一种改进的自适应FAST角点检测算法

An improved adaptive FAST corner detection algorithm

在FAST角点检测算法中存在阈值设置相对固定的问题,该问题会导致对同一张图片中不同区域的检测效果较差。针对上述问题,提出一种基于OTSU的区域自适应FAST角点检测算法,以此来解决区域检测效果较差的问题。该算法分为两个部分:首先,对图像进行灰度处理,并对图像进行M×N的区域网状分割,再对每一个网格区域内的图像部分使用OTSU算法得到该区域的阈值;其次,采用FAST角点检测算法检测图像中每个区域的角点,并将上一步得到的区域阈值结果导入FAST角点检测算法中。最后进行了实验测试,将该算法与常见的角点检测算法进行对比,实验结果表明,经过区域自适应算法改进后的FAST角点检测算法提高了检测效率和检测精度,漏检和错检的情况大幅度减少。该改进算法能够提高图像区域的检测效果。

The relatively fixed threshold setting in the FAST corner detection algorithm makes the FAST corner detection algorithm have a poor detection effect in different areas in the same image,so a region adaptive FAST corner detection algorithm based on OTSU is proposed to solve the problem of poor region detection effect.The algorithm is divided into two parts.The image is dealt with by gray-scale processing and segmented by M×N area mesh,and then the OTSU algorithm is used for the image parts in each mesh area to get the threshold of the area.The FAST corner detection algorithm is used to detect the corner points of each region in the image,and the region threshold values obtained in the previous step are imported into the FAST corner detection algorithm.An experimental test was carried out to compare the algorithm proposed in this paper with the common corner detection algorithms.The experimental results show that the FAST corner detection algorithm improved by the region adaptive algorithm can improve the detection efficiency and accuracy,and the missed detection and false detection can be greatly reduced.The improved algorithm can improve the detection effect of the image area.