Automated grape leaf nutrition deficiency disease detection and classification Equilibrium Optimizer with deep...

Area of Science:

• Agricultural Science
• Computer Science
• Plant Pathology

Background:

• Grapevine cultivation is vital globally, but diseases significantly impact yield and quality.
• Accurate and early detection of grape leaf diseases is crucial for effective management.
• Current disease identification methods can be time-consuming and require expert knowledge.

Purpose of the Study:

• To develop and evaluate an automated system for detecting and classifying grape leaf diseases.
• To enhance the accuracy and efficiency of disease diagnosis in grapevine cultivation.
• To leverage deep learning for improved agricultural disease management.

Main Methods:

• Image preprocessing using Contrast Limited Adaptive Histogram Equalisation (CLAHE) and Adaptive Bilateral Filtering (ABF).
• Feature extraction via the SqueezeNet model.
• Hyperparameter optimization using the Equilibrium Optimizer (EO) algorithm.
• Classification performed by a Stacked Autoencoder (SAE) model.

Main Results:

• The proposed Equilibrium Optimizer-Deep Transfer Learning-Grape Leaf Disease Classification (EODTL-GLDC) technique achieved high precision.
• Achieved 96.31% precision on testing datasets and 96.88% on training datasets (80:20 split).
• Demonstrated superior performance compared to other deep learning and machine learning methods.

Conclusions:

• The developed automated system effectively detects and categorizes grape leaf diseases.
• The integration of SqueezeNet, EO, and SAE models offers a robust solution for agricultural disease diagnosis.
• This approach shows significant potential for practical application in precision agriculture and disease management.