Abstract:Plant disease detection is a critical task in agriculture, essential for ensuring crop health and productivity. Traditional methods in this context are often labor-intensive and prone to errors, highlighting the need for automated solutions. While computer vision-based solutions have been successfully deployed in recent years for plant disease identification and localization tasks, these often operate independently, leading to suboptimal performance. It is essential to develop an integrated solution combining these two tasks for improved efficiency and accuracy. This research proposes the innovative Plant Disease Localization and Classification model based on Vision Transformer (PDLC-ViT), which integrates co-scale, co-attention, and cross-attention mechanisms and a ViT, within a Multi-Task Learning (MTL) framework. The model was trained and evaluated on the Plant Village dataset. Key hyperparameters, including learning rate, batch size, dropout ratio, and regularization factor, were optimized through a thorough grid search. Early stopping based on validation loss was employed to prevent overfitting. The PDLC-ViT model demonstrated significant improvements in plant disease localization and classification tasks. The integration of co-scale, co-attention, and cross-attention mechanisms allowed the model to capture multi-scale dependencies and enhance feature learning, leading to superior performance compared to existing models. The PDLC-ViT model evaluated on two public datasets achieved an accuracy of 99.97%, a Mean Average Precision (MAP) of 99.18%, and a Mean Average Recall (MAR) of 99.11%. These results underscore the model's exceptional precision and recall, highlighting its robustness and reliability in detecting and classifying plant diseases. The PDLC-ViT model sets a new benchmark in plant disease detection, offering a reliable and advanced tool for agricultural applications. Its ability to integrate localization and classification tasks within an MTL framework promotes timely and accurate disease management, contributing to sustainable agriculture and food security.

Deep learning for mango leaf disease identification: A vision transformer perspective

Crop Disease Identification by Fusing Multiscale Convolution and Vision Transformer.

Vision transformer meets convolutional neural network for plant disease classification

A Fresh Look at Tomato Leaf Disease Recognition using Vision Transformers

Aggregating Different Scales of Attention on Feature Variants for Tomato Leaf Disease Diagnosis from Image Data: A Transformer Driven Study

Transform and Deep Learning Algorithms for the Early Detection and Recognition of Tomato Leaf Disease

A deep learning based approach for automated plant disease classification using vision transformer

Explainable vision transformer enabled convolutional neural network for plant disease identification: PlantXViT

ViT-SmartAgri: Vision Transformer and Smartphone-Based Plant Disease Detection for Smart Agriculture

Automatic mango leaf disease detection using different transfer learning models

Agricultural innovation through deep learning: a hybrid CNN-Transformer architecture for crop disease classification

Processing Plant Diseases Using Transformer Model

Deep Learning for Tomato Diseases: Classification and Symptoms Visualization

Detection and classification of plant diseases in leaves through machine learning

Recognition of mango leaf disease using convolutional neural network models: a transfer learning approach

Vision Transformers For Weeds and Crops Classification Of High Resolution UAV Images

A Multitask Learning-Based Vision Transformer for Plant Disease Localization and Classification

Crop Disease Identification Using Transfer Learning

Early and Accurate Detection of Tomato Leaf Diseases Using TomFormer

Efficient Mobile Deployment of Lightweight CNN-Transformer Model for Crop Disease Diagnosis in Smart Agriculture

Transformer based fruits disease classification