Consolidated Convolutional Neural Network for Hyperspectral Image Classification

• Published in: Remote sensing (Basel, Switzerland) Switzerland), 2022-04, Vol.14 (7), p.1571
• Main Authors: Chang, Yang-Lang, Tan, Tan-Hsu, Lee, Wei-Hong, Chang, Lena, Chen, Ying-Nong, Fan, Kuo-Chin, Alkhaleefah, Mohammad
• Format: Article
• Language: English
• Subjects: Accuracy; Artificial neural networks; Band spectra; Classification; Complexity; Computer applications; Computing time; consolidated convolutional neural network; high performance computing; hyperspectral image classification; Hyperspectral imaging; image augmentation; Image classification; Machine learning; Neural networks; principal component analysis; Principal components analysis; Redundancy; Regularization; Remote sensing; Spatial analysis; Spatial discrimination; Spatial resolution; Spectral bands
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs14071571

The performance of hyperspectral image (HSI) classification is highly dependent on spatial and spectral information, and is heavily affected by factors such as data redundancy and insufficient spatial resolution. To overcome these challenges, many convolutional neural networks (CNN) especially 2D-CNN-based methods have been proposed for HSI classification. However, these methods produced insufficient results compared to 3D-CNN-based methods. On the other hand, the high computational complexity of the 3D-CNN-based methods is still a major concern that needs to be addressed. Therefore, this study introduces a consolidated convolutional neural network (C-CNN) to overcome the aforementioned issues. The proposed C-CNN is comprised of a three-dimension CNN (3D-CNN) joined with a two-dimension CNN (2D-CNN). The 3D-CNN is used to represent spatial–spectral features from the spectral bands, and the 2D-CNN is used to learn abstract spatial features. Principal component analysis (PCA) was firstly applied to the original HSIs before they are fed to the network to reduce the spectral bands redundancy. Moreover, image augmentation techniques including rotation and flipping have been used to increase the number of training samples and reduce the impact of overfitting. The proposed C-CNN that was trained using the augmented images is named C-CNN-Aug. Additionally, both Dropout and L2 regularization techniques have been used to further reduce the model complexity and prevent overfitting. The experimental results proved that the proposed model can provide the optimal trade-off between accuracy and computational time compared to other related methods using the Indian Pines, Pavia University, and Salinas Scene hyperspectral benchmark datasets.