A COMPLEX-VALUED CONVOLUTIONAL NEURAL NETWORK WITH DIFFERENT ACTIVATION FUNCTIONS IN POLARIMETRIC SAR IMAGE CLASSIFICATION

被引：3
|
作者：
Zhang, Yun [1 ]
Hua, Qinglong [1 ]
Xu, Dan [1 ]
Li, Hongbo [1 ]
Mu, HuiLin [1 ]
机构：
[1] Harbin Inst Technol, Sch Elect Informat Engn, Harbin, Peoples R China
来源：
2019 INTERNATIONAL RADAR CONFERENCE (RADAR2019) | 2019年
基金：
中国国家自然科学基金;
关键词：
Complex-valued convolutional neural network (CV-CNN); optimization algorithm; activation function; synthetic aperture radar (SAR); terrain classification; deep learning;
D O I：
10.1109/RADAR41533.2019.171298
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
It is well known that the activation function and the gradient descent optimization algorithm have a great influence on convolutional neural network (CNN). Based on the Adam optimization algorithm, this paper proposes a CvAdam optimization algorithm suitable for complex-valued convolutional neural network (CV-CNN), and then in the typical polarization SAR image classification task, Adam and CvAdam were compared using four different activation functions of sigmoid, tanh, Leakey-ReLU and ELU. Experiments on the benchmark dataset of Oberpfaffenhofen show that CvAdam performs better than Adam in both convergence speed and accuracy, no matter which activation function is used.
引用
收藏
页码：749 / 752
页数：4
相关论文
共 50 条
  • [31] Polarimetric SAR Image Classification Using Deep Convolutional Neural Networks
    Zhou, Yu
    Wang, Haipeng
    Xu, Feng
    Jin, Ya-Qiu
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2016, 13 (12) : 1935 - 1939
  • [32] On activation functions for complex-valued neural networks - Existence of energy functions
    Kuroe, Y
    Yoshida, M
    Mori, T
    ARTIFICAIL NEURAL NETWORKS AND NEURAL INFORMATION PROCESSING - ICAN/ICONIP 2003, 2003, 2714 : 985 - 992
  • [33] KERNEL-BASED CLASSIFICATION IN COMPLEX-VALUED FEATURE SPACES FOR POLARIMETRIC SAR DATA
    Moser, Gabriele
    Serpico, Sebastiano B.
    2014 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2014, : 1257 - 1260
  • [34] End-to-End Full Complex-Valued Domain SAR Target Classification Neural Network
    Fang, Cheng
    Guan, Fang-Heng
    Li, Tian-Chi
    Zou, Zheng-Feng
    Yang, Lei
    Tien Tzu Hsueh Pao/Acta Electronica Sinica, 2024, 52 (07): : 2449 - 2460
  • [35] CVGG-Net: Ship Recognition for SAR Images Based on Complex-Valued Convolutional Neural Network
    Zhao D.
    Zhang Z.
    Lu D.
    Kang J.
    Qiu X.
    Wu Y.
    IEEE Geoscience and Remote Sensing Letters, 2023, 20
  • [36] Through-Wall Human Activity Classification Using Complex-Valued Convolutional Neural Network
    Wang, Xiang
    Chen, Pengyun
    Xie, Hangchen
    Cui, Guolong
    2021 IEEE RADAR CONFERENCE (RADARCONF21): RADAR ON THE MOVE, 2021,
  • [37] Complex-valued unsupervised convolutional neural networks for sleep stage classification
    Zhang, Junming
    Wu, Yan
    COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2018, 164 : 181 - 191
  • [38] FPGA Implementation of Complex-Valued Neural Network for Polar-Represented Image Classification
    Ahmad, Maruf
    Zhang, Lei
    Chowdhury, Muhammad E. H.
    SENSORS, 2024, 24 (03)
  • [39] PolSARNet: A Deep Fully Convolutional Network for Polarimetric SAR Image Classification
    Mullissa, Adugna G.
    Persello, Claudio
    Stein, Alfred
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2019, 12 (12) : 5300 - 5309
  • [40] Multistability of complex-valued neural networks with discontinuous activation functions
    Liang, Jinling
    Gong, Weiqiang
    Huang, Tingwen
    NEURAL NETWORKS, 2016, 84 : 125 - 142
12345