CRNSim: A New Similarity Index Capturing Global and Local Spectral Differences in Hyperspectral Data

• DOI: 10.54941/ahfe1005923
• Publication Date: 2025-02-17T04:41:21Z
• Authors: Jungkwon Kim; Sangmin Kim; Jungi Lee; Kwangsun Yoo; Seokjoo Byun
• Keywords: Similarity (geometry); Spectral Analysis; Spectral index

Hyperspectral imaging (HSI) enables detailed spectral analysis across numerous bands, offering transformative potential in diverse domains such as remote sensing, agriculture, and medical diagnostics. However, the inherent challenges of inter-class similarity, intra-class variability, limitations existing similarity metrics hinder its effectiveness. To address these challenges, we propose CRNSim, a novel index that integrates three complementary components: Chebyshev-based term to capture extreme deviations, RMSE-based account for global trends, nonlinear adjustment factor enhance sensitivity subtle variations while mitigating outlier influence. Experimental evaluations on benchmark hyperspectral datasets, including Indian Pines Salinas Valley, demonstrate superiority CRNSim improving separability, outperforming traditional Chebyshev RMSE. These findings highlight CRNSim’s advance HSI methodologies, making it robust tool fine-grained differentiation applications.