基于SURF算法和改进RANSAC算法的无人机影像匹配

UAV image matching based on SURF algorithm and improved RANSAC algorithm

影像匹配是诸多遥感影像处理和影像分析的一个关键环节,结合加速鲁棒性特征(SURF)算法和随机采样一致性(RANSAC)算法对影像进行处理,得到特征稳定、匹配点可靠的配准影像。首先提取影像的SURF特征,利用特征点的欧式距离比来完成影像之间的粗匹配;然后使用RANSAC算法对粗匹配点进行筛选;最后计算出图像间的变换矩阵,完成匹配。文中选择某城郊地区的无人机航拍影像,结合SURF算法,并改进RANSAC算法来对影像进行处理,实现影像的匹配,验证文中方法的可行性。

Image match is an important step in remote sensing image process and image analysis. Combined Speed-Up Robust Features （SURF） and Random Sample Consensus （RANSAC） algorithms for image process, it can get registration image with stable feature and reliable match point. First, the SURF feature is extracted to achieve the initial matched with Euclidean distance between images; then, filtering matched points are chozen with RANSAC algorithm; at last, the transformation matrix is calculated between the images in order to complete match. This paper processes the UAV images of suburban area with SURF algorithms and improved RANSAC algorithms and realizes the image matched, which proves the feasibility of this method.