Disease Detection in Cocoa Fruit Using YoloV4 Convolutional Neural Network Architecture
Journal of Electronics Technology Exploration · 08.02.2026 · 3 (2) · 87-94
DOI: 10.52465/joetex.v3i2.648 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Electronics Technology Exploration, online erschienen am 08.02.2026. Verfasst von Farihah Farihah und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Detecting diseases in cocoa beans is important to prevent a decline in crop quality and economic losses. This study aims to detect diseases in cocoa beans using the YOLOv4 Convolutional Neural Network (CNN) architecture. The dataset used consists of images of cocoa beans, which then undergo preprocessing, data division, data augmentation, and modeling. The test results show that the YOLOv4 model is capable of detecting diseases in cocoa beans with an accuracy rate of 97%, demonstrating good performance in the classification and object detection process. However, this study still has limitations in terms of the amount of data and the variety of image capture conditions. Therefore, further research is expected to use a larger dataset and test the model in more diverse environmental conditions to improve the reliability of the system.
Zitieren
APA
Farihah, F., & Luthfie, A. A. (2026). Disease Detection in Cocoa Fruit Using YoloV4 Convolutional Neural Network Architecture. Journal of Electronics Technology Exploration, 3(2), 87-94. https://doi.org/10.52465/joetex.v3i2.648
Vancouver
Farihah F, Luthfie AA. Disease Detection in Cocoa Fruit Using YoloV4 Convolutional Neural Network Architecture. Journal of Electronics Technology Exploration. 2026;3(2):87-94. doi:10.52465/joetex.v3i2.648
BibTeX
@article{Farihah2026,
author = {Farihah, Farihah and Luthfie, Alfaiz Alafi},
title = {Disease Detection in Cocoa Fruit Using YoloV4 Convolutional Neural Network Architecture},
journal = {Journal of Electronics Technology Exploration},
year = {2026},
volume = {3},
number = {2},
pages = {87--94},
doi = {10.52465/joetex.v3i2.648}
}
EUVIMED