Non-destructive quality evaluation of fruit by color based on RGB LEDs system

Abstract

Fruits and vegetables are generally valued based on their visual appearance, particularly their color. Color is an important quality attribute that affects consumer acceptance and preference in the food industry. Human perception of color has been used as an indirect measure for classifying fruits in indices as fruits matured and ripened. This study was conducted to measure fruit quality by using visible optical fiber sensors that contains of RGB LEDs which are red, green and blue with peak wavelengths at 635 nm, 525 nm, and 470 nm, respectively. Moreover, this study aims to provide an innovative and low-cost approach for nondestructive fruit quality measurement. Multiple linear regressions was applied to develop calibration models for classifying fruits based on the index. Results showed that the proposed optical instruments can produce good and accurate measurements when evaluating the index. The highest coefficient of determination (R2 = 0.879), was obtained by using a combination of red, green, and blue data sets at various ripeness indices. Meanwhile, the optical fiber of red system generated by the monochromatic wavelength exhibited better precision (R2 = 0.795) compared with the other two wavelengths. In conclusion, the application of these systems leads to the rapid and efficient assessment of the quality measurement of mangoes.