WITHDRAWN: An Intelligent Computer Vision Framework for Automated Tur Dal Variety Identification Using Machine Learning

Abstract This research work presents a computer vision-based approach to identify and classify tur dal varieties using ReseNet50, Support Vector Machine (SVM) and Random Forest(RF). A dataset of tur dal images consisting of different varieties is captured using a high-resolution camera. The color, morphological and texture features are extracted from the images. The classifiers namely ReseNet50, SVM and RF and are employed in the classification process. The accuracies of RestNet50, SVM and RF are found to be 93.5%, 91% and 90% respectively. The accuracy of ResNet50 is found to be highest when compared with the other two models. In this work, a methodology for the automated and cost-effective identification and categorization of tur dal varieties is developed.