| 基于SAR极化比和纹理特征的海面溢油识别方法 |
| |
| 引用本文： | 陈韩,谢涛,方贺,孟雷,赵立,艾润冰.基于SAR极化比和纹理特征的海面溢油识别方法[J].海洋学报,2019,41(9):181-190. |
| |
| 作者姓名： | 陈韩 谢涛 方贺 孟雷 赵立 艾润冰 |
| |
| 作者单位： | 南京信息工程大学海洋科学学院,江苏南京,210044;青岛海洋科学与技术试点国家实验室区域海洋动力学与数值模拟功能实验室,山东青岛266237;南京信息工程大学遥感与测绘工程学院,江苏南京210044;北京市511信箱,北京,100094 |
| |
| 基金项目： | 国家自然科学基金项目（41776181）；国家重点研发计划项目（2016YFC1401007）；全球变化研究国家重大科学研究计划项目（2015CB953901）；江苏省研究生科研创新计划（KYCX18_1012）。 |
| |
| 摘 要： | 针对海洋表面SAR影像的特点，采用基于灰度共生矩阵的纹理特征方法是提取海面溢油信息的常用方法，但实际海洋表面复杂的信息使得SAR图像上产生类似溢油现象的暗斑区域，这导致在利用纹理特征方法提取溢油信息时存在虚警率，降低了溢油信息的提取精度。基于RADARSAT-2 SAR四极化影像，本文提出基于SAR极化比影像的纹理特征识别方法对海面油膜进行识别提取。结果显示，基于SAR极化比影像的纹理特征识别方法可以有效且准确地提取海面溢油信息，相比于VV极化影像的纹理特征识别方法，溢油监测过程中的虚警率降低了17.96%，溢油监测总体精度达到96.83%。
|
| 关 键 词： | 合成孔径雷达 溢油识别 极化比 纹理特征 |
| 收稿时间： | 2018/9/8 0:00:00 |
| 修稿时间： | 2018/11/25 0:00:00 |
Sea surface oil spill identification method based on SAR polarization ratio and texture feature |
| |
| Chen Han,Xie Tao,Fang He,Meng Lei,Zhao Li and Ai Runbing.Sea surface oil spill identification method based on SAR polarization ratio and texture feature[J].Acta Oceanologica Sinica (in Chinese),2019,41(9):181-190. |
| |
| Authors: | Chen Han Xie Tao Fang He Meng Lei Zhao Li and Ai Runbing |
| |
| Institution: | 1.School of Marine Science, Nanjing University of Information Science and Technology, Nanjing 210044, China2.Laboratory for Regional Oceanography and Numerical Modeling, Pilot National Laboratory for Marine Science and Technology (Qingdao), Qingdao 266237, China3.School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China4.Beijing City 5111 Mailbox, Beijing 100094, China |
| |
| Abstract: | Aiming at the characteristics of SAR images on the ocean surface, the texture feature method based on gray level co-occurrence matrix is a common method for extracting oil spill information from the sea surface, but the complex information on the actual ocean surface makes the SAR image produce a dark spot area similar to the oil spill phenomenon. The false alarm rate is obtained when the oil feature information is extracted by the texture feature method, and the extraction precision of the oil spill information is reduced. Based on the RADARSAT-2 SAR quadratic polarization image, this paper proposes a texture feature recognition method based on SAR polarization ratio image to identify and extract the oil film on the sea surface. The results show that the texture feature recognition method based on SAR polarization ratio image can effectively and accurately extract the oil spill information on the sea surface. Compared with the texture feature recognition method of VV polarization image, the false alarm rate in the oil spill monitoring process is reduced by 17.96%, the overall accuracy of oil spill monitoring reached 96.83%. |
| |
| Keywords: | synthetic aperture radar oil spill identification polarization ratio texture feature |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《海洋学报》浏览原始摘要信息 |
| 点击此处可从《海洋学报》下载免费的PDF全文 |
|
