Adaptive Granulation-Based Convolutional Neural Networks With Single Pass Learning for Remote Sensing Image Classification

被引:1
|
作者
Pal, Sankar K. K. [1 ]
Kumar, Dasari Arun [1 ,2 ]
机构
[1] Indian Stat Inst, Ctr Soft Comp Res, Kolkata, West Bengal, India
[2] KSRM Coll Engn, Dept Elect & Commun Engn, Kadapa 516005, Andhra Pradesh, India
关键词
Deep adaptive granulation; fuzzy rough feature selection; image classification; pixel uncertainty; remote sensing; roughness measure; weighted class-membership; MULTILAYER PERCEPTRON; INFORMATION; FUSION; ALGORITHM; SETS; SAR;
D O I
10.1109/JSTARS.2022.3223180
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
Convolutional neural networks (CNNs) with the characteristics like spatial filtering, feed-forward mechanism, and back propagation-based learning are being widely used recently for remote sensing (RS) image classification. The fixed architecture of CNN with a large number of network parameters is managed by learning through a number of iterations, and, thereby increasing the computational burden. To deal with this issue, an adaptive granulation-based CNN (AGCNN) model is proposed in the present study. AGCNN works in the framework of fuzzy set theoretic data granulation and adaptive learning by upgrading the network architecture to accommodate the information of new samples, and avoids iterative training, unlike conventional CNN. Here, granulation is done both on the 2-D input image and its 1-D representative feature vector output, as obtained after a series of convolution and pooling layers. While the class-dependent fuzzy granulation on input image space exploits more domain knowledge for uncertainty modeling, rough set theoretic reducts computed on them select only the relevant features for input to CNN. During classification of unknown patterns, a new principle of roughness-minimization with weighted membership is adopted on overlapping granules to deal with the ambiguous cases. All these together improve the classification accuracy of AGCNN, while reducing the computational time significantly. The superiority of AGCNN over some state-of-the-art models in terms of different performance metrics is demonstrated for hyperspectral and multispectral images both quantitatively and visually.
引用
收藏
页码:57 / 70
页数:14
相关论文
共 50 条
  • [1] Remote Sensing Image Scene Classification Based on Convolutional Neural Networks
    Liu, Yumei
    Informatica (Slovenia), 2025, 49 (09): : 45 - 54
  • [2] FULLY CONVOLUTIONAL NEURAL NETWORKS FOR REMOTE SENSING IMAGE CLASSIFICATION
    Maggiori, Emmanuel
    Tarabalka, Yuliya
    Charpiat, Guillaume
    Alliez, Pierre
    2016 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2016, : 5071 - 5074
  • [3] Adaptive Granular Neural Networks for Remote Sensing Image Classification
    Kumar, D. Arun
    Meher, Saroj K.
    Kumari, K. Padma
    IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2018, 11 (06) : 1848 - 1857
  • [4] Selective convolutional neural networks and cascade classifiers for remote sensing image classification
    Wan, Lihong
    Liu, Na
    Huo, Hong
    Fang, Tao
    REMOTE SENSING LETTERS, 2017, 8 (10) : 917 - 926
  • [5] FEATURE SPARSITY IN CONVOLUTIONAL NEURAL NETWORKS FOR SCENE CLASSIFICATION OF REMOTE SENSING IMAGE
    Huang, Wei
    Wang, Qi
    Li, Xuelong
    2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2019), 2019, : 3017 - 3020
  • [6] Pruning Convolutional Neural Networks with an Attention Mechanism for Remote Sensing Image Classification
    Zhang, Shuo
    Wu, Gengshen
    Gu, Junhua
    Han, Jungong
    ELECTRONICS, 2020, 9 (08) : 1 - 19
  • [7] Hyperspectral Remote Sensing Image Classification Based on Convolutional Neural Network
    Dai, Xiangyang
    Xue, Wei
    2018 37TH CHINESE CONTROL CONFERENCE (CCC), 2018, : 10373 - 10377
  • [8] A review of Convolutional Neural Networks in Remote Sensing Image
    Liu, Xinni
    Han, Fengrong
    Ghazali, Kamarul Hawari
    Mohamed, Izzeldin Ibrahim
    Zhao, Yue
    2019 8TH INTERNATIONAL CONFERENCE ON SOFTWARE AND COMPUTER APPLICATIONS (ICSCA 2019), 2019, : 263 - 267
  • [9] REMOTE SENSING IMAGE CLASSIFICATION BASED ON CONVOLUTIONAL NEURAL NETWORKS WITH TWO-FOLD SPARSE REGULARIZATION
    Liu, Han
    He, Lin
    Li, Jun
    2017 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS), 2017, : 992 - 995
  • [10] Bayesian Convolutional Neural Networks for Limited Data Hyperspectral Remote Sensing Image Classification
    Joshaghani, Mohammad
    Davari, Amirabbas
    Hatamian, Faezeh Nejati
    Maier, Andreas
    Riess, Christian
    IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2023, 20