Main > Laser & Optoelectronics Progress >  Volume 57 >  Issue 12 >  Page 122803 > Article
  • Abstract
  • Abstract
  • View Summary
  • Figures (6)
  • Tables (2)
  • Equations (0)
  • References (24)
  • Get PDF(in Chinese)
  • Paper Information
  • Received: Oct. 1, 2019

    Accepted: Oct. 29, 2019

    Posted: Jun. 1, 2020

    Published Online: Jun. 3, 2020

    The Author Email: Zhao Jingyi (zjylwsr@126.com)

    DOI: 10.3788/LOP57.122803

  • Get Citation
  • Copy Citation Text

    Li Hu, Rui Shan, Fang Wang, Guoqian Jiang, Jingyi Zhao, Zhi Zhang. Hyperspectral Image Classification Based on Dual-Channel Dilated Convolution Neural Network[J]. Laser & Optoelectronics Progress, 2020, 57(12): 122803

    Download Citation

  • Category
  • Remote Sensing and Sensors
  • Share
Laser & Optoelectronics Progress, Vol. 57, Issue 12, 122803 (2020)

Hyperspectral Image Classification Based on Dual-Channel Dilated Convolution Neural Network

Hu Li1, Shan Rui1, Wang Fang1, Jiang Guoqian2, Zhao Jingyi3,*, and Zhang Zhi4

Author Affiliations

  • 1School of Science, Yanshan University, Qinhuangdao, Hebei 0 66001, China
  • 2School of Electrical Engineering, Yanshan University, Qinhuangdao, Hebei 0 66001, China
  • 3School of Mechanical Engineering, Yanshan University, Qinhuangdao, Hebei 0 66001, China
  • 4Beijing Institute of Space Mechanics & Electricity, Beijing 100094, China;

Abstract

Based on the excellent hole convolution performance observed using the obtained image information, we propose a framework for performing hyperspectral image classification based on the dual-channel dilated convolution neural network (DCD-CNN) to improve the classification accuracy. The receptive field of the filters can be expanded via dilated convolution, which effectively avoides the loss of image information and improves the classification accuracy. In this proposed framework, one-dimensional CNN and two-dimensional CNN, exhibiting an empty convolution, are used to extract the spectral and spatial features of the hyperspectral images. Subsequently, these extracted features are combined using a weighted fusion method. Finally, the combined features are input into the support vector machine for performing final classification. The expreimental results on the two commonly used hyperspectral image datasets by the proposed framework are compared with that by the four existing classification methods, showing that the proposed framework exhibits improved classification performance.

keywords

Please Enter Your Email: