Portable Visible–Near Infrared Spectroscopy for the Nondestructive Evaluation of Soluble Solids Contents and Firmness in Peaches

Read full text
Authors: Fengqing Tian, Yangyang Dang, Xiaolong Zhang, Wenxiao Zhang, Zhaohui Hao
Cited by: 1
References: 38

Abstract

The soluble solids content (SSC) and firmness are two key parameters that determine edible quality and maturity of peaches; therefore, they have garnered significant attention from researchers in the peach industry. To enable rapid and nondestructive detection of SSC and firmness in peaches, a portable device based on visible–near infrared (Vis/NIR) spectroscopy was developed during this study. The device consists of three main components: a spectral acquisition module; a control and display module; and a power management module. Additionally, a corresponding data processing program was designed for this device. Using the developed instrument, diffuse reflectance spectra of peaches were collected. After spectral preprocessing to eliminate interference, feature wavelength selection algorithm was used for dimensionality reduction, and partial least squares regression (PLSR) model was established for SSC and firmness prediction. For the prediction set, the coefficients of determination ( ) for SSC and firmness were 0.884 and 0.852, the root mean square errors of prediction (RMSEPs) were 0.466°Brix and 5.532 N, and the residual predictive deviations (RPDs) were 2.938 and 2.602, respectively. The optimal models were embedded into the device to evaluate its stability and accuracy using an independent validation set. The results indicated that the average coefficients of variation for SSC and firmness measurements are 3.808% and 4.415%. The independent validation set yielded of 0.854 and 0.817 and the RMSEPs of 0.512°Brix and 5.659 N, respectively. Overall, the portable device developed for nondestructive multi-quality detection of peaches demonstrated stability and accuracy that fully meet the requirements of on-site and real-time measurements.

Fields of Study & Subjects

Spectroscopy and Chemometric Analyses Smart Agriculture and AI Spectroscopy Techniques in Biomedical and Chemical Research

Publication details

Journal / source
HortScience
Publisher
American Society for Horticultural Science
Volume
61
Issue
4
Pages
727–736
Publication date
5 March 2026