海上溢油扩散图像处理的非线性方法研究  

Research on oil spill at sea by means of nonlinear method

作　　者：罗智慧[1] 龙新峰[1] 

机构地区：[1]华南理工大学化工与能源学院,广东广州510640

出　　处：《海洋环境科学》2009年第4期449-453,共5页Marine Environmental Science

基　　金：广东省自然科学基金项目(05006518)

摘　　要：针对海上溢油应急处理问题,本文主要对溢油的初级行为——扩散进行探讨。提出了采用溢油扩散图像来预测溢油扩散未来发展趋势的3步非线性处理过程,即图像边缘检测、面积计算和面积预测,对这一过程中应采用的方法进行了分析与比较。并以奥里油在静止海水中扩散为范例,在MATLAB环境下,利用非线性方法对扩散图像进行边缘监测,通过对梯度算子阀值的调整来计算完整的油膜覆盖面积,并将其计算结果进行BP神经网络预测。结果表明:采用Roberts边缘算子可以得到很好的边缘检测结果,采用BP神经网络获得的面积预测结果与实验结果吻合程度较好。Focus on the problems about the oil spill, the first behavior of diffusion is discussed mainly when the oil spill at sea appeared. Three steps of nonlinear process, namely edge image detected and area computation and area forecast, is put forward to forecast the developmental tend of oil layer by the oil diffusion image. At the same time, the methods used above the process are also com- pared and analyzed. Based on the experiment of Orimulsion diffusing in the still water, with the help of MATLAB, the edge of Orimulsion diffusing image is detected using nonlinear means, and then the intact area covered by Orimulsion is calculated. Finally, the diffusion tend is forecasted with BP neural network. The results show that the edge curve gained by Robert arithmetic operators is good, and the forecast result gained by BP neural network is also agreement with the result of experiment.

关 键 词：溢油扩散 非线性 图像处理 BP神经网络 

分 类 号：TK514[动力工程及工程热物理—热能工程] X87[环境科学与工程—环境工程]