Research On Detection Of Oil Spill At Sea Based On Hyperspectral Imagery

Now, with people’s increased awareness of protecting the marine environment, the problem of the oil spill detection is widely concerned. When the oil spill accident occurs, the oil spill will form an isolation layer and damage to marine ecological environment. Meanwhile, the oil spill contains large amounts of toxic substances. Thus, it is particularly important to detect and deal with sea oil spill. In the area of the detecting oil spill at sea, hyperspectral image technology has been paid more attention.Usually, there were two detecting methods of Oil spill, artificial auxiliary detecting method which was obtaining the oil spill information by simulating the oil spill in the laboratory, and automatic detecting method which was obtaining the oil spill information automatically when the spectral information of oil spill and the information of background were unknown. Compared with the two methods, artificial auxiliary detection has higher detecting accuracy, but it was influenced by various environmental factors in the practical application. Automatic detection can deal with the complex environment. In recent years, the oil spill detection mostly based on artificial assistant method. In this paper, the detection of oil spill was from the perspective of automation.The detection of oil spill is usually divided into two steps. First, it is important to get the oil spill and the background reference spectral information. And then in order to detect the domain of oil spill, the detection algorithm is also necessary. In this paper, the steps of the detection of oil spill as follows: Firstly, the spectral information of pure pixels was extracted from hyperspectral images by using MNF transform and PPI algorithm. And then the author research on a new method for identifying oil spill spectra based on absorption band of oil spill, which is used to identify the spectral information of the oil spill. Meanwhile, the threshold segmentation method is used to obtain the background parameters. Finally, the detection of oil spill is accomplished by using the adaptive matching filter algorithm.