypoints1,Img2,Keypoints2,matches,show,Scalar::all(-1),CV_RGB(255,255,255),Mat(),4);
    putText(show, "drawMatchKeyPoints", cvPoint(10,30), FONT_HERSHEY_SIMPLEX, 1 ,Scalar :: all(-1));
    imshow("match",show);
}

//測試OpenCV：class BOWTrainer
void BOWKeams(const Mat& img, const vector<KeyPoint>& Keypoints,
    const Mat& Descriptors, Mat& centers)
{
    //BOW的kmeans算法聚類;
    BOWKMeansTrainer bowK(ClusterNum,
        cvTermCriteria (CV_TERMCRIT_EPS + CV_TERMCRIT_ITER, 10, 0.1),3,2);
    centers = bowK.cluster(Descriptors);
    cout<<endl<<"< cluster num: "<<centers.rows<<" >"<<endl;

    Ptr<DescriptorMatcher> descriptorMatcher = DescriptorMatcher::create( "BruteForce" );
    vector<DMatch> matches;
    descriptorMatcher->match(Descriptors,centers,matches);//const Mat& queryDescriptors, const Mat& trainDescriptors第一個參數是待分類節點，第二個參數是聚類中心;
    Mat demoCluster;
    img.copyTo(demoCluster);

    //為每一類keyPoint定義一種顏色
    Scalar color[]={CV_RGB(255,255,255),
     CV_RGB(255,0,0),CV_RGB(0,255,0),CV_RGB(0,0,255),
     CV_RGB(255,255,0),CV_RGB(255,0,255),CV_RGB(0,255,255),
     CV_RGB(123,123,0),CV_RGB(0,123,123),CV_RGB(123,0,123)};


    for (vector<DMatch>::iterator iter=matches.begin();iter!=matches.end();iter++)
    {
        cout<<"< descriptorsIdx:"<<iter->queryIdx<<"  centersIdx:"<<iter->trainIdx
            <<" distincs:"<<iter->distance<<" >"<<endl;
        Point center= Keypoints[iter->queryIdx].pt;
        circle(demoCluster,center,2,color[iter->trainIdx],-1);
    }
    putText(demoCluster, "KeyPoints Clustering: 一種顏色代表一種類型",
        cvPoint(10,30), FONT_HERSHEY_SIMPLEX, 1 ,Scalar :: all(-1));
    imshow("KeyPoints Clusrtering",demoCluster);

}




int main()
{
    cv::initModule_nonfree();//使用SIFT/SURF create之前，必須先initModule_<modulename>();

    cout << "< Creating detector, descriptor extractor and descriptor matcher ...";
    Ptr<FeatureDetector> detector = FeatureDetector::create( "SIFT" );

    Ptr<DescriptorExtractor> descriptorExtractor = DescriptorExtractor::create( "SIFT" );

    Ptr<DescriptorMatcher> descriptorMatcher = DescriptorMatcher::create( "BruteForce" );



    cout << ">" << endl;

    if( detector.empty() || descriptorExtractor.empty() )
    {
        cout << "Can not create detector or descriptor exstractor or descriptor matcher of given types" << endl;
        return -1;
    }
    cout << endl << "< Reading images..." << endl;
    Mat img1 = imread("D:/demo0.jpg");
    Mat img2 = imread("D:/demo1.jpg");
    cout<<endl<<">"<<endl;


    //detect keypoints;
    cout << endl << "< Extracting keypoints from images..." << endl;
    vector<KeyPoint> keypoints1,keypoints2;
    detector->detect( img1, keypoints1 );
    detector->detect( img2, keypoints2 );
    cout <<"img1:"<< keypoints1.size() << " points  img2:" <<keypoints2.size()
        << " points" << endl << ">" << endl;

    //compute descriptors for keypoints;
    cout << "< Computing descriptors for keypoints from images..." << endl;
    Mat descriptors1,descriptors2;
    descriptorExtractor->compute( img1, keypoints1, descriptors1 );
    descriptorExtractor->compute( img2, keypoints2, descriptors2 );

    cout<<endl<<"< Descriptoers Size: "<<descriptors2.size()<<" >"<<endl;
    cout<<endl<<"descriptor's col: "<<descriptors2.cols<<endl
        <<"descriptor's row: "<<descriptors2.rows<<endl;
    cout << ">" << endl;

    //Draw And Match img1,img2 keypoints
    //匹配的過程是對特征點的descriptors進行match;
    DrawAndMatchKeypoints(img1,img2,keypoints1,keypoints2,descriptors1,descriptors2);

    Mat center;
    //對img1提取特征點，並聚類
    //測試OpenCV：class BOWTrainer
    BOWKeams(img1,keypoints1,descriptors1,center);


    waitKey();

}

 

 

通過Qt實現DrawKeypoints：

void Qt_test1::on_DrawKeypoints_clicked()
{
    //initModule_nonfree();
    Ptr<FeatureDetector> detector = FeatureDetector::create( "FAST" );
    vector<KeyPoint> keypoints;
    detector->detect( src, keypoints );

    Mat DrawKeyP;
    drawKeypoints(src,keypoints,DrawKeyP,Scalar::all(-1),0);
    putText(DrawKeyP, "drawKeyPoints", cvPoint(10,30),
        FONT_HERSHEY_SIMPLEX, 0.5 ,Scalar :: all(255));
    cvtColor(DrawKeyP, image, CV_RGB2RGBA);
    QImage img = QImage((const unsigned char*)(image.data),
        image.cols, image.rows, QImage::Format_RGB32);
    QLabel *label = new QLabel(this);
    label->move(50, 50);//圖像在窗口中所處的位置;
    label->setPixmap(QPixmap::fromImage(img));
    label->resize(label->pixmap()->size());
    label->show();
}

由於initModule_nonfree()總是出錯，無法對SIFT與SURF特征點提取，

而且無法實現聚類因為運行/BOW的kmeans算法聚類：BOWKMeansTrainer bowK(ClusterNum, cvTermCriteria (CV_TERMCRIT_EPS + CV_TERMCRIT_ITER, 10, 0.1),3,2);總是出錯，不知道咋解決~~~~~(>_<)~~~~ 需要繼續學習

 

from: http://blog.csdn.net/yangtrees/article/details/8456237