Abstract：Apples, sweet and sour, juicy and rich in nutrients, are one of the most popular fruits. In recent years, with the improvement of living standards, consumers' demands for apple quality are also increasing, especially for the improvement of internal quality. The external quality of apples, including color, size, peel damage, defects, etc., is the most direct feature that affects whether consumers buy or not, and the internal quality including soluble solids, titratable acid, hardness, etc., is essential to affect the nutritional value and sensory experience. Therefore, both internal and external qualities affect the market value of apples. However, before reaching consumers, apples must navigate through multiple stages including picking, storage, transportation, and sales. At these stages, apples are susceptible to pests and diseases, mechanical damage, and other adverse factors. Not only can these challenges decrease the internal and external qualities of apples, but they can also lead to serious financial losses. In addition, China is a major apple producer, but relevant data show that China's apple exports account for a relatively low amount, the main reasons for which are that the fruit quality sorting technology is not sophisticated enough, the degree of postharvest automation is low, and the fruit quality is not stable enough. Therefore, it is very important to detect and evaluate the internal and external quality of apples after harvest, which is not only conducive to improving the quality of apples, but also conducive to improving the market competitiveness of apples.Traditional quality detection methods rely mainly on manual and instrumental methods, including mass spectrometry, high performance liquid chromatography, refractometry, and direct observation method. However, although these methods are highly precise, they have the disadvantages of being time-consuming, destructive, and unable to detect them on a large scale, especially when detecting external qualities with the naked eye, which has a certain degree of subjectivity. Hence, there are great limitations in practical application. Therefore, in order to reduce the limitations of traditional methods in fruit quality detection, developing an accurate, rapid and non-destructive fruit quality analysis method for quality detection and grading is essential.In recent years, non-destructive detecting has been widely used in fruit quality detecting. At present, the commonly used non-destructive detecting technologies include near-infrared detection technology, fluorescence imaging detection technology, hyperspectral imaging technology, etc. Compared with the first two methods, hyperspectral imaging technology not only combines imaging technology with spectral technology, but can also obtain two-dimensional spatial information and one-dimensional spectral data at the same time, which consequently enables obtaining multiple dimension information with higher resolution. The image information obtained by hyperspectral imaging technology can be used to detect and evaluate the external quality, while the spectral information can be used to detect the internal quality. Therefore, hyperspectral imaging technology is expected to achieve non-destructive and accurate measurement and evaluation of the internal and external quality of apples. At present, various studies were reported on the application of hyperspectral imaging technology to the quality detection of apples, and the feasibility of this technology in the non-destructive detecting technology of apple quality has been preliminarily confirmed.In order to deeply explore the research progress of hyperspectral imaging technology in apple quality detection and make hyperspectral imaging technology more widely used, this paper first introduces the basic components of hyperspectral system, imaging principles, and common methods of data processing in research. Secondly, the application progress of hyperspectral imaging technology in assessing apple quality (both internal and external) was reviewed. Finally, the current challenges in the field of hyperspectral imaging are discussed, and the future direction of the more extensive and integrated application of this technology in the future is proposed.Research progress in the application of hyperspectral imaging technology in the internal and external quality of apples includes: (1) For internal quality, the technology can accurately quantify soluble solids, firmness, and moisture content, which are essential for assessing flavor and ripeness. However, there are few studies on the prediction of titratable acids using hyperspectral imaging, which may be due to their lower levels in apple fruits. Therefore, future work can consider combining multiple technologies for further research. (2) In terms of external quality, hyperspectral imaging can detect the shape, size, color, surface defects, contaminants, pest and disease infestations, and pesticide residues of apples by analyzing two-dimensional spatial information or combining image and spectral data, which is essential for post-harvest evaluation and grading. In addition, some studies have shown that hyperspectral imaging can distinguish between internal pests and diseases in transmission patterns, which is important to ensure consumer safety and satisfaction. Although many studies have confirmed the application prospects of hyperspectral imaging in apple quality detection, there are still some challenges in the application of this technology, such as the different data processing methods used in different origins and varieties, the low robustness of the model, the high cost of the instrument, and the transition from laboratory to actual field use. Therefore, future work can improve the accuracy of the model through the combination of multiple technologies and the development of more refined algorithms, so as to provide a better reference for non-destructive testing of apple quality.
PDF （）