基于小波变换的声纳图像边缘特征检测研究  

Research of Edge Detection Based on Wavelet Transform for Sonar Image

作　　者：贾伟广[1] 李希玲[1] 

机构地区：[1]国家海洋标准计量中心,天津300112

出　　处：《海洋技术》2006年第3期103-105,共3页Ocean Technology

摘　　要：声纳图像的边缘特征检测是其目标识别技术的重要技术基础。声纳图像背景复杂、噪声污染严重,而传统的边缘检测方法对图像噪声非常敏感,所以针对这一特点,利用小波变换易于消除噪声、运算方便的数学特征,提出了一种基于小波变换的声纳图像边缘特征检测算法。由计算机仿真结果可以得到,与传统的边缘检测算法相比,此算法在有效地抑制噪声的同时,还可以得到较高的边缘定位精度,能够很好地检测到原始声纳图像的边缘。Edge detection of sonar image is the important technique basis of image interpretation. Sonar image＇s background is complicate, noise is serious. At the same time, traditional edge detection is quite sensitive to the image noise, So aiming at this characteristics, making use of the mathematical characteristics of wavelet trans formation＇s easily to dissolving the noise and conveniently to calculating, bringing up a sonar image edge detection calculating method based on wavelet transform. Con trary to the traditional edge detection, computer simulation results show that it is effectively to remove the noise and extract the better location accuracy of image＇s edge by the method in this paper, so it can detect the original sonar image efficiently.

关 键 词：图像 边缘检测 小波变换 图像处理 声纳 

分 类 号：TN911.6[电子电信—通信与信息系统]