Pixel-Wise PolSAR Image Classification via a Novel Complex-Valued Deep Fully Convolutional Network

被引:43
|
作者
Cao, Yice [1 ]
Wu, Yan [1 ]
Zhang, Peng [2 ]
Liang, Wenkai [1 ]
Li, Ming [2 ]
机构
[1] Xidian Univ, Sch Elect Engn, Remote Sensing Image Proc & Fusion Grp, Xian 710071, Shaanxi, Peoples R China
[2] Xidian Univ, Natl Lab Radar Signal Proc, Xian 710071, Shaanxi, Peoples R China
基金
中国国家自然科学基金;
关键词
complex-valued deep fully convolutional neural network (CV-FCN); polarimetric synthetic aperture radar (PolSAR) image classification; pixel-level labeling; POLARIMETRIC SAR IMAGERY; NEURAL-NETWORK; SEMANTIC SEGMENTATION; LAND CLASSIFICATION;
D O I
10.3390/rs11222653
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Although complex-valued (CV) neural networks have shown better classification results compared to their real-valued (RV) counterparts for polarimetric synthetic aperture radar (PolSAR) classification, the extension of pixel-level RV networks to the complex domain has not yet thoroughly examined. This paper presents a novel complex-valued deep fully convolutional neural network (CV-FCN) designed for PolSAR image classification. Specifically, CV-FCN uses PolSAR CV data that includes the phase information and uses the deep FCN architecture that performs pixel-level labeling. The CV-FCN architecture is trained in an end-to-end scheme to extract discriminative polarimetric features, and then the entire PolSAR image is classified by the trained CV-FCN. Technically, for the particularity of PolSAR data, a dedicated complex-valued weight initialization scheme is proposed to initialize CV-FCN. It considers the distribution of polarization data to conduct CV-FCN training from scratch in an efficient and fast manner. CV-FCN employs a complex downsampling-then-upsampling scheme to extract dense features. To enrich discriminative information, multi-level CV features that retain more polarization information are extracted via the complex downsampling scheme. Then, a complex upsampling scheme is proposed to predict dense CV labeling. It employs the complex max-unpooling layers to greatly capture more spatial information for better robustness to speckle noise. The complex max-unpooling layers upsample the real and the imaginary parts of complex feature maps based on the max locations maps retained from the complex downsampling scheme. In addition, to achieve faster convergence and obtain more precise classification results, a novel average cross-entropy loss function is derived for CV-FCN optimization. Experiments on real PolSAR datasets demonstrate that CV-FCN achieves better classification performance than other state-of-art methods.
引用
收藏
页数:29
相关论文
共 50 条
  • [41] Complex-Valued Convolutional Neural Networks for Real-Valued Image Classification
    Popa, Calin-Adrian
    2017 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2017, : 816 - 822
  • [42] Deep convolutional neural network-based pixel-wise landslide inventory mapping
    Su, Zhaoyu
    Chow, Jun Kang
    Tan, Pin Siang
    Wu, Jimmy
    Ho, Ying Kit
    Wang, Yu-Hsing
    LANDSLIDES, 2021, 18 (04) : 1421 - 1443
  • [43] CMID: Crossmodal Image Denoising via Pixel-Wise Deep Reinforcement Learning
    Guo, Yi
    Gao, Yuanhang
    Hu, Bingliang
    Qian, Xueming
    Liang, Dong
    SENSORS, 2024, 24 (01)
  • [44] A novel complex-valued convolutional network for real-world single image dehazing
    Xie, Xinxiu
    Li, Chaofeng
    Guan, Tuxin
    Zheng, Yuhui
    Wu, Xiaojun
    JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION, 2023, 97
  • [45] Superpixel-Oriented Classification of PolSAR Images Using Complex-Valued Convolutional Neural Network Driven by Hybrid Data
    Qin, Xianxiang
    Zou, Huanxin
    Yu, Wangsheng
    Wang, Peng
    IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2021, 59 (12): : 10094 - 10111
  • [46] COMPLEX-VALUED SELF-SUPERVISED POLSAR IMAGE CLASSIFICATION INTEGRATING ATTENTION MECHANISM
    Kuang, Zuzheng
    Bi, Haixia
    Li, Fan
    IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, 2023, : 5958 - 5961
  • [47] PSIC-Net: Pixel-Wise Segmentation and Image-Wise Classification Network for Surface Defects
    Lei, Linjian
    Sun, Shengli
    Zhang, Yue
    Liu, Huikai
    Xu, Wenjun
    MACHINES, 2021, 9 (10)
  • [48] Efficiently utilizing complex-valued PolSAR image data via a multi-task deep learning framework
    Zhang, Lamei
    Dong, Hongwei
    Zou, Bin
    ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2019, 157 : 59 - 72
  • [49] A COMPLEX-VALUED CONVOLUTIONAL NEURAL NETWORK WITH DIFFERENT ACTIVATION FUNCTIONS IN POLARIMETRIC SAR IMAGE CLASSIFICATION
    Zhang, Yun
    Hua, Qinglong
    Xu, Dan
    Li, Hongbo
    Mu, HuiLin
    2019 INTERNATIONAL RADAR CONFERENCE (RADAR2019), 2019, : 749 - 752
  • [50] A Deep Convolutional Neural Network for Semantic Pixel-Wise Segmentation of Road and Pavement Surface Cracks
    Jenkins, Mark David
    Carr, Thomas Arthur
    Iglesias, Maria Insa
    Buggy, Tom
    Morison, Gordon
    2018 26TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO), 2018, : 2120 - 2124