摘要
研究了一种基于尺度空间理论的Harris角点检测方法。建立Harris函数的尺度空间表示,检测每个尺度水平上的极值,利用迭代算法验证每个尺度水平上LoG算子是否获得最大值,从而得到特征角点的位置及其尺度。该方法在保持Harris角点不受光照条件及摄像机姿态变化影响的同时,还能检测出多尺度下的特征点。通过实验验证该方法具有尺度不变特性,适用于尺度变化较大的视觉系统。
An improved method of Harris corner based on the theory of scale-space was described, and a scale-space representation of Harris corner by which local maximum points were detected at each scale level was established. The extrema over scale of the Laplacian of Gaussian (LOG) which was used to select the scale of interest points were applied. For each point, an iterative algorithm can be used to detect the location and the scale of interest points simultaneously. This method not only maintains the advantages of Harris corner which is invariant to the changes of intensity and camera pose but also can be used in multi-scale. It is proved to be scale invariant by experiments and can be applied to the vision system with significant scale changes.
出处
《中南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2005年第5期751-754,共4页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(60234030)