Font Size: a A A

Monitoring Of Discontinuous Oil Spill On Sea Surface Based On UAV Camera And LIF

Posted on:2022-05-27Degree:MasterType:Thesis
Country:ChinaCandidate:Y LiFull Text:PDF
GTID:2491306536996599Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Oil spill pollution on the sea is one of the most common and most severe marine pollution problems in the world,which seriously endangers the ecological environment and social development.It is urgent to strengthen the treatment and improvement of oil spill pollution.Accurate and rapid location and identification of oil spills on the sea is the basis and premise for the treatment of marine oil spills.However,there are no available theoretical research results and mature detection methods for the discontinuous distribution of oil spills on the sea,so accurate detection and evaluation results can not be obtained.Therefore,it is one of the important work to study the accurate and efficient monitoring means of discontinuous oil spill.In order to make up for the lack of this research,this paper studies the discontinuous oil spill monitoring based on UAV camera and ocean laser-induced fluorescence technology.Firstly,in order to improve the problem that SUSAN edge detection algorithm can not detect the edge of oil spill image with different contrast at the same time,an improved SUSAN edge detection algorithm based on non maximum suppression to determine the threshold t is proposed.In this algorithm,Sobel template is used to traverse the image to get the gradient amplitude matrix of the image,and non maximum suppression is used to process the gradient amplitude matrix to get a new gradient amplitude matrix,so as to get the adaptive threshold t,which can change with the change of image contrast,so as to better detect the edge of the oil spill image on the sea.Secondly,based on the edge detection and laser-induced fluorescence technology,the method to judge whether the oil spill is continuous in the monitoring area is introduced.In this method,the local multi-point sampling method is used to obtain the sampling points of each part of the image,and the real physical location of each sampling point is determined by camera calibration.Laser irradiation is used to induce fluorescence reaction at the sampling point,and the fluorescence intensity after treatment is calculated to determine whether the Raman scattering intensity of the sample point is the same as that of the detected oil-free seawater.Combined with the position relationship between the oil spill areas,the oil spill continuity in the whole monitoring area can be judged.Finally,the effectiveness of the improved algorithm and the method to judge whether the oil spill is continuous is verified by experiments.
Keywords/Search Tags:discontinuous oil spill on the sea, SUSAN edge detection algorithm, nonmaximum suppression, laser induced fluorescence technology, camera calibration, local multipoint sampling
PDF Full Text Request
Related items