Detecting maturity of persimmon fruit based on image processing technique

• The color-based grading of persimmon fruits using a vision system was studied.
• Physical, mechanical, and nutritional properties of fruits were evaluated.
• The written algorithm processed the images and extracted features for classification.
• The best performance was obtained for QDA classifier with overall accuracy rate of 90.24%.

Grading of persimmon fruits into three commercially maturity stages was conducted by image analysis technique. An automatic algorithm was developed to classify the fruits based on the external color of them. Physical, mechanical and nutritional properties of fruits were determined to compare the results of image analysis and visual classification. During the process of image segmentation, the black spots on persimmon fruits were removed to dilute the effect of them on the features to be extracted and used for classification. Among the features, there were significant differences between maturity stages for mean values of R, G, b*, gray scale and S channels. Two classifiers based on linear (LDA) and quadratic discriminant analysis (QDA) were used to assess the applicability of vision system. The results showed that QDA classifier could be valuable in categorizing the fruits with better overall accuracy rate of 90.24%.