Parallel Connected Generative Adversarial Network with Quadratic Operation for SAR Image Generation and Application for Classification

被引:13
作者
He, Chu [1 ,2 ]
Xiong, Dehui [1 ]
Zhang, Qingyi [1 ]
Liao, Mingsheng [2 ,3 ]
机构
[1] Wuhan Univ, Elect & Informat Sch, Wuhan 430072, Hubei, Peoples R China
[2] Collaborat Innovat Ctr Geospatial Technol, 129 Luoyu Rd, Wuhan 430079, Hubei, Peoples R China
[3] Wuhan Univ, State Key Lab Informat Engn Surveying Mapping & R, Wuhan 430079, Hubei, Peoples R China
基金
中国国家自然科学基金;
关键词
Synthetic Aperture Radar (SAR); image classification; Generative Adversarial Network (GAN); quadratic operation; TARGET RECOGNITION;
D O I
10.3390/s19040871
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
Thanks to the availability of large-scale data, deep Convolutional Neural Networks (CNNs) have witnessed success in various applications of computer vision. However, the performance of CNNs on Synthetic Aperture Radar (SAR) image classification is unsatisfactory due to the lack of well-labeled SAR data, as well as the differences in imaging mechanisms between SAR images and optical images. Therefore, this paper addresses the problem of SAR image classification by employing the Generative Adversarial Network (GAN) to produce more labeled SAR data. We propose special GANs for generating SAR images to be used in the training process. First, we incorporate the quadratic operation into the GAN, extending the convolution to make the discriminator better represent the SAR data; second, the statistical characteristics of SAR images are integrated into the GAN to make its value function more reasonable; finally, two types of parallel connected GANs are designed, one of which we call PWGAN, combining the Deep Convolutional GAN (DCGAN) and Wasserstein GAN with Gradient Penalty (WGAN-GP) together in the structure, and the other, which we call CNN-PGAN, applying a pre-trained CNN as a discriminator to the parallel GAN. Both PWGAN and CNN-PGAN consist of a number of discriminators and generators according to the number of target categories. Experimental results on the TerraSAR-X single polarization dataset demonstrate the effectiveness of the proposed method.
引用
收藏
页数:20
相关论文
共 44 条
[1]  
[Anonymous], J ELECT ENG
[2]  
[Anonymous], P 3 INT C LEARNING R
[3]  
[Anonymous], CONVOLUTIONAL NEURAL
[4]  
[Anonymous], PROC CVPR IEEE
[5]  
[Anonymous], 2017, ARXIV171102010
[6]  
[Anonymous], 2017, COMMUN ACM, DOI DOI 10.1145/3065386
[7]  
[Anonymous], 2017, ARXIV170107875
[8]  
[Anonymous], 2012, ARXIV 1212 5701
[9]  
[Anonymous], P 2018 IEEE INT GEOS
[10]   Classification of multifrequency polarimetric SAR imagery using a dynamic learning neural network [J].
Chen, KS ;
Huang, WP ;
Tsay, DH ;
Amar, F .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 1996, 34 (03) :814-820