Iranian wheat varieties classification by using a fusion of texture features

Wheat is one of the important nutritional products in agriculture. Planting a specific variety in each region depends on the climatic conditions of that region and farm efficiency. Therefore the classification of different varieties is one of the most important challenges for producers. For this purpose, various methods of image texture extraction have been presented, and each method has a specific accuracy. In order to use all the extracted features and modeling based on them, in this research, the Particle Swarm Optimization (PSO) method was used. For this purpose, using 34 algorithms for extracting texture features of 7 varieties of Iranian wheat, 3519 features were extracted and modeled with Linear Discriminate Analysis (LDA), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) modeling methods. In the following, using PSO method, the amount of accuracy improvement of each modeling method was extracted and compared. The results of the research showed that the PSO method can increase the accuracy of different modeling methods up to 24% and improve the performance of the classifier.