Hyperspectral imaging technology for nondestructive identification of quality deterioration in fruits and vegetables: a review

With the increasing demand for high quality agri-food commodities, the issues of internal and external quality of fruits and vegetables have received widespread attention globally. To obtain the healthy fruits and vegetables, it is essential to develop advanced nondestructive detection technologies for identification of quality deterioration of target sample. Hyperspectral imaging (HSI) technology contains rich spectral and imaging information, which is capable of acquiring a detailed response of quality deterioration in fruits and vegetables. The review delves into the fundamental mechanism and damage type of quality deterioration caused by physical, chemical and biological factors within the domain of fruits and vegetables analysis. Various forms of deterioration encompassing surface defects, chilling injury, mechanical damage, wilting, browning, and microbial infection are summarized. Moreover, this overview also provides recent advances of HSI technology coupled with machine learning algorithms for quality evaluation and discrimination of different varieties fruits and vegetables. It also critically discusses the existing challenges and future prospects of the HSI technology in actual applications. Despite the extant limitations resulting from high-dimensional hyperspectral data and limited number of samples, the ongoing evolution of multi-sensor fusion architectures and artificial intelligence algorithms will promote HSI technology from laboratory to on-line monitoring in industrial applications.