图像拼接源码分析向：BestOf2NearestMatcher opencv 代码解析

//

BestOf2NearestMatcher::BestOf2NearestMatcher(bool try_use_gpu, float match_conf, int num_matches_thresh1, int num_matches_thresh2)
{
    CV_UNUSED(try_use_gpu);
    // 验证 GPU 是否可用
#ifdef HAVE_OPENCV_CUDAFEATURES2D
    if (try_use_gpu && getCudaEnabledDeviceCount() > 0)
    {
        impl_ = makePtr<GpuMatcher>(match_conf);
    }
    else
#endif
    {
        impl_ = makePtr<CpuMatcher>(match_conf);
    }
    // 设置两个门限值
    is_thread_safe_ = impl_->isThreadSafe();
    num_matches_thresh1_ = num_matches_thresh1;
    num_matches_thresh2_ = num_matches_thresh2;
}


void BestOf2NearestMatcher::match(const ImageFeatures &features1, const ImageFeatures &features2,
                                  MatchesInfo &matches_info)
{
    CV_INSTRUMENT_REGION();
    // 输入：图片1特征，图片2特征，图片之间的pairwise矩阵
    (*impl_)(features1, features2, matches_info);

    // Check if it makes sense to find homography
    // 假如现有匹配数量小于第一个门限值，则直接终止
    if (matches_info.matches.size() < static_cast<size_t>(num_matches_thresh1_))
        return;

    // Construct point-point correspondences for homography estimation
    // 为后续单应性矩阵的估计建立点和点之间的对应关系
    Mat src_points(1, static_cast<int>(matches_info.matches.size()), CV_32FC2);
    Mat dst_points(1, static_cast<int>(matches_info.matches.size()), CV_32FC2);
    for (size_t i = 0; i < matches_info.matches.size(); ++i)
    {
        const DMatch& m = matches_info.matches[i];

        Point2f p = features1.keypoints[m.queryIdx].pt;
        p.x -= features1.img_size.width * 0.5f;
        p.y -= features1.img_size.height * 0.5f;
        src_points.at<Point2f>(0, static_cast<int>(i)) = p;

        p = features2.keypoints[m.trainIdx].pt;
        p.x -= features2.img_size.width * 0.5f;
        p.y -= features2.img_size.height * 0.5f;
        dst_points.at<Point2f>(0, static_cast<int>(i)) = p;
    }

    // Find pair-wise motion
    // 计算单应性矩阵，这里面有点复杂，暂时没看明白
    matches_info.H = findHomography(src_points, dst_points, matches_info.inliers_mask, RANSAC);
    if (matches_info.H.empty() || std::abs(determinant(matches_info.H)) < std::numeric_limits<double>::epsilon())
        return;

    // Find number of inliers
    // 计算内点数量
    matches_info.num_inliers = 0;
    for (size_t i = 0; i < matches_info.inliers_mask.size(); ++i)
        if (matches_info.inliers_mask[i])
            matches_info.num_inliers++;

    // These coeffs are from paper M. Brown and D. Lowe. "Automatic Panoramic Image Stitching
    // using Invariant Features"
    matches_info.confidence = matches_info.num_inliers / (8 + 0.3 * matches_info.matches.size());

    // Set zero confidence to remove matches between too close images, as they don't provide
    // additional information anyway. The threshold was set experimentally.
    // 假如图片相似程度很大，那么将置信度设置为 0 即可
    matches_info.confidence = matches_info.confidence > 3. ? 0. : matches_info.confidence;

    // Check if we should try to refine motion
    if (matches_info.num_inliers < num_matches_thresh2_)
        return;

    // Construct point-point correspondences for inliers only
    // 在得到内点之后，只为内点建立点与点之间的对应关系
    src_points.create(1, matches_info.num_inliers, CV_32FC2);
    dst_points.create(1, matches_info.num_inliers, CV_32FC2);
    int inlier_idx = 0;
    for (size_t i = 0; i < matches_info.matches.size(); ++i)
    {
        if (!matches_info.inliers_mask[i])
            continue;

        const DMatch& m = matches_info.matches[i];

        Point2f p = features1.keypoints[m.queryIdx].pt;
        p.x -= features1.img_size.width * 0.5f;
        p.y -= features1.img_size.height * 0.5f;
        src_points.at<Point2f>(0, inlier_idx) = p;

        p = features2.keypoints[m.trainIdx].pt;
        p.x -= features2.img_size.width * 0.5f;
        p.y -= features2.img_size.height * 0.5f;
        dst_points.at<Point2f>(0, inlier_idx) = p;

        inlier_idx++;
    }

    // Rerun motion estimation on inliers only
    // 只计算内点的单应性矩阵
    matches_info.H = findHomography(src_points, dst_points, RANSAC);
}

待补充：计算单应性矩阵部分的代码