BASS Net: Band-Adaptive Spectral-Spatial Feature Learning Neural Network for Hyperspectral Image Classification
This addresses landcover classification in hyperspectral images, which is important for remote sensing applications, but appears incremental as it builds on existing deep learning methods.
The authors tackled hyperspectral image classification by proposing BASS Net, an end-to-end deep learning architecture that extracts band-specific spectral-spatial features, which outperformed the highest reported accuracies on popular datasets.
Deep learning based landcover classification algorithms have recently been proposed in literature. In hyperspectral images (HSI) they face the challenges of large dimensionality, spatial variability of spectral signatures and scarcity of labeled data. In this article we propose an end-to-end deep learning architecture that extracts band specific spectral-spatial features and performs landcover classification. The architecture has fewer independent connection weights and thus requires lesser number of training data. The method is found to outperform the highest reported accuracies on popular hyperspectral image data sets.