A Sea Area Oil Spill Detection Method Based on SAR Image

Abstract

The invention provides a sea oil spill detection method based on an SAR image. Sea oil spill can be accurately detected for different ocean scenes. The method comprises the steps that before target extraction, mean value filtering is carried out on the SAR image; a maximum between-cluster variance method is used to carry out threshold segmentation of the first time on the SAR image after mean value filtering, and dark sea and partial oil spill dark spots are segmented from the whole image; a local contrast drawing method based on context characteristics is used to extract the oil spill dark spots from the dark sea; for the SAR sea image with inhomogeneous contrast, an extensive dark sea region can be acquired by using threshold segmentation of the first time; morphological operation is carried out, namely small scatter dark spots in the extracted image of the dark sea region are removed; and false alarm rejection is carried out by using a method based on context characteristics.

Description

technical field [0001] The invention relates to a sea area oil spill detection method based on SAR images, and belongs to the technical field of target detection and recognition. Background technique [0002] The main sources of oil spills in sea areas are ship oil spills, illegal waste oil discharge, and oil spills from offshore oil exploitation, which have seriously affected the marine ecological environment. In order to effectively detect and control oil spill pollution in sea areas, the current sea area oil spill detection methods based on SAR images can be mainly divided into: oil spill detection algorithms based on gray features, detection algorithms based on texture features, detection algorithms based on edge features, etc. . [0003] The oil spill detection algorithm based on grayscale features is mainly based on the fact that the oil spill area presents a black dark spot area in the SAR sea area image, that is, the pixel value of the oil spill area is lower than t...

AI Technical Summary

Problems solved by technology

Although this method can accurately extract the boundary, it is greatly affected by interference. For example, when there is a sea breeze or the image has land, the gradient image with very rich edges will be obtained through the edge detection method, which will make it difficult to judge the oil spill area.

Images