Journal of Frontiers of Computer Science and Technology ›› 2022, Vol. 16 ›› Issue (10): 2405-2414.DOI: 10.3778/j.issn.1673-9418.2102073

• Graphics and Image • Previous Articles    

Improved Two-Branch Capsule Network for Hyperspectral Image Classification

ZHANG Haitao, CHAI Simin+()   

  1. Software College, Liaoning Technical University, Huludao, Liaoning 125105, China
  • Received:2021-03-01 Revised:2021-06-07 Online:2022-10-01 Published:2021-06-23
  • About author:ZHANG Haitao, born in 1974, Ph.D., profes-sor, M.S. supervisor, member of CCF. His resea-rch interests include machine learning, graphics and image processing, etc.
    CHAI Simin, born in 1996, M.S. candidate. His research interests include machine learning, graphics and image processing, etc.
  • Supported by:
    Pre-research Foundation of National Ministries and Commissions;Natural Science Foundation of Liaoning Province (General Program)(20170540426)

改进双分支胶囊网络的高光谱图像分类

张海涛, 柴思敏+()   

  1. 辽宁工程技术大学 软件学院,辽宁 葫芦岛 125105
  • 通讯作者: + E-mail: 1149278938@qq.com
  • 作者简介:张海涛(1974—),男,博士,教授,硕士生导师,CCF会员,主要研究方向为机器学习、图形图像处理等。
    柴思敏(1996—),男,硕士研究生,主要研究方向为机器学习、图形图像处理等。
  • 基金资助:
    国家部委预研基金;辽宁省自然科学基金面上项目(20170540426)

Abstract:

The method based on the dual-channel capsule network extracts spectral information and spatial informa-tion separately in two channels, which not only retains the feature extraction method of the dual-channel convolu-tional neural network, but also improves the classification accuracy. However, when researchers train the capsule network, the dynamic routing process generates a large number of training parameters because the hyperspectral image (HSI) usually consists of hundreds of channels. To address this limitation, 1D and 2D constraint windows are proposed to reduce the number of capsules from two extraction channels. It uses the capsule vector group as the calculation unit to perform convolution operations and reduce the amount of parameters and computational com-plexity of the capsule network. Based on this parameter reduction optimization method, a new dual-branch capsule neural network (DuB-ConvCapsNet-MRF) is proposed and applied to the task of hyperspectral image classifica-tion. In addition, in order to further improve the classification accuracy, Markov random field (MRF) is introduced to smooth the spatial region and the final output is got. The results of performing ablation experiments on two repre-sentative hyperspectral image datasets and comparing the proposed method with six existing classification methods show that DuB-ConvCapsNet-MRF is superior to other methods in classification performance, and effectively re-duces the cost of training capsule network.

Key words: remote sensing, hyperspectral image classification, capsule neural network, constraint window, Markov random field (MRF)

摘要:

基于双分支的胶囊网络分类方法在两个通道分别提取光谱信息和空间信息,既保留了双分支卷积神经网络的特征提取方式,又提高了分类精度。但由于高光谱图像(HSI)通常由几百个通道组成,在训练胶囊网络时,动态路由过程产生了大量的训练参数。为此提出1D和2D约束窗口分别减少来自两个提取通道的胶囊数量。它以胶囊向量组为计算单位进行卷积运算,来减少胶囊网络的参数量和计算复杂度。基于该降参优化方法提出一个新的双分支胶囊神经网络(DuB-ConvCapsNet-MRF),并将其应用在高光谱图像分类任务中。此外,为进一步提高分类性能,引入马尔可夫随机场(MRF)对空间区域进行平滑后处理,获得最终输出。对两个代表性高光谱图像数据集进行消融实验并与现有的6个分类方法进行比较,结果表明,DuB-ConvCapsNet-MRF在分类精度上都优于其他方法,并且有效降低了胶囊网络的训练代价。

关键词: 遥感, 高光谱图像分类, 胶囊神经网络, 约束窗口, 马尔可夫随机场(MRF)

CLC Number: