摘要
基于特征点的匹配是图像匹配的关键方法。提出一种可以实现大量特征点匹配的算法。该算法实现了图像行校正,改进了Harris角点检测算法,提出了一种新的角点指示器,它可以检测出足够多的特征点进行特征点匹配;然后使用sift特征描述运算符描述每个特征点;最后,使用最近邻居距离比算法完成特征点匹配。实验结果表明,提出的算法可以实现大量特征点匹配,对图像匹配算法具有一定的参考意义。
Matching based on feature points is a key method of image matching.This paper proposes an algorithm that can achieve a large number of feature point matching.The algorithm realizes the image row rectification,improves the Harris corner detection algorithm,and proposes a new corner indicator,which can detect enough feature points for feature point matching.Then use the sift feature description operator to describe each feature point.Finally,this paper uses the nearest neighbor distance ratio algorithm to complete feature point matching.Experimental results show that the algorithm proposed in this paper can achieve a large number of feature points matching,which has certain reference significance for image matching algorithms.
出处
《工业控制计算机》
2021年第6期27-29,共3页
Industrial Control Computer
