An Adaptive Capsule Network for Hyperspectral Remote Sensing Classification

被引:13
|
作者
Ding, Xiaohui [1 ,2 ,3 ,4 ]
Li, Yong [1 ,2 ,3 ,4 ]
Yang, Ji [1 ,2 ,3 ,4 ]
Li, Huapeng [5 ]
Liu, Lingjia [6 ]
Liu, Yangxiaoyue [1 ,2 ,3 ,4 ]
Zhang, Ce [7 ,8 ]
机构
[1] Guangdong Acad Sci, Guangzhou Inst Geog, Guangzhou 510070, Peoples R China
[2] Southern Marine Sci & Engn Guangdong Lab Guangzho, Guangzhou 511458, Peoples R China
[3] Guangdong Open Lab Geospatial Informat Technol &, Guangzhou 510070, Peoples R China
[4] Key Lab Guangdong Utilizat Remote Sensing & Geog, Guangzhou 510070, Peoples R China
[5] Chinese Acad Sci, Northeast Inst Geog & Agroecol, Changchun 130102, Peoples R China
[6] Jiangxi Normal Univ, Sch Geog & Environm, Nanchang 330027, Jiangxi, Peoples R China
[7] Univ Lancaster, Lancaster Environm Ctr, Lancaster LA1 4YQ, England
[8] UK Ctr Ecol & Hydrol, Lib Ave, Lancaster LA1 4AP, England
基金
中国国家自然科学基金; 中国博士后科学基金;
关键词
capsule network; hyperspectral remote sensing; adaptive routing algorithm; deep learning; CONVOLUTIONAL NEURAL-NETWORK; IMAGE CLASSIFICATION; MODEL; CNN;
D O I
10.3390/rs13132445
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
The capsule network (Caps) is a novel type of neural network that has great potential for the classification of hyperspectral remote sensing. However, the Caps suffers from the issue of gradient vanishing. To solve this problem, a powered activation regularization based adaptive capsule network (PAR-ACaps) was proposed for hyperspectral remote sensing classification, in which an adaptive routing algorithm without iteration was applied to amplify the gradient, and the powered activation regularization method was used to learn the sparser and more discriminative representation. The classification performance of PAR-ACaps was evaluated using two public hyperspectral remote sensing datasets, i.e., the Pavia University (PU) and Salinas (SA) datasets. The average overall classification accuracy (OA) of PAR-ACaps with shallower architecture was measured and compared with those of the benchmarks, including random forest (RF), support vector machine (SVM), 1-dimensional convolutional neural network (1DCNN), two-dimensional convolutional neural network (CNN), three-dimensional convolutional neural network (3DCNN), Caps, and the original adaptive capsule network (ACaps) with comparable network architectures. The OA of PAR-ACaps for PU and SA datasets was 99.51% and 94.52%, respectively, which was higher than those of benchmarks. Moreover, the classification performance of PAR-ACaps with relatively deeper neural architecture (four and six convolutional layers in the feature extraction stage) was also evaluated to demonstrate the effectiveness of gradient amplification. As shown in the experimental results, the classification performance of PAR-ACaps with relatively deeper neural architecture for PU and SA datasets was also superior to 1DCNN, CNN, 3DCNN, Caps, and ACaps with comparable neural architectures. Additionally, the training time consumed by PAR-ACaps was significantly lower than that of Caps. The proposed PAR-ACaps is, therefore, recommended as an effective alternative for hyperspectral remote sensing classification.
引用
收藏
页数:17
相关论文
共 50 条
  • [1] Hyperspectral Remote Sensing Classification Using Multi-Scale Adaptive Capsule Network
    Zhang Gen
    Ding Xiaohui
    Yang Ji
    Wang Hui
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2022, 59 (24)
  • [2] Hyperspectral Remote Sensing Image Classification Using Deep Convolutional Capsule Network
    Lei, Runmin
    Zhang, Chunju
    Liu, Wencong
    Zhang, Lei
    Zhang, Xueying
    Yang, Yucheng
    Huang, Jianwei
    Li, Zhenxuan
    Zhou, Zhiyi
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2021, 14 : 8297 - 8315
  • [3] A non-local capsule neural network for hyperspectral remote sensing image classification
    Lei, Runmin
    Zhang, Chunju
    Du, Shihong
    Wang, Chen
    Zhang, Xueying
    Zheng, Hui
    Huang, Jianwei
    Yu, Min
    [J]. REMOTE SENSING LETTERS, 2021, 12 (01) : 77 - 86
  • [4] Multiscale Feature Aggregation Capsule Neural Network for Hyperspectral Remote Sensing Image Classification
    Lei, Runmin
    Zhang, Chunju
    Zhang, Xueying
    Huang, Jianwei
    Li, Zhenxuan
    Liu, Wencong
    Cui, Hao
    [J]. REMOTE SENSING, 2022, 14 (07)
  • [5] Spectral-spatial classification of hyperspectral remote sensing image based on capsule network
    Jia, Sen
    Zhao, Baojun
    Tang, Linbo
    Feng, Fan
    Wang, WenZheng
    [J]. JOURNAL OF ENGINEERING-JOE, 2019, 2019 (21): : 7352 - 7355
  • [6] Hyperspectral remote sensing image classification based on multidirectional adaptive aware network
    Liu, Qian
    Wu, Zebin
    Xu, Yang
    Zheng, Peng
    Zheng, Shangdong
    Wei, Zhihui
    [J]. National Remote Sensing Bulletin, 2024, 28 (01) : 168 - 186
  • [7] An Adaptive Artificial Immune Network for Supervised Classification of Multi-/Hyperspectral Remote Sensing Imagery
    Zhong, Yanfei
    Zhang, Liangpei
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2012, 50 (03): : 894 - 909
  • [8] Land scene classification for remote sensing images with an improved capsule network
    Chen, Xiao
    Zhu, Juan
    [J]. JOURNAL OF APPLIED REMOTE SENSING, 2022, 16 (02)
  • [9] Deep Feature Aggregation Network for Hyperspectral Remote Sensing Image Classification
    Zhang, Chunju
    Li, Guandong
    Lei, Runmin
    Du, Shihong
    Zhang, Xueying
    Zheng, Hui
    Wu, Zhaofu
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2020, 13 : 5314 - 5325
  • [10] Hyperspectral Remote Sensing Image Classification Based on Convolutional Neural Network
    Dai, Xiangyang
    Xue, Wei
    [J]. 2018 37TH CHINESE CONTROL CONFERENCE (CCC), 2018, : 10373 - 10377