Classification of Fresh and Rotten Fruits in an iOS-Based Application for the Visually Impaired
Repository Politeknik Negeri Batam
Date
2024-07-15
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Visually impaired people face various challenges, including difficulties in preparing food, particularly in selecting fresh fruits, which are essential for nutritional needs. iOS-based aplication using machine learning can assist in distinguishing between fresh and rotten fruits. This research aims to develop an iOS application to classify fresh and rotten fruits for the visually impaired using Create ML for model development. The classification model, built with Create ML, is integrated into the iOS application. The model and aplication's performance is evaluated using accuracy, precision, recall, and F1 score metrics. The model achieved 99% accuracy in training, 96% in validation, and 96% in testing. When tested with 100 samples from the fruit dataset, the application achieved an overall accuracy of 93%. Testing with images outside the dataset resulted in 88% accuracy. This application aims to help visually impaired individuals identify fresh and rotten fruits, enhancing their independence.
Description
Keywords
TECHNOLOGY
Citation
IEEE