USING IMAGE PROCESSING TO EVALUATE THE QUALITY OF ORANGE FRUITS IN A NON-DESTRUCTIVE MANNER.

The aim of this research is to develop an image processing system that relies on machine vision to evaluate the chemical and physical properties in a non-destructive, fast and effective way to evaluate the quality of the orange fruits. Chemical and physical features such as TSS, titrated acidity, pH, TSS/T.acidity, liquid percentage, chlorophyll a, chlorophyll b, total chlorophyll and carotenoids were estimated. The results of the study showed that there is a relationship between the chemical and physical properties and the ripening of fruits. Relationships between R/G ratio range, G/R ratio, R/(R+G+B) range, NDVI1 index, VARI index, and VARI1 with some properties such as acidity, liquid percentage, pH, (TSS), TSS/T.Acidity, chlorophyll a, chlorophyll b as well as the concentration of carotenoids at different ripening days. Correlation coefficient and multiple regression analysis were obtained by testing the correlation between (TSS), acidity, TSS/T. acidity, chlorophyll a, chlorophyll b and carotenoids, and ratios R/G ratio, R/(R+G+B), NDVI index and VARI index and VARI1 for orange fruits. The results showed that the mean of the indices of VARI, VARI1, NDVI1, and the R/G range provided a better indicator of the concentrations (TSS), acidity, TSS/T. acidity, and chlorophyll a and b for orange fruits. R/(R + G + B) ratio gave the highest regression coefficient with carotenoids (R²=95***). while NDVI1 index gave the highest regression coefficient with chlorophyll a and chlorophyll b (R²=91***and R²=92***), respectively. [ABSTRACT FROM AUTHOR]