Using a machine learning-based hyperspectral image classification method for stray light pollution level assessment

doi:10.1016/j.optlastec.2025.113935

Using a machine learning-based hyperspectral image classification method for stray light pollution level assessment

2025-09-22

0

PRE

AI

DOI：10.1016/j.optlastec.2025.113935

Original

Original request for help

Save

Abstract

En 中文

• Propose a classification-centered framework for assessing the degree of stray light pollution in hyperspectral images. • Develop the RCDC-WJSaCR algorithm to address limited and imbalanced training samples in classification. • Introduce a pollution assessment index (PAI) map to realize visual distribution of pollution degrees. • Adopt independently acquired experimental hyperspectral images as the dataset for experimental verification. • Validate the effectiveness of the assessment framework through experiments.

Using a machine learning-based hyperspectral image classification method for stray light pollution level assessment

Abstract

Journal

Organization