Published Paper Details:

Sugarcane is a prominent cash crop in India, especially in Western Maharashtra, where vast areas are devoted to its cultivation. However, the crop is highly susceptible to various diseases that can lead to considerable yield losses. This work introduces a novel approach to sugarcane disease detection using Convolutional Neural Networks (CNNs)--a type of deep learning architecture well-suited for image classification tasks. The system is designed to automatically learn distinctive features from sugarcane leaf images and classify them as healthy or diseased. The dataset used in this work comprises images from three major disease categories and was enhanced through preprocessing and data augmentation techniques to improve model performance and generalizability. The CNN model was trained and evaluated on this enriched dataset, achieving an impressive test accuracy of 92.7%. Comparative analysis against other state-of-the-art classification algorithms revealed that the CNN-based model consistently outperformed its counterparts, highlighting its effectiveness and robustness. This system serves as a practical and reliable tool for both farmers and agricultural researchers, enabling early and accurate disease diagnosis. By facilitating timely intervention, the model has the potential to significantly reduce crop losses and improve overall yield.