Intelligent Analysis for Tomato Disease Detection Based on Images Using a Hybrid Neural Network Architecture VGG16–RNN

This paper presents an approach to automatically identify tomato diseases from leaf images using a hybrid VGG16-RNN neural network architecture. The aim of the study is to develop an intelligent analysis of visual disease features that ensures high classification accuracy by combining convolutional and recurrent neural networks. The proposed model utilizes VGG16 to extract spatial features from images, and RNN to analyze sequential dependencies between the extracted features. This combined approach allows for consideration of both local textural features of lesions and their structural relationships at the level of contours and spots.

Hali tarjima qilinmagan