Identifying Ripeness in Chokanan Mango Fruit Using K-Nearest Neighbor

• Published in: 2024 5th International Conference on Artificial Intelligence and Data Sciences, AiDAS 2024 - Proceedings
• Main Author: Syafiqah Noramli N.A.; Saidi R.M.; Ghazalli H.I.M.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85209639657&doi=10.1109%2fAiDAS63860.2024.10730537&partnerID=40&md5=90d3ea8df819094ce97bb428e950f235

Chokanan are popularly grown in Malaysia for both local consumption and export markets which is a sweet mango variety originating from Thailand, India, Bangladesh, and Pakistan. The evaluation of mango fruit maturity is crucial for quality control, supply chain management, and consumer satisfaction. However, existing methods suffer from several limitations, including time-consuming procedures, subjectivity, and potential inaccuracies. In this study, we utilize an approach using the K-Nearest Neighbor (KNN) algorithm to recognize the maturity level of Chokanan mango fruit. The proposed approach involves the extraction of relevant features from images of mango fruits, followed by the training of the KNN classifier using a labelled dataset. The trained model is then utilized to classify unseen mango fruit samples into different maturity classes. The experimental results indicate that the developed system shows strong potential in accurately recognizing the maturity level of mango fruits, achieving a high classification accuracy. While the system demonstrated a high level of accuracy in the tests conducted, further validation and testing are needed to confirm its effectiveness across broader conditions. These findings suggest promising potential for improving mango maturity recognition, reducing effort, and enhancing overall fruit quality assessment. © 2024 IEEE.