Original Article
Mansoureh Mozaffari Gonbari; Bahareh Jamshidi; Jaber Soleymani; Parisa zargaripour
Abstract
Monitoring potato quality attributes during storage is essential for optimizing supply chain management, ensuring timely product distribution, and minimizing postharvest losses. In this study, a non-destructive monitoring system for potato quality assessment was developed and evaluated using laser light ...
Read More
Monitoring potato quality attributes during storage is essential for optimizing supply chain management, ensuring timely product distribution, and minimizing postharvest losses. In this study, a non-destructive monitoring system for potato quality assessment was developed and evaluated using laser light backscattering imaging in the visible and near-infrared spectral regions. Backscattering images were acquired throughout the storage period under three storage conditions: cold storage at 4°C, cold storage at 7°C, and traditional storage. Two potato cultivars, Agria and Jelly, were included in the experiments. Key quality attributes, including moisture content, starch content, total soluble solids (TSS), and texture firmness, were measured periodically during storage. Features based on intensity and texture analyses of images were extracted. An artificial neural network model was employed to establish the relationship between the mentioned properties and image features. The results indicated that the 680 nm wavelength was the most effective for predicting moisture content, whereas the 880 nm wavelength provided superior performance for predicting firmness. The combination of both wavelengths performed well in predicting starch content and total soluble solid content. The highest correlation coefficients obtained for the prediction of moisture content, total soluble solids, firmness, and starch content were 0.76, 0.82, 0.71, and 0.73, respectively. These findings demonstrate the possibility of using laser backscattering imaging system for assessing the attributes of stored potatoes.
