舰船尾流图像的边缘分割算法研究

Research on edge segmentation method of ship wake image

舰船尾流是指舰船航行过程中由于螺旋桨空化、水与舰船侧壁摩擦等原因,在舰船尾部形成的气泡群。利用舰船尾流多波束成像技术,可以探测到尾流的声散射强度数据。为了更好地研究尾流的特性,对原始声呐数据进行图像可视化处理。由于处理后的原始尾流图像掺杂着噪声,尤其当海面反射较强时,造成部分尾流图像与海面的反射图像完全混合在一起,因此需要寻找合适的算法将尾流图像的边缘提取出来。文章研究了微分算子、数学形态学算法、分形算法等边缘分割算法在尾流图像处理中的应用,并通过比较总结了每种算法的优缺点。

Ship wake is the bubble group formed at the tail of the ship due to the cavitation of the propellers and the friction between the water and the side wall of the ship in the course of navigation.The multi-beam imaging technology for ship wake can be used to detect the sound scattering intensity of the wake.In order to better study the characteristics of the wake,the image visualization processing of the original sonar data is adopted.However,the original wake image after processing is doped with noise,and especially when the sea surface reflection is strong,part of the wake image could be completely mixed with the reflected image of the sea surface.Therefore,it is necessary to find a suitable algorithm to extract the edge of the wake image.The applications of several edge segmentation algorithms,such as the differential operators,the mathematical morphology-based and fractal-based algorithms are studied in this paper.The advantages and disadvantages of different algorithms are summarized by comparison and verification.