DeepLeaf: an optimized deep learning approach for automated recognition of grapevine leaf diseases
Neural Computing and Applications · 2025-05 · 37 (15) · 8799-8823
DOI: 10.1007/s00521-025-11038-3 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neural Computing and Applications, online erschienen am 28.02.2025. Verfasst von Fatma M. Talaat und 3 weiteren Autor:innen. Die Arbeit wurde bisher 32-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Kafr El Shiekh University.
Abstract (Originalsprache)
Plant diseases can cause severe losses in agricultural production, impacting food security and safety. Early detection of plant diseases is crucial to minimize crop damage and ensure agricultural sustainability. Manual monitoring is often impractical due to the complexity and time involved, making automated disease recognition essential. This study presents a new Plant Disease Detection Algorithm (PDDA) called DeepLeaf focused on identifying four common grapevine diseases: leaf blight, black rot, stable, and black measles. The PDDA integrates three key modules: an Image Preprocessing Module, a Feature Extraction Module, and an Optimized Convolutional Neural Network (OCNN)-based Classification Module. The OCNN forms the core of the classification system, with its hyperparameters fine-tuned using fuzzy optimization to enhance performance. Preprocessing techniques are applied to analyze diseased leaves, and a logistic regression algorithm is used to downsample the features for better analysis. The CNN is trained on images from the Plant Village dataset, allowing it to detect and classify grapevine leaf diseases accurately. The proposed model's efficiency in the automated diagnosis of grapevine diseases is demonstrated by its remarkable 99.7% accuracy rate. This high accuracy indicates that the PDDA may help with more effective and scalable plant disease monitoring, which will ultimately allow better agricultural practices.
Förderung
- Kafr El Shiekh University
Zitieren
APA
Talaat, F. M., Shams, M. Y., Gamel, S. A., & ZainEldin, H. (2025). DeepLeaf: an optimized deep learning approach for automated recognition of grapevine leaf diseases. Neural Computing and Applications, 37(15), 8799-8823. https://doi.org/10.1007/s00521-025-11038-3
Vancouver
Talaat FM, Shams MY, Gamel SA, ZainEldin H. DeepLeaf: an optimized deep learning approach for automated recognition of grapevine leaf diseases. Neural Computing and Applications. 2025;37(15):8799-8823. doi:10.1007/s00521-025-11038-3
BibTeX
@article{Talaat2025,
author = {Talaat, Fatma M. and Shams, Mahmoud Y. and Gamel, Samah A. and ZainEldin, Hanaa},
title = {DeepLeaf: an optimized deep learning approach for automated recognition of grapevine leaf diseases},
journal = {Neural Computing and Applications},
year = {2025},
volume = {37},
number = {15},
pages = {8799--8823},
doi = {10.1007/s00521-025-11038-3}
}
EUVIMED