Opencv3中ORB算法的使用

相信很多小伙伴在使用ORB算法的時候，一般會從網上搜一些代碼作為參考，那么問題來了：在好多ORB程序中都會這么寫：

    ORB orb;

如果你使用的是Opencv3的版本，編譯器就會報錯：ORB是一個純虛類，無法進行實例化。但在opencv2的版本中可以正常使用。這是為什么呢？

於是乎就在opencv3官方的Documents中尋找答案，ORB屬於features2d模塊中。在它的文檔中終於發現了原因

Public Member Functions:
virtual int     getEdgeThreshold () const =0
virtual int     getFastThreshold () const =0
virtual int     getFirstLevel () const =0
virtual int     getMaxFeatures () const =0
virtual int     getNLevels () const =0
virtual int     getPatchSize () const =0
virtual double     getScaleFactor () const =0
virtual int     getScoreType () const =0
virtual int     getWTA_K () const =0
virtual void     setEdgeThreshold (int edgeThreshold)=0
virtual void     setFastThreshold (int fastThreshold)=0
virtual void     setFirstLevel (int firstLevel)=0
virtual void     setMaxFeatures (int maxFeatures)=0
virtual void     setNLevels (int nlevels)=0
virtual void     setPatchSize (int patchSize)=0
virtual void     setScaleFactor (double scaleFactor)=0
virtual void     setScoreType (int scoreType)=0
virtual void     setWTA_K (int wta_k)=0

Static Public Member Functions：
static Ptr< ORB >     create (int nfeatures=500, float scaleFactor=1.2f, int nlevels=8, int edgeThreshold=31, int firstLevel=0, int WTA_K=2, int scoreType=ORB::HARRI
S_SCORE, int patchSize=31, int fastThreshold=20)

#include<iostream>
#include<vector>

#include<opencv2\core\core.hpp>
#include<opencv2\features2d\features2d.hpp>
#include<opencv2\highgui\highgui.hpp>

using namespace std;
using namespace cv;


int main()
{
    //讀取圖片
    Mat rgbd1 = imread("自己的圖片Path");
    Mat rgbd2 = imread("自己的圖片Path");
    //imshow("rgbd1", depth2);
    //waitKey(0);
    Ptr<ORB> orb = ORB::create();
    vector<KeyPoint> Keypoints1,Keypoints2;
    Mat descriptors1,descriptors2;
    orb->detectAndCompute(rgbd1, Mat(), Keypoints1, descriptors1);
    orb->detectAndCompute(rgbd2, Mat(), Keypoints2, descriptors2);
    
    //cout << "Key points of image" << Keypoints.size() << endl;
    
    //可視化，顯示關鍵點
    Mat ShowKeypoints1, ShowKeypoints2;
    drawKeypoints(rgbd1,Keypoints1,ShowKeypoints1);
    drawKeypoints(rgbd2, Keypoints2, ShowKeypoints2);
    imshow("Keypoints1", ShowKeypoints1);
    imshow("Keypoints2", ShowKeypoints2);
    waitKey(0);

    //Matching
    vector<DMatch> matches;
    Ptr<DescriptorMatcher> matcher =DescriptorMatcher::create("BruteForce");
    matcher->match(descriptors1, descriptors2, matches);
    cout << "find out total " << matches.size() << " matches" << endl;

    //可視化
    Mat ShowMatches;
    drawMatches(rgbd1,Keypoints1,rgbd2,Keypoints2,matches,ShowMatches);
    imshow("matches", ShowMatches);
    waitKey(0);


    return 0;
}

這是一個對兩幅圖片進行ORB提取特征，並進行Match的程序。可以看到對於orb的使用程序變成了

Ptr<ORB> orb = ORB::create();

細心的讀者應該已經發現Match類也變成了一個純虛類使用方法與ORB類似：

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

static Ptr<ORB> cv::ORB::create    (    
        int     nfeatures = 500,
        float     scaleFactor = 1.2f,
        int     nlevels = 8,
        int     edgeThreshold = 31,
        int     firstLevel = 0,
        int     WTA_K = 2,
        int     scoreType = ORB::HARRIS_SCORE,
        int     patchSize = 31,
        int     fastThreshold = 20 
)        

下面是關於具體參數的解釋有點多，就附上網址吧：

http://docs.opencv.org/3.2.0/db/d95/classcv_1_1ORB.html#adc371099dc902a9674bd98936e79739c

點贊 12
————————————————
版權聲明：本文為CSDN博主「bingoplus」的原創文章，遵循 CC 4.0 BY-SA 版權協議，轉載請附上原文出處鏈接及本聲明。
原文鏈接：https://blog.csdn.net/bingoplus/article/details/60133565