海洋遥感图像中基于纹理的海面油膜识别

基于海面纹理特征,利用Borda算法进行了航空遥感图像中油膜和非油膜的识别研究。首先,对航空遥感图像进行纹理特征分析和提取:二值化处理并且提取谱相关性、方向粒度和条状粒度的纹理特征值;然后,建立海面油膜数据库,包括原始图像、特征向量,以及图像和纹理特征间的映射表;最后引入Borda算法,根据被测图像的特征来识别其是否含有油膜。实验结果表明,本模型能有效的识别油膜,能为大范围海面溢油检测奠定研究基础。

Recognition of Marine Oil Spill in Marine Sensing Image Based on Ocean Water Surface

A way of recognizing oil spill ocean surface is proposed by analyzing remote sensing image and using the Borda count method. First, surface patches of images of visible remote sensing are binarized and features including spectral coherence, directionality fineness and strip transect are extracted; Second, a database including raw data, feature data and the map of the feature data to raw image data are built; Finally, suspicious patches are test by using Borda count method to determine whether it is a real oil spill patch. The efficiency of the algorithm is demonstrated by a preliminary experiment with 278 training patches and 112 testing patches. This paper reports initial experiments on our oil spill reorganization and lays a foundation for larger open sea oil spill detection.