Convolutional Neural Network-Based Xception, MobileNetV2 and InceptionV3 Models for Plant Disease Identification in Sub-Saharan Africa
Year: 2025
Type: article
Abstract: Plant disease identification in Sub-Saharan Africa poses a significant challenge, hindered by costly laboratory tests or subjective visual assessments.Recent advances in image-based disease identification show promise, but existing methods are limited in accuracy and efficiency.This study addresses these shortcomings by presenting a convolutional neural network (CNN)-based plant disease classifier, leveraging transfer learning from pre-trained models Xception, MobileNetV2, and InceptionV3.A high generalization rate of 98.76% is achieved in the test data, demonstrating the potential for efficient and accurate identification of plant disease.This research contributes to innovative agricultural management solutions in Sub-Saharan Africa, with implications for improving crop yields, food security, and sustainable agriculture.
Authors: Aminou Halidou, Youssoufa Mohamadou, Pascalin Tiam Apen, Daramy Vandi Von Kallon, William John Baraza, Mbouna Gildas Patrick, Djiembou Tientcheu Victor Nico, Robndoh Mardochée
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: diamond
OpenAlex ID: W4409840922