Distinguishing Seed Cultivars of Quince (Cydonia oblonga Mill.) Using Models Based on Image Textures Built Using Traditional Machine Learning Algorithms

Different cultivars of seeds may have different properties. Therefore, distinguishing be important for seed processing and product quality. This study was aimed at revealing the usefulness innovative models developed based on selected image textures built using traditional machine algorithms cultivar classification quince seeds. The belonging to four ‘Uspiech’, ‘Leskovac’, ‘Bereczki’, ‘Kaszczenko’ were considered. In total, 1629 from color channels each extracted images acquired a flatbed scanner. Texture parameters used build combined set all channels, sets spaces RGB, Lab, XYZ, individual R, G, B, L, a, b, X, Y, Z groups. most successful Logistic (group Functions), IBk (Lazy), LogitBoost (Meta), LMT (Trees), naïve Bayes (Bayes). accuracy reached 98.75% in case model algorithm. For models, greatest misclassification cases observed between ‘Bereczki’ ‘Kaszczenko’. procedure can practice distinguish terms avoid mixing with properties intended further processing.