BASS Net: Band-Adaptive Spectral-Spatial Feature Learning Neural Network for Hyperspectral Image Classification

TL;DR

This paper introduces BASS Net, a deep learning model that adaptively learns spectral-spatial features for hyperspectral image classification, addressing high dimensionality and limited labeled data challenges.

Contribution

It proposes a novel band-adaptive spectral-spatial neural network that requires fewer parameters and outperforms existing methods on standard hyperspectral datasets.

Findings

Outperforms state-of-the-art accuracy on hyperspectral datasets

Requires fewer training samples due to fewer connection weights

Effectively captures spectral-spatial features for landcover classification

Abstract

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.