Deep Feature Extraction and Classification of Hyperspectral Images Based on Convolutional Neural Networks

TLDR

The extracted deep features are useful for image classification and target detection. The study proposes a regularized deep feature extraction method using a 3‑D CNN with combined regularization and a virtual sample enhancement to improve hyperspectral image classification. The method uses multiple convolutional and pooling layers to extract nonlinear, discriminant, invariant deep features, applies L2 regularization and dropout to mitigate overfitting, and employs a virtual sample enhancement, evaluated on Indian Pines, University of Pavia, and Kennedy Space Center datasets. The models with sparse constraints achieve competitive performance compared to state‑of‑the‑art methods, indicating that the proposed deep feature extraction opens avenues for further research.

Abstract

Due to the advantages of deep learning, in this paper, a regularized deep feature extraction (FE) method is presented for hyperspectral image (HSI) classification using a convolutional neural network (CNN). The proposed approach employs several convolutional and pooling layers to extract deep features from HSIs, which are nonlinear, discriminant, and invariant. These features are useful for image classification and target detection. Furthermore, in order to address the common issue of imbalance between high dimensionality and limited availability of training samples for the classification of HSI, a few strategies such as L2 regularization and dropout are investigated to avoid overfitting in class data modeling. More importantly, we propose a 3-D CNN-based FE model with combined regularization to extract effective spectral-spatial features of hyperspectral imagery. Finally, in order to further improve the performance, a virtual sample enhanced method is proposed. The proposed approaches are carried out on three widely used hyperspectral data sets: Indian Pines, University of Pavia, and Kennedy Space Center. The obtained results reveal that the proposed models with sparse constraints provide competitive results to state-of-the-art methods. In addition, the proposed deep FE opens a new window for further research.