An enhanced lightweight T-Net architecture based on convolutional neural network (CNN) for tomato plant leaf disease...

Area of Science:

• Agricultural Science
• Computer Science
• Plant Pathology

Background:

• Tomato cultivation is vital globally, but diseases like Fusarium wilt and bacterial blight significantly reduce yield and quality.
• Current disease detection methods are labor-intensive and time-consuming, hindering efficient crop management.
• Automated disease detection is crucial for enhancing global agricultural productivity and food security.

Purpose of the Study:

• To develop a rapid and accurate automated system for detecting tomato leaf diseases.
• To introduce the T-Net model, a novel deep learning approach for classifying tomato diseases.
• To provide farmers with practical tools for effective disease management.

Main Methods:

• Proposed the T-Net model, integrating convolutional neural networks (CNNs) with transfer learning from VGG-16, Inception V3, and AlexNet.
• Utilized a layered architecture within the deep learning framework for enhanced feature extraction.
• Conducted extensive experimentation and comparative analysis against existing methods.

Main Results:

• The T-Net model achieved a high accuracy rate of 98.97% in classifying tomato leaf diseases.
• Demonstrated superior performance compared to previous automated disease detection techniques.
• Validated the model's effectiveness through rigorous testing and validation protocols.

Conclusions:

• The T-Net model offers a dependable and efficient solution for automated tomato disease diagnosis.
• This advancement in agricultural technology provides farmers with practical insights for crop protection.
• The developed framework represents a significant step towards sustainable and productive tomato farming.