Grey Blight Disease Detection on Tea Leaves Using Improved Deep Convolutional Neural Network

Area of Science:

• Agricultural Science
• Computer Science
• Plant Pathology

Background:

• Grey blight disease poses a significant threat to tea production.
• Accurate and early diagnosis of plant diseases is crucial for effective management.
• Existing diagnostic methods may lack the precision and speed required for large-scale agricultural applications.

Purpose of the Study:

• To develop and evaluate a novel deep convolutional neural network (DCNN) for the automated diagnosis of grey blight disease in tea leaves.
• To improve the accuracy and efficiency of tea plant disease detection using advanced machine learning techniques.

Main Methods:

• A custom DCNN architecture was designed, incorporating inverted residuals and linear bottleneck layers.
• A dataset of 1320 tea leaf images was collected and augmented using various image transformation techniques, resulting in 5280 images.
• The DCNN model was trained and validated on 5016 images and tested on 264 images, comparing its performance against state-of-the-art methods.

Main Results:

• The proposed DCNN model achieved a classification accuracy of 98.99% on the test dataset.
• The model demonstrated superior performance with precision, recall, and F-measure rates of 98.51%, 98.48%, and 98.49%, respectively.
• The DCNN model significantly outperformed existing techniques in detecting tea grey blight disease.

Conclusions:

• The developed DCNN model offers a highly accurate and efficient solution for diagnosing grey blight disease in tea leaves.
• This automated approach has the potential to aid farmers in early disease detection and management, thereby improving tea crop yield and quality.