Investigating the qualitative properties of sugar beet using hyperspectral imaging

In recent decades, various non-destructive technologies have been used to evaluate the quality of fruits and vegetables. Among these methods, hyperspectral imaging has been noticed by researchers as a non-destructive, fast and minimal environmental pollution method to evaluate the properties of agricultural products. The purpose of the research was to investigate the effect of variety, duration and storage conditions on the quality properties of sugar beet and also the feasibility of using hyperspectral images as a new and non-destructive technique in measuring quality properties. Three sugar beet cultivars including Agrit, Muriel and Shokofa cultivars in three different storage conditions (1- in storage conditions with protection 2- without protection i.e., in atmospheric conditions 3- subsoil and the same field conditions) for 45 days of storage and then ultraspectral imaging in the mode Reflectance and in the wavelength range of 400 to 1100 nm was measured once every 15 days. In the next step, the properties of the samples (including soluble solids (SSC), mechanical properties, sugar content, moisture percentage and pH) were determined based on destructive methods. Because the acquired spectra have noise, the effects of disturbing factors were minimized through various pre-processing to provide clearer data for modeling the qualitative characteristics of the samples. Finally, modeling was done using the resulting data from hyperspectral imaging and reference tests based on the least square regression (PLS) method, and in the last step, after validating the resulting models, the most appropriate model was selected. The results show that the hyperspectral imaging method can be used as a valuable tool to predict the quality properties of sugar beet during the storage period.