CNN-based Classification of Bean Leaf Diseases for Crop Monitoring

Bean crops are susceptible to numerous diseases that can severely effect harvest and quality. Timely and correct detection of these illnesses is life-threatening for ensuring crop health and food security. This study proposes a Convolutional Neural Network (CNN)-based method for automatic detection and classification of bean shrub diseases using greenery images. The dataset used consists of three types: angular leaf spot, bean rust, and healthy leaves. To improve model generalization, images received preprocessing and augmentation. This image dataset was used to train and evaluate the CNN model, which generated a 94.26% prediction accuracy. The model performed well when distinguishing between infected and healthy leaves as well as between other disease categories. In the agricultural sector, this approach provides a scalable and economical substitute for real-time disease intensive care. The results support deep learning's potential to promote early disease detection and sustainable farming methods.