Non-Destructive Discrimination and Traceability of Exocarpium Citrus grandis Aging Years via Feature-Optimized Hyperspectral Imaging and Broad Learning System
Year: 2025
Type: article
Abstract: Exocarpium Citrus grandis is a traditional Chinese medicinal and edible herb whose pharmacological efficacy is closely tied to its aging duration. The accurate discrimination of aging years is essential for quality control but remains challenging due to limitations in current analytical techniques. This study proposes a novel feature-optimized classification framework that integrates hyperspectral imaging (HSI) with a Broad Learning System (BLS). Bilateral spectral data (side A and side B) were collected to capture more comprehensive sample information. A combination of normalization (NOR) preprocessing and the Iterative Variable Importance for Spectral Subset Selection Algorithm (iVISSA) was found to be optimal. The NOR–iVISSA–BLS model achieved classification accuracies of 94.09 ± 1.01% (side A) and 95.10 ± 0.82% (side B). Furthermore, cross-validation between the two sides (A→B: 94.92%, B→A: 94.11%) confirmed the model’s robustness and generalizability. This dual-side spectral validation strategy offers a rapid, nondestructive, and reliable solution for the vintage authentication of Exocarpium Citrus grandis, contributing to the modernization of quality control in medicinal foodstuffs.
Source: Photonics
Authors: Wenqi Liu, Shihua Zhong
Institution: Hunan Normal University
Language: English
Subfield: Analytical Chemistry
Field: Chemistry
Domain: Physical Sciences
SDG: Zero hunger
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W4412517531