Feature Selection and Recognition Based on SAR Image
Zhenzhen Wan, Chaoshu Jiang, Yanfang Liu, Jie Lin, Wei Xu, Xiaojie Qu, Xiangtao Min, Xiaoyu Zhang · 2021 CIE International Conference on Radar (Radar) · 2021
Synthetic Aperture Radar (SAR) image feature selecting can effectively reduce the data dimension of subsequent processing through feature extraction and selection based on a small number of sample images, thereby reduce the complexity of subsequent processing and the workload in the process of feature extraction for a large number of images. In this paper, a method composed of feature difference analysis and Relief algorithm is proposed to select the features. The proposed method is called D-Relief method. Then support vector machine (SVM) is utilized to verify the influence of different feature selection methods on the classification results. The experimental results show that the proposed D-Relief method can further reduce the dimension of feature vector and ensure the performance of classifier.