Review of advances in hyper-spectral imaging-based non-destructive testing methods in food application
Sha Wu, Lianfei Huang, Renshuai Huang
Abstract
Ensuring food quality and safety is a global priority, and non-destructive testing (NDT) plays a critical role in monitoring products throughout the supply chain. Hyperspectral imaging (HSI), originally developed for remote sensing, has emerged over the past two decades as a powerful NDT tool for food analysis by integrating two-dimensional imaging with spectroscopy to acquire both spatial and spectral information at the pixel level. This review provides a comprehensive and application-oriented overview of HSI-based NDT methods in food science and food quality and safety control. First, the fundamental principles, sensing and acquisition modes, and typical hardware configurations of HSI systems are summarized. Second, key steps in hyperspectral data processing—including image calibration, spectral preprocessing, feature extraction, and model construction—are discussed, with particular emphasis on recent algorithmic advances such as deep learning-based feature learning, spectral–spatial modeling, and multisource data fusion. Subsequently, representative applications of HSI are critically reviewed in three major domains: food composition analysis (moisture, protein, fat, carbohydrates, and functional components), quality assessment (appearance, texture, freshness, and storage stability), and food safety (pesticide residues, heavy metals, mycotoxins, microbial spoilage, adulteration, and traceability). Finally, current challenges and future perspectives are outlined, including the need for standardised acquisition and validation protocols, miniaturised and low-cost HSI devices, integration with Internet of Things (IoT) technologies for supply chain monitoring, and open benchmark datasets to support robust deep learning models. By synthesising methodological advances and recent applications, this review aims to clarify how HSI can contribute to reliable, efficient, and non-destructive monitoring of food quality and safety.