Differentiation of peach cultivars by image analysis based on the skin, flesh, stone and seed textures

Abstract The peaches belonging to different cultivars can be characterized by differentiation in properties. aim of this study was evaluate the usefulness individual parts fruit (skin, flesh, stone and seed) for cultivar discrimination based on textures determined using image analysis. Discriminant analysis performed classifiers Bayes net, logistic, SMO, multi-class classifier random forest a set combined selected from all color channels R, G, B, L, a, b, X, Y, Z separately RGB, Lab XYZ spaces. In case sets (R, Z), accuracy 100% observed stones seeds classifiers. RGB space produced correctness equal flesh peaches. spaces, slightly lower accuracies than were obtained reaching noted only research proved texture parameters successful peach with an 100%. distinguishing between may important breeders, consumers industry ensuring adequate processing conditions equipment parameters. identification human large errors. molecular or chemical methods require special time-consuming. ensure objective, rapid relatively inexpensive procedure high discrimination.