International Journal of Science, Engineering and TechnologyAuthors: Sayan Roy, Dr. Saravanan Santhanam
Abstract: Plant diseases are a major concern in agriculture, as they directly affect crop yield, quality, and overall productivity. Early identification of these diseases is essential to reduce losses and ensure better crop management. This project presents an artificial intelligence-based system that detects plant diseases by analyzing images of plant leaves. The system is designed to provide users with a simple and efficient way of determining whether a plant is healthy or affected by disease without relying on expert knowledge. The proposed approach makes use of deep learning techniques, specifically transfer learning with a model such as EfficientNetB4. These models are trained on a dataset of plant leaf images after applying preprocessing and augmentation techniques to improve their ability to handle real-world variations. Each model offers a different balance between speed and accuracy, allowing a comparative analysis to identify the most suitable option for practical use. The system processes an input image, extracts relevant features, and classifies the plant condition as healthy or diseased. The results demonstrate that deep learning models can provide accurate and reliable predictions, with EfficientNetB4 offering faster performance while achieving higher accuracy. The proposed system can assist farmers in making timely decisions, reduce dependency on manual inspection, and contribute to more efficient and sustainable agricultural practices.
DOI:
