摘要
针对局部特征匹配面临的实时性和鲁棒性难以兼顾的问题,提出了一种基于Harris算子的快速图像匹配算法。通过分析Harris算子的基本原理,提出了用特征检测的中间数据描述局部特征,并建立了一种基于Harris自相关矩阵之迹的低维特征描述子,在保持算法鲁棒性的同时有效减少了算法的计算量,最后用特征描述子之间的绝对值距离作为相似性度量匹配特征点以降低计算复杂度。实验结果表明,本算法不仅对图像尺度缩放、旋转、模糊、亮度变化和较小视角变化保持不变,而且匹配速度较快。
In order to solve the problem that it is difficult to balance the real-time perform- ance and robustness in image matching using local feature, a fast image matching algorithm based on Harris operator is presented. By analyzing the basic theory of Harris operator, it is proposed that local feature can be described with the temporary data of feature detecting. Furthermore, a low dimensions feature descriptor is built with the trace of Harris autocorre-lation matrix, which can both preserve the robustness and reduce the computation effectively. Finally, the absolute distance between descriptors is used as a similarity measurement to match feature points for decreasing the computing complexity. Simulation results indicate
出处
《武汉大学学报（信息科学版）》
EI
CSCD
北大核心
2012年第4期406-409,414,共5页
Geomatics and Information Science of Wuhan University
基金
国家863计划资助项目
