Cross-domain CNN for Hyperspectral Image Classification

TL;DR

This paper introduces a novel cross-domain CNN that leverages shared and dataset-specific parameters to improve hyperspectral image classification across multiple datasets, addressing data scarcity issues.

Contribution

It is the first to propose an end-to-end CNN that jointly learns from multiple hyperspectral datasets using shared and non-shared parameters.

Findings

Outperforms baseline networks trained on single datasets

Effective handling of dataset-specific spectral characteristics

Addresses data scarcity in hyperspectral classification

Abstract

In this paper, we address the dataset scarcity issue with the hyperspectral image classification. As only a few thousands of pixels are available for training, it is difficult to effectively learn high-capacity Convolutional Neural Networks (CNNs). To cope with this problem, we propose a novel cross-domain CNN containing the shared parameters which can co-learn across multiple hyperspectral datasets. The network also contains the non-shared portions designed to handle the dataset specific spectral characteristics and the associated classification tasks. Our approach is the first attempt to learn a CNN for multiple hyperspectral datasets, in an end-to-end fashion. Moreover, we have experimentally shown that the proposed network trained on three of the widely used datasets outperform all the baseline networks which are trained on single dataset.