Research on image processing and machine learning of some species of Onobrychis genus

This thesis aims to classify the fruits of the Onobrychis (sainfoin) plant, which grows in Turkey, using Artificial Neural Networks (ANN), providing fundamental information on ANNs and Local Binary Patterns (LBP). In the study, a database was created using a total of 448 fruit images from four different Onobrychis species (O. cappadocica, O. argyrea, O. hypargyrea, and O. tournefortii). Six different methods were used to classify the sainfoin fruit varieties: Support Vector Machine, Naive Bayes, Decision Trees, K-Nearest Neighbor, Artificial Neural Networks, and Logistic Regression. The performances of these methods were compared, and the Support Vector Machine method was identified as the most suitable model with an accuracy rate of 99.6%.