Utilizing Machine Learning Techniques forthe Detection of Plant Leaf Diseases

Abstract

Identification of plant diseases is crucial for preserving crops and ensuring food security. Analysis of detectable chemicals in plants is essential to understand transmission mechanisms and develop effective strategies for disease control measures to conserve agricultural products and prevent losses. However, manual monitoring of plant health is labor-intensive and time-consuming, requiring specialized skills and knowledge. To overcome these challenges, random forest systems are emerging as a powerful tool for disease detection and classification in plants. The process involves several steps, including image acquisition, preprocessing, and segmentation, followed by feature extraction, model training, and testing. Leveraging machine learning techniques, the random forest algorithm enables accurate classification of healthy and diseased leaves based on selected features. Image classification techniques are utilized to extract color information, while global features such as size and texture are captured through annotation. The dataset used for model training and testing comprises diverse samples, encompassing healthy and diseased plants. The random forest model is trained on 70% of the data to ensure robust learning, while the remaining 30% is reserved for testing, facilitating the exploration of model performance and overall feasibility