DOI QR코드

DOI QR Code

Hyperspectral Image Classification using EfficientNet-B4 with Search and Rescue Operation Algorithm

  • S.Srinivasan (School of Computer Science and Engineering, VIT University) ;
  • K.Rajakumar (School of Computer Science and Engineering, VIT University)
  • Received : 2023.12.05
  • Published : 2023.12.30

Abstract

In recent years, popularity of deep learning (DL) is increased due to its ability to extract features from Hyperspectral images. A lack of discrimination power in the features produced by traditional machine learning algorithms has resulted in poor classification results. It's also a study topic to find out how to get excellent classification results with limited samples without getting overfitting issues in hyperspectral images (HSIs). These issues can be addressed by utilising a new learning network structure developed in this study.EfficientNet-B4-Based Convolutional network (EN-B4), which is why it is critical to maintain a constant ratio between the dimensions of network resolution, width, and depth in order to achieve a balance. The weight of the proposed model is optimized by Search and Rescue Operations (SRO), which is inspired by the explorations carried out by humans during search and rescue processes. Tests were conducted on two datasets to verify the efficacy of EN-B4, with Indian Pines (IP) and the University of Pavia (UP) dataset. Experiments show that EN-B4 outperforms other state-of-the-art approaches in terms of classification accuracy.

Keywords

References

  1. Ghosh P. Deep Learning to Diagnose Diseases and Security in 5G Healthcare Informatics. InMachine Learning and Deep Learning Techniques for Medical Science 2022 (pp. 279-331). CRC Press.
  2. Kwon HE, Kim JN, Kwon MJ, Lee JR, Kim SC, Nam JH, Kim BJ. The traditional medicine bojungikki-tang increases intestinal motility. Pharmacognosy Magazine. 2021 Jan 1;17(5):1.
  3. Manoharan, J. S., Sakthivel, T. G., Sangeetha, M., Sivamani. & Md. Saad Patel. (2022). A hybrid fuzzy based cross neighbor filtering (HF-CNF) for image enhancement of fine and coarse powder scanned electron microscopy (SEM) images. Journal of Intelligent and Fuzzy Systems, 42(6), 6159-6169. doi:10.3233/JIFS-212561.
  4. Wang Y, Mao H, Xu G, Zhang X, Zhang Y. A Rapid Detection Method for Fungal Spores from Greenhouse Crops Based on CMOS Image Sensors and Diffraction Fingerprint Feature Processing. Journal of Fungi. 2022 Apr 6;8(4):374.
  5. Li L, Li W, Du Q, Tao R. Low-rank and sparse decomposition with mixture of Gaussian for hyperspectral anomaly detection. IEEE Trans Cybern 2021;51(9):4363e72.
  6. Xu X, Zhao M, Yang J, Xiong Y, Pang F, Tan Z, Luo M. 3D laser scanning strategy based on cascaded deep neural network. Def Techno 2021.
  7. Chen, Y.; Zhu, K.; Zhu, L.; He, X.; Ghamisi, P.; Benediktsson, J.A. Automatic design of convolutional neural network for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 2019, 57, 7048-7066. https://doi.org/10.1109/TGRS.2019.2910603
  8. Huang, H.; Duan, Y.; He, H.; Shi, G. Local linear spatial-spectral probabilistic distribution for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 2020, 58, 1259-1272. https://doi.org/10.1109/TGRS.2019.2945255
  9. Li Y, Zhang H, Shen Q. Spectral-spatial classification of hyperspectral imagery with 3D convolutional neural network. Rem Sens 2017;9(1):67.
  10. Zhu L, Chen Y, Ghamisi P, Benediktsson J. Generative adversarial networks for hyperspectral image classification. IEEE Trans Geosci Rem Sens 2018;56(9): 5046e63.
  11. Roy, S.K.; Krishna, G.; Dubey, S.R.; Chaudhuri, B.B. HybridSN: Exploring 3-D-2-D CNN feature hierarchy for hyperspectral image classification. IEEE Geosci. Remote Sens. Lett. 2020, 17, 277-281. https://doi.org/10.1109/LGRS.2019.2918719
  12. Zhu, L.; Chen, Y.; Ghamisi, P.; Benediktsson, J.A. Generative adversarial networks for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 2018, 56, 5046-5063. https://doi.org/10.1109/TGRS.2018.2805286
  13. Paoletti, M.E.; Haut, J.M.; Fernandez-Beltran, R.; Plaza, J.; Plaza, A.J.; Pla, F. Deep pyramidal residual networks for spectral-spatial hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 2019, 57, 740-754. https://doi.org/10.1109/TGRS.2018.2860125
  14. Zhu, M.; Jiao, L.; Liu, F.; Yang, S.; Wang, J. Residual spectral-spatial attention network for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 2021, 59, 449-462. https://doi.org/10.1109/TGRS.2020.2994057
  15. Ma, W.; Yang, Q.; Wu, Y.; Zhao, W.; Zhang, X. Double-branch multiattention mechanism network for hyperspectral image classification. Remote Sens. 2019, 11, 1307.
  16. Li, R.; Zheng, S.; Duan, C.; Yang, Y.; Wang, X. Classification of Hyperspectral Image Based on Double-Branch Dual-Attention Mechanism Network. Remote Sens. 2020, 12, 582.
  17. Cui, Y.; Yu, Z.; Han, J.; Gao, S.; Wang, L. Dual-Triple Attention Network for Hyperspectral Image Classification Using Limited Training Samples. IEEE Geosci. Remote Sens. Lett. 2021, 19, 1-5.
  18. Roy, S.K.; Manna, S.; Song, T.; Bruzzone, L. Attention-Based Adaptive Spectral Spatial Kernel ResNet for Hyperspectral Image Classification. IEEE Trans. Geosci. Remote Sens. 2020, 59, 7831-7843. https://doi.org/10.1109/TGRS.2020.3043267
  19. P. Zhong B. Du L. Zhang S. Wan, C. Gong and J. Yang. Multiscale dynamic graph convolutional network for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens., 58(5):3162-3177, 2019. https://doi.org/10.1109/TGRS.2019.2949180
  20. Sun, W., Yang, G., Peng, J. and Du, Q., 2019. Lateral-slice sparse tensor robust principal component analysis for hyperspectral image classification. IEEE Geoscience and Remote Sensing Letters, 17(1), pp.107-111. https://doi.org/10.1109/LGRS.2019.2915315