Evaluation of Tea Leaf Disease Identification Based on Convolutional Neural Networks VGG16, ResNet50, and DenseNet169 Image Recognitions
Xianghong Deng, Chonlatee Photong
Abstract
Tea is the most favorite drink in the world and its demand seems to continuously increase; therefore, large amount of tea production while retaining high-quality tea is required for the marketplaces. However, based on many research literacies, the yield and quality of tea are possibly affected by a diversity of diseases, which would need high accuracy in identification in order to determine the most suitable treatment or specific preventive actions. In this study, the tea leaf disease identification based on the three most widely used convolutional Neural Networks VGG16, Resnet50, and DenseNet169 were examined and evaluated. By accommodating the relevant parameters of these three identification models and the learning rates, the performances of the three convolutional neural networks were analyzed and compared. The simulated test results showed that VGG16, Resnet50, and DenseNet169 achieved the highest accuracy rates of 90.6%, 95.2%, and 99.0%, respectively. DenseNet169 provided the fastest convergence speed and the highest accuracy among the models. In addition, the DenseNet169 showed the possibility of the precise diagnosis and identification of other crop leaf diseases as well.