Evaluation of Bruised Plum Quality Using Laser Induced Diffuse Reflectance Imaging

Abstract

Mechanical damage affects the quality of harvested fruit. The damage makes bruising and results in losses of economic value during storage and postharvest handling. This study aimed to evaluate the effect of mechanical damage on plum as well as develop prediction models of soluble solids content (SSC) and firmness. Plum samples of cultivar “Topend” were subjected to 0.722 J ± 0.087 J mechanical impact. Bruised and control samples were stored at 1°C, 10°C, and 22°C. A low power line laser module of 1 mW and 635 nm was used to induce backscattering on the fruit surface. Three parameters were extracted from the signal: amplitude, shape factor (gradient sharpness), and full width at half maximum. All of them achieved significant correlation (p < 0.001) with the reference parameters. Support Vector Machine (SVM) was applied to develop models. Among the available kernel functions, the radial kernel obtained the best accuracy with increased weight coefficient (γ). The cross validation with bootstrapping confirmed the applicability of the proposed technique with the R² = 0.870 and RMSE = 0.66% for SSC and the R² = 0.895 and RMSE = 1.89 N for firmness. The promising performance of laser induced backscattering imaging suggests its application in quality evaluation of plum fruit. Since the line laser can be utilized to scan the fruit surface, it opens the opportunity to new developments toward spatial analysis.