Application of Hyperspectral Imagers in Non-destructive Testing of Fruit and Vegetable Quality (Internal and External)

The quality of fruits and vegetables is a crucial indicator for quality control and a key aspect of quality and safety management. Traditional testing methods are cumbersome, time-consuming, and can cause significant damage to the samples. Hyperspectral imaging technology can be used to rapidly, non-destructively, qualitatively, and quantitatively test the quality of fruits and vegetables. This article introduces the application of hyperspectral imagers in non-destructive testing of the internal and external quality of fruits and vegetables.

The Importance of Fruit and Vegetable Quality Testing: Fruits and vegetables are indispensable foods, providing humans with essential vitamins, fiber, minerals, and trace elements. The quality and safety of fruits and vegetables have always been a hot topic of social concern. The external quality of fruits and vegetables is primarily assessed based on their color, texture, size, shape, and surface defects. External quality is their most direct quality characteristic, directly impacting market price and consumer purchasing desire. Internal quality mainly depends on their sugar content, acidity, firmness, soluble solids content, starch content, moisture content, maturity, and the content of other nutrients. Internal quality reflects their value. Quality and safety are primarily assessed through the detection of foreign contamination such as animal feces, various diseases, quality deterioration, bacterial infection, and pesticide residues. This is related to consumer food safety and health and is the most important quality characteristic of fruits and vegetables.

Currently, the quality and safety of fruits and vegetables are mainly tested using traditional chemical methods, which are time-consuming, labor-intensive, and destructive testing techniques. With the rapid development of imaging and spectroscopy technologies, hyperspectral imaging technology has been widely applied to the rapid, non-destructive testing of agricultural product quality and safety. Numerous successful cases have proven that hyperspectral imaging technology is a scientifically effective tool for food and agricultural product quality and safety testing. Hyperspectral imaging technology integrates the advantages of traditional imaging and spectral techniques, acquiring hyperspectral images with the characteristic of "image-spectrum integration," meaning they simultaneously contain image and spectral information. Image information can be used to detect the external quality of fruits and vegetables, while spectral information can be used to detect their internal quality and safety.

Applications of Hyperspectral Imagers in Non-destructive Testing of Internal and External Quality of Fruits and Vegetables: As a non-invasive, non-contact, and non-traditional technology, hyperspectral imaging provides spatial and spectral information of the research object. Its key characteristics include high spectral resolution; a wide spectral response range with numerous narrow bands; "spectrum-image integration"; large data volume and rich information; and multiple data description models for flexible analysis. These characteristics determine the unique advantages of hyperspectral imaging technology in the internal and external quality testing of food. Because image data reflects the external characteristics of a product, while spectral data can analyze the internal physical structure and chemical composition of an object, hyperspectral imaging technology can be considered a perfect combination of image and spectral techniques. The data generated by hyperspectral imaging technology can be vividly described using "three-dimensional data blocks," making it more reliable than traditional machine vision or spectral techniques. Its applications in fruit and vegetable quality testing are as follows:

1. Non-destructive testing of fruits and vegetables

Traditional fruit and vegetable quality testing methods cause some damage to the fruits and vegetables and require significant time, manpower, and resources, failing to meet the demands of modern fruit and vegetable quality testing. Hardness and soluble solids content are important intrinsic indicators determining the maturity and harvest time of fruits and vegetables. In addition, important fruit and vegetable quality indicators include taste, appearance, aroma, and chemical composition. Based on hyperspectral imaging technology, the determination of these indicators requires only a small number of samples and can achieve rapid, non-destructive testing of relevant quality indicators. Testing personnel only need to extract a "three-dimensional" image block from the model sample, including two-dimensional image pixel information and third-dimensional wavelength information. Then, a multivariate analysis model is established, correlating the extracted spectral data with the measured values of sample attributes to establish a quantitative relationship model. This allows for quality testing and grading of fruit and vegetable samples, and effective evaluation of fruit and vegetable quality.

2. Detection of contaminants on the surface of fruits and vegetables

Research shows that when testing personnel irradiate fruits and vegetables with light of different wavelengths, the fruits and vegetables exhibit phenomena such as scattering, absorption, reflection, and transmission of light. Studies have shown that when light of specific wavelengths shines into the interior of fruits and vegetables, the spectral data collected through transmission and scattering carries rich information about the food's internal components. In recent years, hyperspectral technology has been applied to detect damage to fruits and vegetables. Hyperspectral imaging allows for non-destructive testing while rapidly and accurately acquiring complete image and spectral information of samples. Inspectors use image and spectral analysis to detect the chemical composition, physical structure, and surface features of fruits and vegetables. Within the visible and near-infrared spectral range of 400 to 1000 nanometers, by combining exposure time, scanning speed, and spectral correction, inspectors can acquire "three-dimensional" image blocks, including two-dimensional image pixel information and third-dimensional wavelength information. Because hyperspectral data has multispectral channels, high spectral resolution, and continuous spectrum, it can distinguish two very similar but different images at different wavelengths, and can obtain continuous spectral curves for any pixel and different spectral curves for different substances. This technology utilizes the significant differences between the spectral values of normal and damaged areas at certain specific wavelengths to achieve non-destructive testing of fruit and vegetable surfaces.

3. Detection of Pesticide Residues on Fruit and Vegetable Surfaces

In practice, there are various pesticide residue detection technologies. Traditional high-precision detection processes require highly skilled operators and are time-consuming, limiting their application to precise laboratory analysis and detection of pesticide residues. Chemical detection methods are generally destructive, consuming organic reagents, involving cumbersome sample preparation, and incurring high costs. Hyperspectral imaging detection, however, achieves efficient, real-time, rapid, and non-destructive detection. Operators obtain hyperspectral images of food using a hyperspectral analyzer, apply principal component analysis to analyze these images, and identify images at characteristic wavelengths. Image processing techniques are then used to analyze these characteristic images, allowing for the detection of pesticide residue levels on the surface of fruits and vegetables.