FOOD SCIENCE ›› 2026, Vol. 47 ›› Issue (20): 11-22.doi: 10.7506/spkx1002-6630-20260430-311

• Food Analysis and Detection Based on Spectroscopy and Chemometrics • Previous Articles    

Recent Advances in Spectroscopic Techniques Combined with Machine Learning for Nondestructive Tomato Quality Assessment

GUO Qi, YAN Xinhuan, ZHENG Xiaodong, TAN Mengnan, JIAO Min, SONG Ye, LIU Xuemei   

  1. 1. Jinan Fruit Research Institute, All China Federation of Supply & Marketing Co-operatives, Jinan 250220, China; 2. Shandong Province Fruit and Vegetable Storage and Processing Technology Innovation Center, Jinan 250220, China; 3. College of Life Sciences, Shandong Normal University, Jinan 250358, China
  • Published:2026-11-06

Abstract: Tomato quality directly determines its market value, commercial processing suitability, and consumer acceptance. Traditional manual or destructive detection methods are time-consuming and subjective, failing to meet the demands for efficient and nondestructive testing (NDT) in modern agriculture. Recently, with the development of artificial intelligence, the deep integration of spectroscopic techniques and machine learning has been progressively applied to the rapid and nondestructive assessment of tomato quality. This paper systematically reviews the current status of near-infrared (NIR) spectroscopy, hyperspectral imaging (HSI), spatially resolved spectroscopy (SRS), Raman spectroscopy, and fluorescence spectroscopy (FS) in this field. It focuses on the critical roles of machine learning in spectral data preprocessing, feature selection, and multimodal data fusion. Furthermore, the latest progress on spectroscopic techniques for evaluating the internal and external quality attributes of tomatoes, as well as the development of portable and online devices, is summarized to provide technical references for the intelligent quality inspection of tomatoes across the entire industry chain.

Key words: spectroscopic detection; machine learning; tomato quality; nondestructive testing; deep learning; data fusion

CLC Number: