基于形状描述算子的仿射点匹配算法

摘要

针对仿射点匹配问题,首先通过建立点模式之间仿射映射关系,运用虚拟对应点估计待匹配点集的仿射参量,然后运用形状描述算子进行点集间匹配关系的求解,通过上述步骤的反复迭代最终求得点匹配问题的最优解集.算法较好地解决了仿射变换下的点集匹配问题,即使在较强的噪声和出格点干扰情况下,仍能获得较高的正确匹配率.实验证明了算法的有效性和鲁棒性.%A new algorithm is proposed to solve the point matching problem. We suppose that the point sets are in the relationship of affine transformation, and use the virtual corresponding points to estimate the affine parameters for searching corresponds. Based on this estimation, we utilize shape context to find the correspondences between the two point sets. The algorithm will iterate the two steps above many times for the final exact affine parameters and corresponds. Experiments on both synthetic point sets with varying degrees of affine deformation, noise and outliers, and on real image point-sets fully demonstrate the robustness and effectiveness of the algorithm.