ARCHIVES

Early detection of plant leaf diseases is essential for improving crop productivity and ensuring sustainable agriculture. This paper presents a simple yet effective leaf disease detection system based on image processing and feature matching techniques. The proposed method begins with image acquisition, followed by preprocessing steps such as resizing and noise reduction. Leaf segmentation is performed using the HSV color space to isolate the leaf region from the background. Subsequently, grayscale conversion and intensity-based thresholding are applied to detect abnormal (diseased) regions. Morphological operations are used to refine the segmented regions and remove noise. The system estimates the percentage of affected leaf area and extracts relevant image features, including color and texture. These features are then compared with a pre-built database using a similarity-based matching approach to classify the disease type. The final results, including disease name, affected area, and confidence score, are displayed through an interactive GUI for user-friendly analysis. Experimental results demonstrate that the proposed system provides reliable disease detection with low computational complexity, making it suitable for practical applications in resource-constrained agricultural environments. The simplicity, efficiency, and real-time visualization capability of the system make it a promising tool for assisting farmers and researchers in early disease diagnosis.