中国海洋大学学报（自然科学版）2017，Vol.47Issue(2)：106-115,10.DOI:10.16441/j.cnki.hdxb.20160196
海上溢油合成孔径雷达探测研究
Study of Oil Spill Detection on SAR Images
摘要
Abstract
Discriminating oil spills from lookalike phenomena is a crucial procedure in oil spill detection.To achieve this purpose,three-step approach is taken in general:firstly,features of oil spills and lookalikes are extracted;then,key features which are beneficial to the oil spill classification are screened out;finally,effective classifier is built and pattern recognition method is used to conduct classification.In this paper,16 kinds of features which include geometric features,surrounding features,backscattering features and textural features of 138 oil spills and lookalikes are extracted from 15 SAR images.The images were acquired during Penglai 19-3 Platform oil spill accident in 2011.The 16 features are sorted from big to small based on the FDR value of the single feature.We find that the standard deviation of backscattering coefficient of the backgrounds has larger FDR value.Therefore,it can be selected as the first feature.Then,the forward selection method of sequential search method are used to determinate the optimal feature subset for oil spill detection.We find that the standard deviation of backscattering coefficient of the backgrounds,inverse difference moment,energy and the mean value of backscattering coefficient can be selected as the optimal feature subset in this work.CART(Classification And Regression Tree) is a kind of binary decision tree which is helpful to improve efficiency of generating tree.However,the disadvantage of the decision tree classifier is that the variance of classification results is quite high.So the decision tree classifier is an unstable classifier.While for bagging algorithm,the only real training set in practice are divided into different training sets through resampling methods.And that is benefit for improving the unstable classifier performance.The bagging method based on decision-making tree combines massive calculation of single classifiers which is helpful to improve the accuracy of oil spill detection.In this paper,we combine CART with bagging algorithm to classify the oil spills from look-alike.Multiple training data sets with the same size are generated by random selection,and then several decision tree models can be established.So the oil spills and lookalikes can be classified by voting.The experiment results show that,the classification accuracy tends to be stable through 100 iterations.In order to get effective classification results,five-fold and ten-fold cross validation are employed to evaluate the classification results,and the results show that the average classification accuracy of oil spills and lookalikes is above 85 ％.At last,we compared the CART that based on bagging algorithm with conventional CART and BP Neural network classification algorithm for oil spill discrimination.We find that the classification accuracy of our method is higher than the other two methods,which indicates that the effectiveness of our method in oil spill detection on SAR image.关键词
合成孔径雷达/特征选择/油膜分类/CART决策树/BaggingKey words
Synthetic Apeture Radar(SAR)/feature selection/oil spill classification/decision tree/bagging分类
信息技术与安全科学引用本文复制引用
张彦敏,徐卓,旭锋..海上溢油合成孔径雷达探测研究[J].中国海洋大学学报（自然科学版）,2017,47(2):106-115,10.基金项目
海洋公益性科研专项(201505002) （201505002）
国家自然科学基金项目(61501520)资助 Supported by the Public Science and Technology Research Funds Projects of Ocean(201505002) （61501520）
the National Natural Science Foundation of China (61501520) （61501520）
