Vis/NIR Spectroscopy and Vis/NIR Hyperspectral Imaging for Non-Destructive Monitoring of Apricot Fruit Internal Quality with Machine Learning

Foods, Journal Year: 2025, Volume and Issue: 14(2), P. 196 - 196

Published: Jan. 10, 2025

The fruit supply chain requires simple, non-destructive, and fast tools for quality evaluation both in the field during post-harvest phase. In this study, a portable visible near-infrared (Vis/NIR) spectrophotometer Vis/NIR hyperspectral imaging (HSI) device were tested to highlight genetic differences among apricot cultivars, develop multi-cultivar multi-year models most important marketable attributes (total soluble solids, TSS; titratable acidity, TA; dry matter, DM). To do this, fruits of seventeen cultivars from single experimental orchard harvested at commercial maturity stage considered. Spectral data emphasized similarities capturing changes pigment content macro components samples. recent years, machine learning techniques, such as artificial neural networks (ANNs), have been successfully applied more efficiently extract valuable information spectral accurately predict traits. prediction developed based on multilayer perceptron network (ANN-MLP) combined with Levenberg-Marquardt algorithm. Regarding dataset, good predictive performances achieved TSS (R2 = 0.855) DM 0.857), while performance TA was unsatisfactory 0.681). contrast, optimal ability found HSI dataset (TSS: R2 0.904; DM: 0.918, TA: 0.811), confirmed by external validation. Moreover, ANN allowed us identify input regions each model. results showed potential spectroscopy an alternative traditional destructive methods monitor qualitative traits fruits, reducing time costs analyses.

Language: Английский

Shining light on seaweed—the utilization of vibrational spectroscopy and machine learning in the seaweed industry

International Journal of Food Science & Technology, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 13, 2025

Abstract Seaweed and macroalgae have been utilized for centuries in human animal nutrition due to their rich composition functional properties. As global demand sustainable food sources grows, the seaweed industry requires effective quality control systems ensure product safety consistency. Vibrational spectroscopy, including near-infrared (NIR), mid-infrared (MIR), Raman offers powerful techniques analysing molecular of seaweed. These methods enable identification characterization key structures, essential ensuring seaweed-based products. The integration machine learning (ML) chemometric enhances analytical capabilities vibrational providing robust tools data interpretation decision-making sustainability. This review highlights recent advancements application learning, practices within industry, emphasizing role improving quality, traceability, safety, resource efficiency. Furthermore, ability IR spectroscopy predict chemical biomass production under different abiotic conditions is discussed. Developing implementing will agile that support management risk evaluation systems, with objective measurements identify hazards during post-harvest processing.

Language: Английский