An Enhanced Random Forest Classifier to detect Crop Disease with Texture and Shape Features OF Corn Images (ERFCTS)
Main Article Content
Abstract
Crop diseases pose a significant threat to global food security, affecting crop yield and quality. Early detection and accurate diagnosis of these diseases are crucial for effective disease management. A novel random forest classification for crop disease detection using texture and shape features is proposed in this research. The input dataset contains corn images with three different types of diseases namely Corn Blight, Rust and Gray Leaf Spot as well as Healthy images. These images are pre-processed using image processing techniques, binarized and the feature vector for each image is created and stored as training feature vector. The Random Forest Classifier is trained with this feature data set. The feature set with two images of each class is taken as test data set, test feature vector is calculated, mapped against training feature vector and finally classified by Random Forest Classifier. The performance of the proposed system is evaluated using the classification metrics such as accuracy, precision, recall, and F1-score. Experimental results demonstrate that the effectiveness of the proposed approach in detecting and classifying the crop diseases is 97% accuracy, thereby aiding farmers and agricultural stakeholders to take disease management strategies timely.