A New Algorithm for SAR Image Target Recognition Based on an Improved Deep Convolutional Neural Network

Abstract

To effectively make use of the automatic feature extraction ability of biologically inspired deep learning technology, and enhance the ability of depth learning method to learn features, this paper proposed a deep learning algorithm combining deep convolutional neural network (DCNN) trained with an improved cost function and support vector machine (SVM). The class separation information, which explicitly facilitates intra-class compactness and inter-class separability in the process of learning features, is added to an improved cost function as a regularization term to enhance the feature extraction ability of DCNN. Then, the improved DCNN is applied to learn the features of SAR images. Finally, SVM is utilized to map the features into output labels. Experiments are performed on SAR image data in moving and stationary target acquisition and recognition (MSTAR) database. The experiment results prove the effectiveness of our method, achieving an average accuracy of 99% on ten types of targets, some variants, and some articulated targets. It proves that our method is effective and CNN enjoys a certain potential to be applied in SAR image target recognition.