论文部分内容阅读
海面溢油对生态环境造成了严重危害,故及早发现和尽快处理对降低事故影响和经济损失起着至关重要的作用。合成孔径雷达(SAR)是观测海面溢油、快速检测和事故态势分析判断的有效技术途径。本文针对SAR图像的海面溢油检测,提出了一种特征概率函数的双阈值分割方法。首先,通过高低阈值分割提取不同层次的灰度信息,再利用密度估计提取灰度的空间分布信息,然后,通过构建概率函数对油膜和类油膜区域进行形态学分类,最后,结合辅助信息,获得最终的海面溢油检测结果。本文利用香港中文大学卫星地面站接收的ENVISAT ASAR图像开展实验,结果表明,本文提出的方法能够准确地排除由风场或者水流场导致的低散射区域,有效地检测和识别生成不久的中型油膜,从而有助于溢油事故的早期预警与处置。
Oil spills on the sea have caused serious damage to the ecological environment. Therefore, early detection and prompt treatment play a crucial role in reducing the impact of accidents and economic losses. Synthetic Aperture Radar (SAR) is an effective technique for observing sea oil spills, rapid detection and accident situation analysis. In this paper, aiming at SAR image sea surface oil spill detection, a dual thresholding method based on feature probability function is proposed. Firstly, the gray-level information of different levels is extracted through high and low threshold segmentation, and then the spatial distribution information of gray-level is extracted by density estimation. Then, the oil film and oil-like film are classified by the probability function. Finally, combined with the auxiliary information, The final sea spill test results. In this paper, the ENVISAT ASAR images received by the Chinese University of Hong Kong satellite ground station are used to carry out experiments. The results show that the proposed method can accurately exclude the low scattering area caused by the wind field or water flow field and effectively detect and identify medium oil film , Thus contributing to the early warning and disposal of oil spills.