IDENTIFICATION OF WHEAT STEM RUST USING MACHINE LEARNING TECHNIQUE – A SVM BASED APPROACH

This study proposes an efficient method for early detection of fungal diseases in wheat crops using image processing and machine learning. The system classifies wheat leaves into two categories: Black rust (stem rust) and Healthy. Images are first pre-processed through resizing to ensure uniformity, followed by feature extraction using Histogram of Oriented Gradients (HOG) to capture texture and shape characteristics. A Support Vector Machine (SVM) classifier is trained on labelled data to perform classification. Model performance is evaluated using confusion matrix–based metrics, including accuracy, precision, recall, and F1-score. Results indicate that the SVM model achieves high classification accuracy, demonstrating its effectiveness as a decision-support tool for early disease detection, enabling timely intervention and improved crop productivity.