Pendekatan U-Net dan EfficientNet Transfer Learning pada Segmentasi Multikelas Penyakit Daun Kentang Shella Ayu Putriani
DOI:
https://doi.org/10.71417/jitie.v1i2.64Keywords:
Deep Learning, EfficientNet, Plant Disease, Potato Leaf, U-NetAbstract
Penyakit daun kentang seperti Early Blight dan Late Blight menyebabkan kerugian ekonomi global mencapai $20 miliar per tahun, menantang identifikasi manual yang subjektif. Penelitian ini bertujuan membandingkan performa U-Net standar dengan U-Net ber-backbone EfficientNet-B0 untuk segmentasi multikelas (Healthy, Early Blight, Late Blight). Penelitian eksperimental menggunakan dataset PlantVillage (3.000 citra, 256×256 piksel) dengan preprocessing, augmentasi, dan pembagian stratified (70:15:15). Model dilatih dengan TensorFlow/Keras di Google Colab, menggunakan metrik IoU, mIoU, dan Dice Coefficient. Hasil menunjukkan U-Net + EfficientNet-B0 mencapai mIoU 75,5%, meningkat 4,6% dari U-Net standar (70,9%). Kesimpulan: Integrasi EfficientNet-B0 meningkatkan akurasi segmentasi dan generalisasi untuk aplikasi pertanian presisi.
Downloads
References
Arshad, M., Rehman, A., Naseer, S., & Alshahrani, A. (2023). End-to-end hybrid deep learning framework for potato leaf disease detection. Alexandria Engineering Journal, 78, 102-115. https://doi.org/10.1016/j.aej.2023.10.045
Fu, J., Zhao, Y., & Wu, G. (2023). Potato leaf disease segmentation method based on improved UNet. Applied Sciences, 13(20), 11179. https://doi.org/10.3390/app132011179
Guerra Ibarra, J. P., & De la Cruz, F. J. (2022). Evaluation of the effectiveness of the UNet model with different backbones in the semantic segmentation of tomato leaves and fruits. Computers and Electronics in Agriculture, 202, 107345. https://doi.org/10.1016/j.compag.2022.107345
Kyaw, P. H., Gutierrez, J., & Ghobakhlou, A. (2024). A systematic review of deep learning techniques for phishing email detection. Electronics, 13(19), 3823. https://doi.org/10.3390/electronics13193823
Marques, M. J. M., et al. (2024). Plant Doctor: A hybrid machine learning and image segmentation software to quantify plant damage in video footage. Ecological Informatics, 82, 102678. https://doi.org/10.1016/j.ecoinf.2024.102678
Nazir, T., Iqbal, M. M., Jabbar, S., Hussain, A., & Albathan, M. (2023). EfficientPNet—An optimized and efficient deep learning approach for classifying disease of potato plant leaves. Agriculture, 13(4), 841. https://doi.org/10.3390/agriculture13040841
Sangar, G., & Rajasekar, V. (2025). Optimized classification of potato leaf disease using EfficientNet-LITE and KE-SVM in diverse environments. Frontiers in Plant Science, 16, Article 1499909. https://doi.org/10.3389/fpls.2025.1499909
Syukriyah, Y., & Purnama, A. (2025). Image segmentation for sweet potato leaf disease detection using U-Net. International Journal of Multidisciplinary Approach Research and Science, 3(3), 833-844. https://doi.org/10.59653/ijmars.v3i03.1848
Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking model scaling for convolutional neural networks. Proceedings of the 36th International Conference on Machine Learning (ICML 2019), 10691-10700.
Vallabhajosyula, S., et al. (2022). Deep ensemble neural networks for automated plant disease classification. Plants, 11(21), 2825. https://doi.org/10.3390/plants11212825
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Shella Ayu Putriani, Muhammad Akbar, Muhammad Irvai, Joni Karman (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
