Few-shot Unknown Class Discovery of Hyperspectral Images with Prototype Learning and Clustering

2025-08-25

AI Summary

To address the fundamental challenge in open-set few-shot hyperspectral image classification—namely, the inability to recognize unknown classes—this paper proposes a novel framework integrating prototype learning with unsupervised clustering. Leveraging only a minimal number of labeled samples from known classes, the method jointly constructs prototypes for known classes and infers latent prototypes for unknown classes, enabling fine-grained discovery and partitioning of unseen categories. A unified metric-learning objective jointly models distances between known and unknown prototypes, allowing simultaneous rejection of unknown-class samples and their internal clustering. Extensive experiments on four benchmark hyperspectral datasets demonstrate that the proposed approach significantly outperforms existing open-set few-shot classification methods, achieving consistent improvements in both unknown-class detection and overall classification accuracy. The source code is publicly available.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Representation Learning for Vision
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Open-set few-shot hyperspectral image (HSI) classification aims to classify image pixels by using few labeled pixels per class, where the pixels to be classified may be not all from the classes that have been seen. To address the open-set HSI classification challenge, current methods focus mainly on distinguishing the unknown class samples from the known class samples and rejecting them to increase the accuracy of identifying known class samples. They fails to further identify or discovery the unknow classes among the samples. This paper proposes a prototype learning and clustering method for discoverying unknown classes in HSIs under the few-shot environment. Using few labeled samples, it strives to develop the ability of infering the prototypes of unknown classes while distinguishing unknown classes from known classes. Once the unknown class samples are rejected by the learned known class classifier, the proposed method can further cluster the unknown class samples into different classes according to their distance to the inferred unknown class prototypes. Compared to existing state-of-the-art methods, extensive experiments on four benchmark HSI datasets demonstrate that our proposed method exhibits competitive performance in open-set few-shot HSI classification tasks. All the codes are available at href{https://github.com/KOBEN-ff/OpenFUCD-main} {https://github.com/KOBEN-ff/OpenFUCD-main}

Problem

• Discovering unknown classes in hyperspectral images with few labeled samples
• Distinguishing unknown from known classes using prototype learning
• Clustering rejected unknown samples into new classes via inferred prototypes

Innovation

• Prototype learning for distinguishing known and unknown classes
• Inferring unknown class prototypes from few labeled samples
• Clustering rejected samples using distance to inferred prototypes