基于SIFT和Affinity Propagation的遥感图像配准算法

随着传感器和光学影像测量等各种技术的快速发展,航空遥感技术已经在电力巡检、森林防火、地理测绘等领域中发挥着越来越重要的作用,而图像配准作为遥感图像的预处理步骤为图像融合等后续处理提供了参考和依据,目前已经成为遥感图像处理领域的研究热点。文中提出了一种基于尺度不变特征变换（Scale-invariant Feature Transform,SIFT）和仿射传播聚类（Affinity Propagation,AP）的图像配准算法。该算法与原有算法相比优势在于无需预先设定参数,并且实验仿真表明该算法能有效地对多源图像进行高精度的配准,与随机抽样一致算法（Random Sample Consensus,RANSAC）相比提高了正确匹配点的数目。

Remote sensing image registration algorithm based on SIFT and affinity propagation

With the rapid development of sensor, optical image measuring and other various technologies, remote sensing data processing has been in power patrol, forest fire prevention, geographic mapping and other fields, now playing an increasingly important role. Image registration as a preprocessing step for the remote sensing image fusion and other subsequent processing provides a reference and basis, which has become the hot research field of remote sensing images. This paper presents an image registration algorithms based on scale-invariant feature transform （SIFT） and affinity propagation （AP）. The advantage of this algorithm compared with the original algorithm is without preset parameters. The simulation results show that the algorithm is capable of multi-source image registration for higher accuracy and improving robustness compared with random sample consensus （RANSAC）.