Stochastic Neighbor Embedding Feature-Based Hyperspectral Image Classification Using 3D Convolutional Neural Network

被引:2
作者
Hossain, Md. Moazzem [1 ,2 ]
Hossain, Md. Ali [1 ]
Miah, Abu Saleh Musa [2 ,3 ]
Okuyama, Yuichi [3 ]
Tomioka, Yoichi [3 ]
Shin, Jungpil [3 ]
机构
[1] Rajshahi Univ Engn & Technol, Dept Comp Sci & Engn, Rajshahi 6204, Bangladesh
[2] Bangladesh Army Univ Sci & Technol, Dept Comp Sci & Engn, Saidpur 5311, Bangladesh
[3] Univ Aizu, Sch Comp Sci & Engn, Aizu Wakamatsu 9658580, Japan
关键词
hyperspectral image; dimensionality reduction; visualization; image classification; principle component analysis; t-Distributed Stochastic Neighbor Embedding; blended convolutional neural network; REDUCTION;
D O I
10.3390/electronics12092082
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The ample amount of information from hyperspectral image (HSI) bands allows the non-destructive detection and recognition of earth objects. However, dimensionality reduction (DR) of hyperspectral images (HSI) is required before classification as the classifier may suffer from the curse of dimensionality. Therefore, dimensionality reduction plays a significant role in HSI data analysis (e.g., effective processing and seamless interpretation). In this article, a sophisticated technique established as t-Distributed Stochastic Neighbor Embedding (tSNE) following the dimension reduction along with a blended CNN was implemented to improve the visualization and characterization of HSI. In the procedure, first, we employed principal component analysis (PCA) to reduce the HSI dimensions and remove non-linear consistency features between the wavelengths to project them to a smaller scale. Then we proposed tSNE to preserve the local and global pixel relationships and check the HSI information visually and experimentally. Lastly, it yielded two-dimensional data, improving the visualization and classification accuracy compared to other standard dimensionality-reduction algorithms. Finally, we employed deep-learning-based CNN to classify the reduced and improved HSI intra- and inter-band relationship-feature vector. The evaluation performance of 95.21% accuracy and 6.2% test loss proved the superiority of the proposed model compared to other state-of-the-art DR reduction algorithms.
引用
收藏
页数:19
相关论文
共 62 条
  • [1] Abadi M., 2016, arXiv, DOI 10.48550/arXiv.1603.04467
  • [2] [Anonymous], 2010, HYPERSPECTRAL IMAGIN, P9, DOI [10.1016/B978-0-12-374753-2.10019-X, DOI 10.1016/B978-0-12-374753-2.10019-X]
  • [3] Semisupervised Hyperspectral Image Classification Using Deep Features
    Aydemir, M. Said
    Bilgin, Gokhan
    [J]. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2019, 12 (09) : 3615 - 3622
  • [4] A Pigment Analysis Tool for Hyperspectral Images of Cultural Heritage Artifacts
    Bai, Di
    Messinger, David W.
    Howell, David
    [J]. ALGORITHMS AND TECHNOLOGIES FOR MULTISPECTRAL, HYPERSPECTRAL, AND ULTRASPECTRAL IMAGERY XXIII, 2017, 10198
  • [5] Classification of Hyperspectral Images With Regularized Linear Discriminant Analysis
    Bandos, Tatyana V.
    Bruzzone, Lorenzo
    Camps-Valls, Gustavo
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2009, 47 (03): : 862 - 873
  • [6] Bartholomew DJ., 2010, International Encyclopedia of Education, P374, DOI DOI 10.1016/B978-0-08-044894-7.01358-0
  • [7] 3-D Deep Learning Approach for Remote Sensing Image Classification
    Ben Hamida, Amina
    Benoit, Alexandre
    Lambert, Patrick
    Ben Amar, Chokri
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2018, 56 (08): : 4420 - 4434
  • [8] A Fast and Compact Hybrid CNN for Hyperspectral Imaging-based Bloodstain Classification
    Butt, Muhammad Hassaan Farooq
    Ayaz, Hamail
    Ahmad, Muhammad
    Li, Jian Ping
    Kuleev, Ramil
    [J]. 2022 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2022,
  • [9] Chang C.I., 2013, Hyperspectral Data Processing:Algorithm Design and Analysis
  • [10] Chao Bu, 2020, Journal of Physics: Conference Series, V1693, DOI 10.1088/1742-6596/1693/1/012100