opencv3中SIFT图像拼接代码详解(findHomography)_the input arrays should be 2d or 3d point sets in

sift特征检测，返回内点个数和透视变换矩阵：

int siftmatch(Mat img1, Mat img2, Mat* H)
{
	//一、检测特征点
	Ptr<xfeatures2d::SIFT>feature = xfeatures2d::SIFT::create();//创建SIFT特征类
	vector<KeyPoint>keypoints1, keypoints2;
	feature->detect(img1, keypoints1);//检测特征点，检测信息保存在keypoint中
	feature->detect(img2, keypoints2);
 
	//二、计算描述矩阵并匹配
 
	Mat description1, description2;//初始化描述矩阵
 
	feature->compute(img1, keypoints1, description1);//计算描述矩阵，保存在description中
	feature->compute(img2, keypoints2, description2);
 
	vector<DMatch>matches; //匹配矩阵
	BFMatcher matcher;
	matcher.match(description1, description2, matches);
 
	//Mat image_match2;
	//drawMatches(img1, keypoints1, img2, keypoints2, matches, image_match2);
	//imshow("匹配后的图片2", image_match2);
 
	//三、采用findHomography函数进行RANSAC筛选
 
	std::vector<Point2f>obj, scene;
 
	for (size_t i = 0; i<matches.size(); i++)
	{
		//-- Get the keypoints from the good matches 
		obj.push_back(keypoints1[matches[i].queryIdx].pt);
		scene.push_back(keypoints2[matches[i].trainIdx].pt);
	}
 
	vector<uchar>inliersMask(obj.size());
	*H = findHomography(scene, obj, CV_FM_RANSAC, 3.0, inliersMask, 100);
 
	vector<DMatch>inliers;
	for (size_t i = 0; i<inliersMask.size(); i++) {
		if (inliersMask[i])
			inliers.push_back(matches[i]);
	}
	matches.swap(inliers);
 
	//画线
	//cout << "内点数为：" << matches.size() << endl;
	//Mat image_match2 = Mat(img1.rows, img1.cols + img2.cols, CV_8UC1, Scalar(0));
	//for (int i = 0; i < img1.rows; i++)
	//{
	//	const uchar* img1ptr = img1.ptr<uchar>(i);
	//	const uchar* img2ptr = img1.ptr<uchar>(i);
	//	uchar* outdata = image_match2.ptr<uchar>(i);
	//	for (int j = 0; j < img1.cols; j++)//将图1和图2拼在一起
	//	{
	//		outdata[j] = img1ptr[j];
	//		outdata[j + img1.cols] = img2ptr[j];
	//	}	
	//}
	//Point pt1, pt2;//连线的两个端点  
	//for (size_t i = 0; i<matches.size(); i++)
	//{
	//	//-- Get the keypoints from the good matches 
	//	pt1 = keypoints1[matches[i].queryIdx].pt;
	//	pt2 = keypoints2[matches[i].trainIdx].pt;
	//	pt2.x += img1.cols; 
	//	line(image_match2, pt1, pt2, CV_RGB(255, 0, 255));
	//}
	//imshow("匹配后的图片2", image_match2);
 
	return matches.size();
}

findHomography函数：

_points1和_points2为初步计算的匹配点，用来通过RANSAC算法筛选
method为筛选方法，这里为CV_FM_RANSAC

cv::Mat cv::findHomography( InputArray _points1, InputArray _points2,
                            int method, double ransacReprojThreshold, OutputArray _mask,
                            const int maxIters, const double confidence)
{
    CV_INSTRUMENT_REGION()//应该是OpenCV相关算法表现性能测试框架，测量函数执行时间，在函数内部追踪函数执行状况
 
    const double defaultRANSACReprojThreshold = 3;//默认拒绝阈值
    bool result = false;
 
    Mat points1 = _points1.getMat(), points2 = _points2.getMat();//用矩阵保存
    Mat src, dst, H, tempMask;
    int npoints = -1;
 
    for( int i = 1; i <= 2; i++ )
    {
        Mat& p = i == 1 ? points1 : points2;
        Mat& m = i == 1 ? src : dst;
        npoints = p.checkVector(2, -1, false);
        if( npoints < 0 )
        {
            npoints = p.checkVector(3, -1, false);
            if( npoints < 0 )
                CV_Error(Error::StsBadArg, "The input arrays should be 2D or 3D point sets");
            if( npoints == 0 )
                return Mat();
            convertPointsFromHomogeneous(p, p);
        }
        p.reshape(2, npoints).convertTo(m, CV_32F);
    }
 
    CV_Assert( src.checkVector(2) == dst.checkVector(2) );
 
    if( ransacReprojThreshold <= 0 )
        ransacReprojThreshold = defaultRANSACReprojThreshold;
 
    Ptr<PointSetRegistrator::Callback> cb = makePtr<HomographyEstimatorCallback>();
 
    if( method == 0 || npoints == 4 )
    {
        tempMask = Mat::ones(npoints, 1, CV_8U);
        result = cb->runKernel(src, dst, H) > 0;
    }
    else if( method == RANSAC )
        result = createRANSACPointSetRegistrator(cb, 4, ransacReprojThreshold, confidence, maxIters)->run(src, dst, H, tempMask);
    else if( method == LMEDS )
        result = createLMeDSPointSetRegistrator(cb, 4, confidence, maxIters)->run(src, dst, H, tempMask);
    else if( method == RHO )
        result = createAndRunRHORegistrator(confidence, maxIters, ransacReprojThreshold, npoints, src, dst, H, tempMask);
    else
        CV_Error(Error::StsBadArg, "Unknown estimation method");
 
    if( result && npoints > 4 && method != RHO)
    {
        compressElems( src.ptr<Point2f>(), tempMask.ptr<uchar>(), 1, npoints );
        npoints = compressElems( dst.ptr<Point2f>(), tempMask.ptr<uchar>(), 1, npoints );
        if( npoints > 0 )
        {
            Mat src1 = src.rowRange(0, npoints);
            Mat dst1 = dst.rowRange(0, npoints);
            src = src1;
            dst = dst1;
            if( method == RANSAC || method == LMEDS )
                cb->runKernel( src, dst, H );
            Mat H8(8, 1, CV_64F, H.ptr<double>());
            createLMSolver(makePtr<HomographyRefineCallback>(src, dst), 10)->run(H8);
        }
    }
 
    if( result )
    {
        if( _mask.needed() )
            tempMask.copyTo(_mask);
    }
    else
    {
        H.release();
        if(_mask.needed() ) {
            tempMask = Mat::zeros(npoints >= 0 ? npoints : 0, 1, CV_8U);
            tempMask.copyTo(_mask);
        }
    }
 
    return H;
}