A robust deep learning framework for wheat disease identification using modified CNN

Wheat is one of the most important food crops in the world, giving billions of people the calories and vital nutrients they need. However, a number of illnesses that can negatively affect yield and quality present serious obstacles to wheat production. It is essential to identify wheat infections early and accurately in order to put control measures in place in a timely manner and avoid significant financial losses. This study suggests a modified Convolutional Neural Network (CNN) architecture designed specifically for wheat crop disease detection. The proposed model integrates attention mechanisms, dense connection patterns, and a novel feature fusion technique to increase the accuracy and efficacy of disease diagnosis. Extensive testing was conducted using a huge dataset of images of wheat plants with various diseases. These included stripe rust, leaf rust, powdery mildew, and leaf blotch. An overall accuracy of 97.8% was achieved by the proposed revised CNN model in comparison to other state-of-the-art models. The results demonstrate how the proposed approach could be applied in real-world precision agriculture systems; automated crop monitoring systems were also considered in the study.