【摘要】 角点是图像目标的重要的局部特征,角点检测是低层次图像处理的一个重要方法。角点检测在光流计算、运动估计、形状分析、相机标定和3D重建、视觉的定位和测量等方面都有重要的应用;根据实现方法不同可将角点检测算法分为基于模板的方法、基于边缘的方法、基于亮度变化的方法,并将现有的角点检测方法作了较为详细的分析、比较。更多还原

【Abstract】 Corner is a significant local feature of images. Corner detection is an important method in low-level image processing. Corner detection have been used in optical flow computation, motion estimate, object tracking, shape analysis, camera calibration, 3D reconstruction, location and measure of machine vision. In this paper, the corner detection algorithms are divided into three classes: template-based corner detection, boundary-based corner detection,and intensity change-based corner detection according to different realizations, as well as description of these algorithms.更多还原