The model, VGG16 achieved high accuracy detecting bean leaf diseases, with future work on ensemble methods and field deployment for diagnosis.
Authors
Tabib E Alahi, Daffodil International University, Dept. of Computer Science and Engineering, Dhaka, Bangladesh
Supta Das Dip, Daffodil International University, Dept. of Computer Science and Engineering, Dhaka, Bangladesh
Mayen Uddin Mojumdar, Daffodil International University, Dept. of Computer Science and Engineering, Dhaka, Bangladesh
Nivedita Haldar, Professor, Jindal Global Business School, O.P. Jindal Global University, Sonipat, Haryana, India
Narayan Ranjan Chakraborty, Daffodil International University, Dept. of Computer Science and Engineering, Dhaka, Bangladesh
Summary
Bangladesh, a predominantly an agricultural country. It’s common that vegetables become peaky in the time of cultivation due to some insect attack or other unavoidable activities. Accurate and early detection of plant diseases became necessary for improving crop health and ensuring better productivity. In this study, the work focused on the detection and classification of healthy and bean disease leaves using a deep learning approach based on the image classifier. The dataset contains two classes with two thousand and thirty-four thousand numbers of images. The proposed model was VGG16 which was trained and evaluated on the dataset, achieving high performance accuracy that demonstrates its robustness in distinguishing between healthy bean leaves and diseased bean leaves. Future work would be focused on further improving model performance, exploring ensemble methods and deploying the project for hand to hand uses into on-field health and disease diagnosis.
Published in: Proceedings of the 2025 12th International Conference on Computing for Sustainable Global Development, INDIACom 2025
To read the full article, please click here.
