Classification of Tea Leaf Diseases Based on ResNet-50 and Inception V3

Technological advances have made a major contribution to controlling plant diseases. One method for resolving issues with disease identification is the use of deep learning digital image processing. Tea leaf that requires fast and effective control. So, in this study, we adopted Convolutional Neural Network (CNN) architectures, namely ResNet-50 Inception V3, classify six types diseases attack leaves. The amount data used was 5867, which were divided into classes, healthy leaf, algal spot, brown blight, gray helopeltis, red spot. process distributing involves randomly splitting it three portions, an allocation 80% training, 10% validation, testing. classification carried out by adjusting batch sizes training maximizehyperparameters. are 16, 32, 64. Using different size scenarios each model, shows has better performance on 32 accuracy value 97.44%, while V3 best 64 97.62%..