Rotten Fruit Detection Using Artificial Intelligence

Abstract

Detecting rotten fruits has become crucial in the agricultural industry as it ensures the separation of spoiled produce from fresh ones, thus maintaining trust and credibility among consumers. Machine learning and Artificial Intelligence are rich in algorithms. These algorithms help in various fields. Unhealthy fruits may cause damage to the other healthy fruits if not classified properly and can also affect productivity. The above issue has been discussed in this paper. A Convolutional Neural Network (CNN) based Deep Learning Technique was proposed to enhance the prediction model for identifying rotten fruits. This model aims to classify fruit images as either fresh or rotten. The study focused on three fruit types: Apple, Banana, and Oranges. A dataset consisting of fresh and rotten fruits was collected and used for training, validation, and testing purposes. The CNN architecture was employed to build the predictive model. The dataset that is downloaded from Kaggle is evaluated for the performance of the proposed model and produces an accuracy of 96.88%. By leveraging the power of CNNs, this improved model can effectively analyze and classify fruit images, providing accurate predictions regarding their freshness or spoilage. Such advancements contribute to the agricultural industry's ability to deliver high-quality products to consumers and reduce potential health risks associated with consuming rotten fruits.