TY - GEN
T1 - Enhancing Hyperspectral Image Classification
T2 - 20th International Computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2023
AU - Butt, Imtiaz Ahmed
AU - Bo, Li
AU - Butt, Muhammad Hassaan Farooq
AU - Tassew, Megabiaw Tewodros
AU - Bin Akhtar Khan, Muhammad Usama
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - In the domain of hyperspectral imaging classification (HSIC) for remote sensing, recent advancements in deep learning have proven transformative. Convolutional neural networks have demonstrated immense potential in resolving these issues of HSIC, including insufficient labeled data, redundant spatial and spectral features, and overfitting. Traditional convolutional neural networks (CNNs) have effectively extracted spectral and spatial features, but 2D CNNs are limited in spatial modeling, while 3D CNNs alone struggle to distinguish spectral and spatial characteristics. Given that the accuracy of HSIC hinges on both spatial and spectral information, we proposed a hybrid-CNN model designed to mitigate the constraints of 2D and 3D CNNs. Our approach involves leveraging hybrid CNNs with spatial and channel attention (CA) mechanisms to address challenges like overfitting and model complexity. The proposed framework also improves generalization performance compared to 2D or 3D CNNs alone. Experiments were conducted on open datasets from Pavia University, Indian Pines, and Salinas to validate the proposed approach. The results demonstrate the efficacy of the hybrid CNN model with spatial and CA in consistently producing excellent classification results through thorough analyses with different deep-learning models.
AB - In the domain of hyperspectral imaging classification (HSIC) for remote sensing, recent advancements in deep learning have proven transformative. Convolutional neural networks have demonstrated immense potential in resolving these issues of HSIC, including insufficient labeled data, redundant spatial and spectral features, and overfitting. Traditional convolutional neural networks (CNNs) have effectively extracted spectral and spatial features, but 2D CNNs are limited in spatial modeling, while 3D CNNs alone struggle to distinguish spectral and spatial characteristics. Given that the accuracy of HSIC hinges on both spatial and spectral information, we proposed a hybrid-CNN model designed to mitigate the constraints of 2D and 3D CNNs. Our approach involves leveraging hybrid CNNs with spatial and channel attention (CA) mechanisms to address challenges like overfitting and model complexity. The proposed framework also improves generalization performance compared to 2D or 3D CNNs alone. Experiments were conducted on open datasets from Pavia University, Indian Pines, and Salinas to validate the proposed approach. The results demonstrate the efficacy of the hybrid CNN model with spatial and CA in consistently producing excellent classification results through thorough analyses with different deep-learning models.
KW - Channel Attention
KW - Classification
KW - Convolutional Neural Networks
KW - Remote sensing
KW - Spectral Spatial
UR - http://www.scopus.com/inward/record.url?scp=85184661631&partnerID=8YFLogxK
U2 - 10.1109/ICCWAMTIP60502.2023.10387100
DO - 10.1109/ICCWAMTIP60502.2023.10387100
M3 - 会议稿件
AN - SCOPUS:85184661631
T3 - 2023 20th International Computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2023
BT - 2023 20th International Computer Conference on Wavelet Active Media Technology and Information Processing, ICCWAMTIP 2023
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 15 December 2023 through 17 December 2023
ER -