Self Supervised Learning for Few Shot Hyperspectral Image Classification
This addresses the impracticality of manual labeling in hyperspectral imaging, offering a more efficient solution for domain-specific applications.
The paper tackles the problem of limited labeled data for hyperspectral image classification by using self-supervised learning with Barlow-Twins pre-training, resulting in models that outperform vanilla supervised learning with only a few labels.
Deep learning has proven to be a very effective approach for Hyperspectral Image (HSI) classification. However, deep neural networks require large annotated datasets to generalize well. This limits the applicability of deep learning for HSI classification, where manually labelling thousands of pixels for every scene is impractical. In this paper, we propose to leverage Self Supervised Learning (SSL) for HSI classification. We show that by pre-training an encoder on unlabeled pixels using Barlow-Twins, a state-of-the-art SSL algorithm, we can obtain accurate models with a handful of labels. Experimental results demonstrate that this approach significantly outperforms vanilla supervised learning.