Strawberry Freshness Assessment by Hyperspectral Imaging

Background. Strawberry is a highly valued and perishable food item. The freshness of these fruits plays a crucial role in their quality, as it determines their shelf life, nutritional content, visual appeal, and safety for human consumption. Traditional methods of assessing fruit freshness are subjective, labor-intensive, and have low productivity. This study aims to develop a methodology for quantitatively assessing the freshness of strawberries using hyperspectral imaging, which can provide objective and accurate measurements of fruit quality. Materials and method. During the research, we evaluated the spectral properties of the outer surface and internal structure of strawberries from "Remontant Elizabeth II" over a period of 26 days after harvesting. The measuring instrument used was an acousto-optical Vis-NIR imaging spectrometer. Digital data processing involved preprocessing spectral images, morphological analysis, and calculating a quantitative metric for spectral reflectance at the most informative wavelengths. Statistical analysis was based on constructing regression models to determine the post-harvest period for strawberries. Model evaluation was done using the coefficients of determination (R2), relative error (RE), and root mean squared error (RMSE). Results. The methodology for assessing the freshness of strawberries using hyperspectral imaging has been proposed. Mathematical models for determining the post-harvest period of "Remontant Elizaveta II" strawberries were obtained using hyperspectral images of the surface and internal structure of the samples. Analysis of the spectral properties of the external surface of fruits showed higher accuracy in determining the postharvest period, with , and .. Regression models with different polynomial orders were assessed, and the cubic polynomial showed the greatest effectiveness. A set of the most informative wavelengths was determined, based on which multiple regression analysis was performed, demonstrating the highest accuracy. Conclusion. The developed methodology for quantitative analysis of strawberry freshness stands out for its precision, objectivity, efficiency, and automation. Assessment of individual stages, including sample preparation, hyperspectral imaging, digital data processing, and statistical analysis will be beneficial to advance methods for spectral diagnostics of food products. Proposed approach could supplement traditional methods of food quality control. Research could be used to develop optimal strategies for transportation, processing, storage and marketing of strawberry batches. EDN: EMYEPG