遥感监测墨西哥湾溢油目标识别算法

Extraction algorithm of oil spill cases in Gulf of Mexico with RS data

为了及时监测溢油事故相关信息,防止污染扩张,该文利用2010年4月29日、5月17日及5月24日3个时相的中分辨率MODIS数据,以2010年发生的墨西哥湾溢油事故为例进行目标提取与识别。提取算法考虑到油膜和背景海水易混分的现象,采用了辅加纹理特征量的光谱角匹配算法进行油膜提取;同时,对比了常用的最小距离和支持向量机分类方法;并基于混淆矩阵对结果进行精度验证。结果显示,改进的光谱角匹配算法可以更准确地提取海面目标,精度高达90%以上,能够较好地监测出大面积溢油,可以用于海面溢油灾害的动态监测。

This paper extracted oil spill using three images on April,29,May,17 and May,24 2010,which happened in Gulf of Mexico base on MODIS data.It was conducted with eigenvectors built by multispectral information and spectral matching with the texture characteristics to recognize targets in consideration of mix points phenomenon between oil slick and water.Meanwhile,it also used minimum distance algorithm and support vector machine to extract oil slick.Then it used confusion matrix to verify the results accuracy.The results showed the improved Spectral Angle Mapper（SAM）had good effect of monitoring oil spill area,and the overall accuracy was up to 90%,and they could meet the extraction accuracy requirement,which could be used in oil spill dynamic monitoring and provide scientific reference and technical support for oil spill trend and dispersion area.