Identifying Coconut Maturity Levels Using CNN and YOLOv8 Deep Learning Algorithms

Abstract

To improve the efficiency and accuracy of determining coconut maturity levels in the processing industry, this study proposes an automated detection system employing Convolutional Neural Networks (CNN) and the You Only Look Once version 8 (YOLOv8) algorithm to classify maturity levels from image data. This study introduces an automated detection system using Convolutional Neural Networks (CNN) and the You Only Look Once version 8 (YOLOv8) algorithm to identify coconut maturity levels from image data. A dataset of 230 coconut images was utilized, classified into two categories: Young Coconut and Mature Coconut. The YOLOv8 model was trained and evaluated using standard object detection metrics, including mean Average Precision (mAP), precision, recall, and F1-score. The proposed model achieved a mAP of 90.5%, precision of 99.3%, recall of 94.2%, and F1-score of 96.6%, demonstrating high accuracy in detecting coconut maturity. This approach offers a practical and efficient alternative to manual assessment, contributing to improved accuracy and operational efficiency in agricultural practices.