Automated skin defect identification system for orange fruit grading based on genetic algorithm

Using machine vision technology to grade oranges can ensure that only good-quality fruits are exported. One of the most prominent issues in the post-harvest processing of oranges is the efficient determination of skin defects with the intention of classifying the fruits depending on their external appearance. Shape, size, colour and texture are the important grading parameters that dictate the quality and value of many fruit products. The accuracy of the evaluation results is increased by proper combination of different grading parameters. This article presents an efficient orange surface grading system (normal and defective) based on the colour and texture features. As a part of the feature selection step, this article presents a wrapper approach with genetic algorithm to search out and identify the informative feature subset for classification. The selected features were subjected to various classifiers such as support vector machine, back propagation neural network and auto associative neural network (AANN) to study the performance analysis among these three classifiers. The results reveal that AANN classification algorithm has the highest accuracy rate of 94.5% among these three classifiers.