Leaf Disease Detection of Multiple Plants Using Machine Learning

Abstract

Diseases and pests cause great economic loss to the multiple industry every year. The detection of various multiple plant diseases is challenging for the farmers as the symptoms produced by different diseases may be very similar, and may be present simultaneously. This project is an attempt to provide the timely and accurate detection and identification of multiple leaf diseases.  In the proposed study, we develop a Machine Learning based models for detection and classification of multiple leaf diseases at the initial stages. Data augmentation is performed on a collected dataset. We applied data augmentation techniques like rotation, translation, reflection and scaling. Further implemented the leaf plant segmentation using threshold and K-Means clustering and edge detection using Canny edge algorithm. This approach involves collecting a diverse dataset of healthy and diseased leaf images from different plant species, preprocessing the images to enhance quality, and extracting relevant features using image processing techniques. Machine learning models, such as Support Vector Machines (SVM), Random Forest, and MACHINE learning architectures like Convolutional Neural Networks (CNNs), are then trained to classify leaf diseases with high accuracy.