基于数学形态学的侧扫声呐图像轮廓自动提取

Automatic extraction of the side-scan sonar imagery outlines based on mathematical morphology

侧扫声呐图像特征自动提取的难点在于特征地貌边缘检测较困难,依据图像灰度突变检测得到的边缘比较粗糙、不连续,而且有断口和小洞。本文在对图像进行预处理和阈值化的基础上,采用数学形态学方法对图像进行处理,即用具有一定形态的结构元素去量度和提取图像中的对应形状,得到连续化、粗化、圆滑的特征区域边缘填充目标内部阴影且消除背景噪声。基于数学形态学的侧扫声呐图像特征自动提取的主要步骤为:首先对侧扫声呐图像进行预处理,然后进行灰度阈值化,接着采用数学形态学方法进行处理,最后对处理后的图像进行边缘检测,提取出特征地貌边缘。实验表明,采用数学形态学方法进行处理后,错断、离散的海底目标物变得连续,背景噪声大大减少,自动提取结果准确可靠。

Automatic extraction of the side-scan sonar imagery outlines is difficult.The results extracted by edge detection based on sharp gray-scale gradient of image are discontinuous and rough,and also have gaps and holes edge detection.After preprocessing the side-scan sonar image and thresholding,some processings are carried out to take the smooth and continuous rims of the geological objectives,and to eliminate the background noises,by measuring and extracting the corresponding shape from the image with a certain form of structural element according to the basic idea of mathematical morphology.The algorithm of feature extraction for the side-scan sonar imagery based on mathematical morphology is as follows：firstly,preprocess the image and do thresholding it;then process the image by mathematical morphology;finally obtained the edges of the geological objectives by edge-detection technology.The numerical experiments show that this method leads to smooth and continuous and accurate detection,meanwhile,greatly reduced background noise.