DETECTION OF PLANT LEAF DISEASES USING IMAGE PROCESSING

Abstract

Agricultural productivity is strongly affected by plant diseases caused by microorganisms, pests, and
bacteria that spread through parts of the plant like leaves, stems, and fruits. Early detection is crucial to prevent
these diseases from reducing crop quality and yield, which harms the economy, especially in agricultural countries.
Traditionally, farmers inspect plants manually for diseases, but this method is slow and often inaccurate, with
detection rates as low as 60%. It also requires expert knowledge, making it impractical for large- scale farms. To
solve this problem, modern technologies like image processing and computer vision are used to automate disease
detection. Image processing involves analyzing plant images to identify symptoms using techniques such as image
enhancement, segmentation, feature extraction, and classification. This process can convert images from RGB to
other color spaces like YCbCr to make disease symptoms clearer. The system then extracts features like color,
texture, and shape to accurately identify the disease. Machine learning methods, such as support vector machines
(SVM), are used for classification. Automated systems can detect diseases with up to 99% accuracy, much higher
than manual methods. Real-time crop image capturing and disease detection systems provide immediate feedback to
farmers about plant health and specific diseases affecting their crops. These systems can also track environmental
data, such as temperature and humidity, which helps in disease management and crop growth. Content-based image
retrieval (CBIR) systems that compare features like color, shape, and texture improve disease detection accuracy.
These systems help farmers make better decisions about irrigation, fertilization, and disease control. By automating
disease detection, farmers can save time, reduce the need for expert help, and increase crop productivity, leading to
more efficient farming and better food security.