Evaluating the Efficacy of Convolutional Neural Networks: Disease Detection in Crops
Authors:
- Stephen Ardey MensahSMStephen Ardey Mensah
- Peace Korshiwor AmoateyPAPeace Korshiwor Amoatey
- Peter Amoako OforiPOPeter Amoako Ofori
- Stella Owusu-NketiaSOStella Owusu-Nketia
- Edward Benjamin SabiESEdward Benjamin Sabi
- Frank Agyen DwomohFDFrank Agyen Dwomoh
- Anthony Raphael SimpsonASAnthony Raphael Simpson
- Evans AsensoEvans Asenso
Open Access
- 1 January 2026
- journal article
- Published by Wiley in Advances in Agriculture
- Vol. 2026 (1)
Abstract
Crop diseases pose a significant risk to agricultural productivity, particularly in sub‐Saharan Africa, where cassava ( Manihot esculenta ), plantain ( Musa paradisiaca ), and cocoa ( Theobroma cacao ) serve as essential crops. There is an urgent need for innovative plant disease detection methods to enhance control measures and preventive strategies to address this menace, as these diseases threaten food security and economic stability in the growing regions. Image recognition, powered by deep learning (DL) models, offers a cost‐effective and scalable technology for disease detection. Thus, technological convergence has led to significant improvements in AI‐assisted plant disease diagnosis. This review evaluates the efficacy of convolutional neural networks (CNNs) for disease detection in these crops, focusing on model architecture, classification capability, and performance metrics such as accuracy, precision, recall, and F1‐score. A comparative analysis of CNN models and machine learning techniques highlights their robustness in detecting diseases like Black Sigatoka, Cassava Mosaic Disease, and Black Pod Disease. Additionally, this study examines the current challenges and limitations in crop disease detection and explores future directions, including the potential for transfer learning and expanded datasets to enhance model accuracy and generalizability. The goal is to improve crop disease detection accuracy, which is a critical step in ensuring the agricultural sector’s viability. The study expects that by greatly increasing detection rates, a strong basis for better disease management techniques will be established, guaranteeing the productivity and health of crops and, consequently, bolstering the financial stability of farming communities that depend on crop production. Results highlight CNNs as a strong foundation for early crop disease identification, supporting the advancement of precision agriculture applications in the region.Keywords
This publication has 103 references indexed in Scilit:
- Performance Evaluation of CNN, ResNet50, and Hybrid Architectures for Automated Grape Leaf Disease DetectionApplied Fruit Science, 2026
- An Effective Approach for Recognition of Crop Diseases Using Advanced Image Processing and YOLO v8Food Science & Nutrition, 2026
- Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agricultureArtificial Intelligence Review, 2025
- Plant disease management: a fine-tuned enhanced CNN approach with mobile app integration for early detection and classificationArtificial Intelligence Review, 2024
- A review of convolutional neural networks in computer visionArtificial Intelligence Review, 2024
- Revolutionizing crop disease detection with computational deep learning: a comprehensive reviewEnvironmental Monitoring and Assessment, 2024
- Image‐based detection and classification of plant diseases using deep learning: State‐of‐the‐art reviewUrban Agriculture & Regional Food Systems, 2024
- Deep learning modelling techniques: current progress, applications, advantages, and challengesArtificial Intelligence Review, 2023
- Convolutional Neural Networks2022
- Advanced Concepts for Intelligent Vision Systems2020