基于SAR图像的海洋溢油分割方法研究

目前,海洋溢油污染在发生频率、分布范围以及所带来的危害程度上均居海洋污染首位。它对沿海城市生态、环境、经济的可持续发展造成严重破坏,因此越来越受到全社会的广泛关注。从众多海洋溢油分割方法中选取边缘检测算子和种子填充算法对SAR图像进行分割。首先运用欧空局的BEST软件进行数据转换以及图像预处理,然后使用ENVI软件对图像进行滤波处理。再用ERDAS软件进行灰度均衡和分割算法的实施,并针对算法的一些缺陷,进一步加入纹理分析使图像中的油膜与类油膜以及海水与海浪等更好的区分开。最后对不同的图像效果进行总结和比较。

Research on split methods of ocean oil spill in SAR image

At present ,the oil spill in the ocean stays in the first place of marine pollution at the frequency of occurrence ,distribution or damage . It raises all the society ’ s concern , because it has made serious damages to the coastal city’s ecological ,environment and the sustainable development of economy .This paper chooses the edge detection operator and seed filling from the numerous methods of oil spill segmentation ,which are used for segmentation of SAR image .First ,the special software named BEST which is given by the ESA(European Space Agency) is used to make a data conversion and do the image preprocessing .Then the software named ENVI is used to do the filtering processing .At last ,use the software named ERDAS is used to carry out the gray balance and segmentation . After that , some disadvantages of those methods are found .For having an open and shut result ,this paper adds a texture analysis which can divide the oil slick ,the alike oil slick and the waves in the sea better .In the end ,it makes some conclusions and comparisons among different image effects .