利用小波边缘增强的可靠性匹配方法

Reliable matching method based on wavelet edge enhancement

为提高景像匹配的可靠性,提出了一种基于小波边缘增强的可靠性匹配方法。采用四阶中心B样条小波对实时图和基准图进行小波分解和边缘增强,以提取可靠的边缘特征。在粗尺度上基于小波边缘增强图进行相关匹配,选择相似度曲面上前5个峰值点作为候选匹配点,保证正确匹配点可以包含在候选匹配点中。对实时图进行旋转校正,利用候选匹配点处的局部灰度特征确定边缘提取的双阈值,应用形态学连接算子来获得二值边缘图,再进行相关匹配,筛选出正确匹配点。在细尺度上获得精确的定位点。应用该方法进行匹配实验,其匹配概率比基于小波二值边缘提取的匹配算法提高了6.95%。

In order to improve the reliability of scene matching, a reliable matching method based on wavelet edge enhancement is presented. The real-time image and the referenced image are decomposed to wavelet pyramids and edge-enhanced using the fourth order center B spline wavelet to extract reliable edge features. The matching is done based on wavelet edge enhanced images in coarse scale. The first five peak values are selected as the candidate matching points to ensure that the true matching point are included. Real-time image is corrected in a way of rotation. Double thresholds are determined on the basis of local gray features around the candidate matching points, and the binary edge images are obtained with morphological linking operator. Correlation matching is implemented again and the maximal peak value is chosen as the true matching point. Then precise matching point is obtained in fine scale.Matching experiment is completed with the method. The results show that the matching probability of the method increases by 6.95% compared with the method based on wavelet binary edge extraction.