Image Analysis Technique for Evaluating the Quality of Navel Orange Fruits

Abstract

The goal of the research is to use an image processing system to evaluate the biochemical properties of orange fruits in a non-destructive, rapid and inexpensive way. Various biochemical properties such as TSS, acidity, TSS/TA, chlorophyll A, chlorophyll B and carotenoids were measured. The ENVI software package was used to analyze the orange fruit images and three RGB bands (red, green and blue) were derived for each image until R, G and B colors were obtained and then color indices such as GRVI, gn, bn, IKAW, ExR and Norm(VARI, VARI1). The results showed that there is a high correlation between color indicators and biochemical characteristics. The ExR gave the highest coefficient of determination with carotenoids (R^2= 0.94***), and also gave the highest coefficient of determination with TSS (R^2=0.90***). while the IKAW index gave the highest coefficient of determination with chlorophyll a and chlorophyll b (R^2=0.91*** and R^2=0.92***), respectively. The IKAW and ExR gave the highest coefficient of determination with TSS/TA (R^2= 0.917*** and R^2= 0.914***).