海上油膜及浒苔的遥感提取*
Offshore Oil Film and Enteromorpha Extraction by Remote Sensing Data

过杰1**,尚伟涛1,姜晓鹏1,刘欣1,过爽2
GUO Jie1,SHANG Weitao1,JIANG Xiaopeng1,LIU Xin1,GUO Shuang2

(1.中国科学院烟台海岸带研究所,山东省海岸带环境过程重点实验室,山东烟台 264003;2.河南城建学院电气与信息工程学院,河南平顶山 467000)
(1.Key Laboratory of Coastal Environmental Processes and Ecological Remediation,Yantai Institute of Coastal Zone Research,Chinese Academy of Sciences,Yantai,Shandong,264003,China;2.School of Electrical and Information Engineering,Henan University of Urban Construction,Pingdingshan,Henan,467000,China)


摘要:【目的】为给海上污染物的清除提供技术支持,应用Radarsat2全极化数据和均值阈值法对海上油膜及浒苔的提取进行研究。【方法】利用图像分割求取后向散射系数(NRCS)均值d,并用d除以重采样(10*10)平均值得比值t,再根据t值选取得到油膜或浒苔的最小阈值,从而快速识别油膜或浒苔(悬浮与漂浮)面积。【结果】该方法提取油膜所得结果好于熵方法;提取悬浮及漂浮浒苔结果与熵方法、平均Alpha角方法所得结果吻合。【结论】该方法对海上油膜和浒苔的提取有效。
关键词:Radarsat2 数据     油膜    浒苔    均值阈值法
中图分类号:P736.22      文献标识码:A      文章编号:1002-7378(2016)02-0107-05
Abstract:【Objective】In order to provide technical support for marine pollutant removal,the oil film and enteromorpha extract methods were explored by Quadpolarization Radarsat2 data.【Methods】Average threshold value methodthe normalized radar cross section (NRCS) are calculated (d) after using image segmentation. The resampling 10*10 average value divided by d was noted for t.According to t value selected a minimum threshold for oil film or enteromorpha,so that the method immediately identified the area of oil film or enteromorpha (suspension and floating) area.【Results】The method used to extract in the oil film is better than that of entropy. Extraction of suspension and floating enteromorpha by entropy and the average Angle of alpha (alpha) method are consistent with that by average threshold value method.【Conclusion】The method is effective for oil film and sea enteromorpha extract.
Key words:Radarsat2 data,oil film,enteromorpha,average threshold value method

*国家自然科学面上基金项目(41176160,41576032),国家自然科学基金委员会与俄罗斯基础研究基金会合作交流项目(4141101049)和国家海洋局北海分局渤海中部公共海域沉积物现场微生物修复项目(QDZC20150420-002)资助。

 PDF全文下载cbimg.cnki.net/Editor/2016/0623/gxkx/5ff1ab91-aef0-4f8d-8903-43e1eb9c9559.pdf

发布日期:2016/6/23