Published Paper Details:

The global population has surged since 2022, intensifying the challenge of ensuring food security amid escalating agricultural demands. Agriculture remains central to addressing this issue but faces persistent threats from plant diseases, which contribute significantly to worldwide crop losses. Accurately identifying these diseases, particularly in their early stages, remains a formidable challenge. Automated plant disease identification and diagnosis systems are thus indispensable for effective disease management. This study explores deep learning approaches to enhance potato crop disease detection in smart farming, focusing on developing specialized databases for potatoes--a crucial crop for global food security. Potatoes are highly susceptible to various bacterial and fungal diseases, necessitating robust disease detection methods. To address this, the study constructs a dataset specific to potato cultivation and employs deep learning-based image classification techniques. A single deep learning model, trained exclusively on the potato dataset, achieves an accuracy of approximately 83% in identifying potato crop diseases. These findings highlight the potential of deep learning in enhancing disease management strategies, reducing reliance on manual inspection, and improving crop health monitoring. By integrating deep learning models into smart farming practices, this research contributes to the automation of plant disease diagnosis, ensuring timely intervention and precision agriculture. The promising results of this study pave the way for scalable and efficient solutions in potato farming and beyond, reinforcing the role of artificial intelligence in advancing modern agricultural practices and addressing global food security challenges.