Conditional variational autoencoder ensemble for noisy hyperspectral image classification

Youqiang Zhang

doi:10.1016/j.optlastec.2025.114602

Return

Conditional variational autoencoder ensemble for noisy hyperspectral image classification

2025-12-30

DOI：10.1016/j.optlastec.2025.114602

Original

Save

View PDF

Abstract

En 中文

• Propose a CVAE2 framework for the classification of noisy hyperspectral images. • Design a label-aware CVAE with ensemble learning to enhance stability. • Introduce a confidence degree that combines reconstruction error and classification confidence. • Achieve superior robustness on four benchmark datasets under varying noise levels.

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.