残冰检测中飞机图像特征点精确提取算法

摘要

传统角点检测算法无法精确提取感兴趣的特征点.为此,提出一种新的特征点提取算法.通过曲率角点检测算法提取飞机图像的特征点,并作为待匹配图像,采用像素相关性匹配算法进行特征点提取,利用飞机机身固定结构约束匹配算法去除误匹配对.实验结果表明,该算法具有较好的适应性,能够精确提取特征点.%Traditional corner detection algorithm can not extract the interested feature point. In order to solve this problem, this paper proposes a new template matching algorithm. In this paper, the special points of airplane image are gotten by curvature corner detection algorithm as the matching image. Template matching algorithm is used to extract the best match points. The features of the image with different illumination conditions and shooting angles are extracted and matched. Experimental results show that the algorithm has better adaptability, and improve the accuracy of matching.