Tea Leaf Disease Identification using Improved Convolution Neural Network

Plant diseases are a significant concern for the agricultural industry, as they can reduce crop yields and cause economic losses. Tea is a popular and widely consumed beverage in India, and the tea crop can be affected by different leaf diseases. Early detection of the diseases is essential to prevent the spread of other crops and minimize production losses. Traditional methods of detecting leaf diseases involve manually interpreting the images, which can be time-consuming and laborious. To address this, artificial intelligence techniques, specifically deep learning models are used for more accurate and efficient detection of tea leaf disease. This study compares the performance of several deep learning models including ResNet50, Resnet50-RS, ResNetlOl-V2, and Modified-ResNet50-V2 (M-RN50V2) and finds that the M-RN50V2 model has the highest accuracy at 90.S4%. Overall, this work aims to promote sustainable agriculture by leveraging cutting-edge technology for disease classification.