On the use of feature selection to improve the detection of sea oil spills in SAR images

详细信息    查看全文

• 作者：David Mera^a ; ^{david.mera@usc.es} ; Veronica Bolon-Canedo^b ; ^{vbolon@udc.es} ; J.M. Cotos^a ; ^{manel.cotos@usc.es} ; Amparo Alonso-Betanzos^b ; ^{ciamparo@udc.es}
• 关键词：Feature Selection ; SVM-RFE ; Classification ; Support Vector Machines ; Synthetic Aperture Radar ; Oil spill detection
• 刊名：Computers & Geosciences
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：100
• 期：Complete
• 页码：166-178
• 全文大小：4913 K
• 卷排序：100

文摘

Five feature selection methods were applied to improve oil spill detection systems. Feature selection discarded irrelevant features and improved the classifier accuracy. The SVM-RFE feature selection method obtained the best accuracy results. A 6-input SVM classifier showed an accuracy of 87.1% and a Kappa statistic of 74.06%. 75.0% of the unique selected features belong to the geometrical category.