Tomato Plant Disease Classification Using Transfer Learning with VGG19

The suggested model uses the VGG19 architecture and transfer learning to classify tomato plant diseases. 18,345 training and 4,585 testing photos from ten classes that includes both healthy and sick plants were used to train the model. To enhance generalization, preprocessing was carried out using data augmentation and rescaling approaches. This model was optimized for training with a batch size of 32 and a learning rate of 0.0001 using Adam and a categorical cross-entropy loss function. Throughout training, this model attained 96.33% training accuracy and 87.5% validation accuracy. These models can learn from this data without overfitting, as illustrated by the decrease in validation losses. The use of early stopping will prevent overfitting; thereby, an optimal performance is achieved. This paper indicates deep learning-based plant disease identification with significant agricultural uses in practice.

Ҳали таржима қилинмаган