With the improvement of people's living standards, consumers are paying more and more attention to the quality and safety of fruits and vegetables. For example, fruit damage not only causes the decay of fruits and vegetables but also seriously affects the health of consumers. Therefore, rapid and effective detection of fruit damage is highly practical. Although the damaged areas of fruits may look similar to the normal areas in appearance, there are subtle changes in the damaged regions that can be detected through spectral analysis at specific wavelengths.
Hyperspectral image technology combines the advantages of spectral analysis and image processing to detect and analyze both internal and external quality characteristics of objects. In this study, hyperspectral imaging is used to identify damaged areas of apples, aiming for accurate and efficient detection.
**2. Test Materials and Methods**
**2.1 Experimental Materials**
In this study, apples were selected as the research object to analyze their decayed areas. The decayed areas on the apples were naturally formed.
**2.2 Experimental Equipment**
The hyperspectral imaging data was collected using the GaiaSorter hyperspectral sorting system from Sichuan Shuangli Hepu Technology Co., Ltd. The system includes a hyperspectral imager (V10E), CCD camera, light source, black box, computer, and other components. The key parameters of the system are listed in Table 1.
**Table 1: GaiaSorter Hyperspectral Sorter System Parameters**
| No. | Item | Parameter |
|-----|--------------------|-----------------|
| 1 | Spectral scanning range | 350–1000 nm |
| 2 | Spectral resolution | 2.8 nm |
| 3 | Collection interval | 1.9 nm |
| 4 | Number of spectral channels | 520 |
**2.3 Image Processing and Analysis**
Hyperspectral data was preprocessed using SpecView and ENVI/IDL software. This included mirror transformation and black-and-white frame calibration. The processed data was then analyzed using ENVI/IDL to extract meaningful information.
**3. Results and Discussion**
**3.1 Spectral Analysis of Apple Decay and Normal Areas**
A total of 200 pixels were sampled from both the decayed and normal areas of the apple. The average spectral reflectance was calculated and compared. As shown in Figure 2, the red line represents the spectral reflectance of the decayed area, the blue line represents the normal area, and the green line represents the pesticide residue area. It was observed that the spectral reflectance of the pesticide residue area was the highest in the range of 400–1000 nm, followed by the normal area, and then the decayed area. All three regions showed a peak at 610 nm, a valley at 650 nm, and a steep slope between 650–680 nm. These features are considered unique to apples.
**3.2 Extraction of Apple Decay Area**
By analyzing the spectral differences between the apple and background areas, a calibrated hyperspectral image was used to generate a pure apple image using band calculations in ENVI/IDL. Principal component analysis was then applied, and the PC2 component was selected to distinguish the decayed area from the normal area. Threshold segmentation was performed to extract the decayed region.
**3.5 Discussion**
The application of hyperspectral imaging technology in detecting fruit surface damage has proven its advantage of "spatial-spectral integration." Subtle damages often occur beneath the skin, making them difficult to detect with the naked eye. Over time, these areas may darken and eventually cause the entire fruit to rot, potentially affecting other fruits. The results of this study show that using hyperspectral imaging combined with principal component analysis and thresholding techniques can effectively detect fruit damage, enabling rapid and accurate identification. This method holds great promise for improving food safety and quality control in the agricultural industry.
Outdoor Fibre Termination Box,Fiber Optic Distribution Cabinet,Wall Mount Fiber Distribution Box,Distribution Box Fiber Optic
Ningbo Fengwei Communication Technology Co., Ltd , https://www.fengweicommunication.com