DeepLearning-Based Tomato Leaf Disease Identification: Enhancing Classification with AlexNet

Diseases of tomato leaves are significant influence factors that affect the health conditions of crops. Therefore, various diseases have their special characteristics in spots, represented by shapes, colours, and locations. The paper proposes a AlexNet architecture-based classification model that can recognise and classify ten different types of diseases using RGB images from a standard normal dataset. The proposed AlexNet model has been designed to focus on several convolutional layers, a max-pooling operation, and fully connected layers that detect and analyze complex features of an image. A number of data augmentation methods have been employed, such as random rotation, shift, shear, and zoom, in order to make the model more robust and reduce some problems of overfitting. Having an accuracy of 97.54% and a low loss of 0.0722 itself depicts the goodness of the model in generalizing on several conditions and classify various diseases of the leaves correctly. This research underlines the potential of the proposed AlexNet-based approach to contribute to precision agriculture with a reliable tool for early and accurate disease detection. Results show significant improvement in disease classification and thus provide useful insights into effective plant disease management.