Abstract

One of the critical challenges across several industry sectors is fruit grading and classification. One of the most popular fruits in food products is the orange fruit because there are many value-added products produced from it. Due to the shared traits of the varieties with various tastes and uses, even in major markets, It would be challenging at times to instantly label numerous classifications of this fruit when there is a need for success. Three varieties of orange fruit that are currently popular in the Iraqi marketplace were studied in this research. We propose a fast and cost-effective way to automate the classification of this fruit, we suggest an effective classification framework system based on the most recent deep learning algorithms and clustering approaches. To determine the most effective model for classifying this fruit, we trained, evaluated, and contrasted the results within many deep-learning models. The accuracy was assessed to be 99.1%, indicating the reliability of the suggested method.