Automated Wheat Plant Disease Detection using Deep Learning: A Multi-Class Classification Approach

Wheat disease Finding is a critical duty in agriculture to guarantee a good crop output and avoid financial losses. Deep learning-based techniques have recently demonstrated promising results in the detection and classification of wheat illnesses from photographs. In this study, we suggest a VGG19 deep convolutional neural network architecture-based system for detecting wheat diseases. The suggested approach makes use of a dataset of photographs of both healthy and ill wheat plants that has been resized, normalized, and enhanced. On this dataset, which contains 10 types of wheat illnesses, including powdery mildew, yellow rust, and leaf rust, among others, we refine the pre-trained VGG19 model. We assess the suggested model's performance using a variety of metrics, including accuracy, precision, recall, and F1 score. The findings demonstrate that the suggested model beats various cutting-edge wheat disease detection algorithms, achieving an accuracy of 97.65% on the validation dataset. The suggested technique can be implemented in practical applications to give farmers a dependable and effective way to identify wheat illnesses. As a result, they will be better able to prevent crop losses and increase agricultural productivity. The suggested method can be used to other crops and diseases, which makes it a significant advancement in the fields of agricultural and computer vision.