Titre :	Efficient plant disease classification using MobileNetV3 and transfer learning
Type de document :	document multimédia
Auteurs :	Oussama Zaidi, Auteur ; Laradj Chellama, Directeur de thèse
Editeur :	Laghouat : Université Amar Telidji - Département d'informatique
Année de publication :	2026
Importance :	54 p.
Accompagnement :	1 disque optique numérique (CD-ROM)
Note générale :	Option : Artificial intelligence and data science
Langues :	Anglais (eng)
Mots-clés :	Plant Disease Detection  Deep Learning  Transfer learning  MobileNetV3-Small  Convolutional neural networks (CNN)  Image classification  PlantVillage dataset  Precision agriculture
Résumé :	Plant diseases pose a significant threat to agricultural productivity and food security. Early and accurate disease detection is essential for minimizing crop losses and supporting effective agricultural management. This work presents a deep learning-based approach for plant disease classification using transfer learning. A MobileNetV3-Small model pretrained on ImageNet is adapted to classify plant leaf images into 38 healthy and diseased categories. The model is trained through a two-stage process consisting of feature extraction and fine-tuning. Experiments are conducted on the PlantVillage dataset, which contains approximately 87,000 images of plant leaves from multiple crop species. The proposed approach achieves a validation accuracy of 99.47%, demonstrating the effectiveness of transfer learning for large-scale plant disease classification. In addition, MobileNetV3 provides a lightweight and computationally efficient solution compared to its predecessor MobileNetV2, making it suitable for deployment on mobile and edge devices. However, evaluation on real-world images reveals limitations caused by differences in lighting conditions, backgrounds, and image quality compared to the training dataset.
note de thèses :	Mémoire de master en informatique

Efficient plant disease classification using MobileNetV3 and transfer learning [document multimédia] / Oussama Zaidi, Auteur ; Laradj Chellama, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2026 . - 54 p. + 1 disque optique numérique (CD-ROM).
Option : Artificial intelligence and data science
Langues : Anglais (eng)

Mots-clés :	Plant Disease Detection  Deep Learning  Transfer learning  MobileNetV3-Small  Convolutional neural networks (CNN)  Image classification  PlantVillage dataset  Precision agriculture
Résumé :	Plant diseases pose a significant threat to agricultural productivity and food security. Early and accurate disease detection is essential for minimizing crop losses and supporting effective agricultural management. This work presents a deep learning-based approach for plant disease classification using transfer learning. A MobileNetV3-Small model pretrained on ImageNet is adapted to classify plant leaf images into 38 healthy and diseased categories. The model is trained through a two-stage process consisting of feature extraction and fine-tuning. Experiments are conducted on the PlantVillage dataset, which contains approximately 87,000 images of plant leaves from multiple crop species. The proposed approach achieves a validation accuracy of 99.47%, demonstrating the effectiveness of transfer learning for large-scale plant disease classification. In addition, MobileNetV3 provides a lightweight and computationally efficient solution compared to its predecessor MobileNetV2, making it suitable for deployment on mobile and edge devices. However, evaluation on real-world images reveals limitations caused by differences in lighting conditions, backgrounds, and image quality compared to the training dataset.
note de thèses :	Mémoire de master en informatique