基于共视目标的侧扫声纳条带图像拼接方法研究

A Mosaic Method Based on Corresponding Features for Side-Scan Sonar Images

针对侧扫声纳受海洋复杂环境影响大的特点,本文提出了一种基于相邻条带共视目标的方法,并结合声纳原始记录文件中航向、姿态等信息,对侧扫声纳位置及航向等进行修正,实现了相邻条带拼接,对拼接后的图像分别采用加权平均和小波变换的方法进行融合处理,并使用信息熵对图像质量进行了评价。试验验证了该方法的有效性和可行性。

According to the characteristics of the side - scan sonar system which affected by the complex ocean environment, this pa- per proposes a method based on adjacent bands which have the common targets. This method can correctly rectify the position and heading of side - scan sonar with sonar original record file such as course, finally implement the adjacent of the side - scan sonar stripe images. Then two image fusion methods, the weighted average and the wavelet transform, are used to fulfill the fusion of the side - scan sonar images. The result demonstrates that the proposed method in this paper is feasible and effective.