Lightweight CNNs for Eggplant Leaf Disease Classification on a Balanced Dataset: A Comparative Study of MobileNetV3 and EfficientNetV2-B2
(1)
(2) Universitas Universal, Batam
(3) Universitas Universal, Batam
(4) Universitas Universal, Batam
(*) Corresponding Author
Abstract
Eggplant (Solanum melongena) is a vital agricultural commodity, but its yield is highly vulnerable to foliar diseases. Early and accurate detection using deep learning is essential for effective crop management. However, deploying automated detection in real-world agricultural settings faces two primary challenges: (1) severe classification bias caused by complex background noise and data imbalance, and (2) extreme computational constraints that hinder the deployment of conventional deep learning models on farmers' edge devices. This study presents a robust methodology for classifying four conditions of eggplant leaves (Healthy, Leaf Spot, Mosaic Virus, and Insect Pest) by implementing an automated background removal technique and targeted data augmentation, resulting in a perfectly balanced dataset of 1,400 images. Furthermore, this research conducts a comparative analysis between two distinct categories of lightweight Convolutional Neural Networks (CNNs): MobileNetV3-Large (representing ultra-lightweight architectures with 224x224 input resolution) and EfficientNetV2-B2 (representing medium-lightweight architectures with 260x260 input resolution). The models were evaluated based on their accuracy, loss convergence, and computational efficiency using an 80:20 data split and early stopping callbacks to prevent overfitting. Experimental results demonstrate that both models achieved exceptional performance. EfficientNetV2-B2 exhibited superior stability and precision, achieving a peak validation accuracy of 97.50% and a validation loss of 0.073. Meanwhile, MobileNetV3-Large reached a validation accuracy of 96.07% with significantly faster training iterations. These findings indicate that while EfficientNetV2-B2 is highly recommended for precision-critical agricultural diagnostics, MobileNetV3-Large remains a formidable alternative for deployment on edge devices with extreme computational constraints
Keywords
Full Text:
PDFReferences
A. Krishnaswamy Rangarajan and R. Purushothaman, “Disease Classification in Eggplant Using Pre-trained VGG16 and MSVM,” Sci. Rep., vol. 10, no. 1, p. 2322, 2020, doi: 10.1038/s41598-020-59108-x.
S. Lika, Y. Pernando, and A. Kurniawan, “Lightweight Deep Learning for Object Detection on Mobile Device,” Bull. Informatics Data Sci., vol. 2, no. 2, pp. 106–114, 2023, doi: 10.61944/bids.v2i2.82.
R. E. Saragih, A. R. Purnajaya, I. Syafrinal, and Y. Pernando, “Mango and Banana Ripeness Detection based on Lightweight YOLOv8,” J. Buana Inform., vol. 15, no. 2, pp. 79–88, 2024. doi: 10.24002/jbi.v15i2.8895
Y. Pernando, E. Febrianti, I. Syafrinal, Y. Roza, and U. Afifah, “Deep Learning for Faces on Orphanage Children Face Detection,” JURTEKSI (Jurnal Teknol. dan Sist. Informasi), vol. 9, no. 1, pp. 25–32, Dec. 2022, doi: 10.33330/jurteksi.v9i1.1858.
Y. Pernando, Y. Heryadi, I. Sonata, L. A. Wulandhari, and A. S. Girsang, “Use of CNN in Diagnosing Banana Leaf Diseases an Agricultural Technology-Based Solution BT - Advances in Smart Knowledge Computing: Towards Post Artificial Intelligence Era,” in Advances in Smart Knowledge Computing, F. Lumban Gaol, T. Matsuo, and T. Ito, Eds., Cham: Springer Nature Switzerland, 2026, pp. 243–255. doi: 10.1007/978-3-032-01133-6_17.
R. E. Saragih, Y. Pernando, A. J. Santoso, Y. Roza, M. KH, and I. Verdian, “Optimizing The Sorting Of Barangan Bananas Through Hybrid And Transfer Learning Approaches,” ICIC Express Lett, vol. 19, no. 4, pp. 459–466, 2025.
J. Naranjo-Torres, M. Mora, R. Hernández-García, R. J. Barrientos, C. Fredes, and A. Valenzuela, “A Review of Convolutional Neural Network Applied to Fruit Image Processing,” Applied Sciences, vol 10, no 10, 2020. doi: 10.3390/app10103443.
Masparudin, I. Fitri, and Sumijan, “Pengembangan Sistem Klasifikasi Buah Apel Menggunakan Convolutional Neural Network (CNN) Arsitektur MobileNet pada Platform Android,” Sist. J. Sist. Inf., vol. 13, no. 1, pp. 230–243, 2024, doi: https://doi.org/10.32520/stmsi.v13i1.3533.
L. Chen, S. Li, Q. Bai, J. Yang, S. Jiang, and Y. Miao, “Review of Image Classification Algorithms Based on Convolutional Neural Networks,” Remote Sensing, vol 13, no 22, 2021. doi: 10.3390/rs13224712.
I. Goodfellow, B. Yoshua, and A. Courville, Deep Learning. The MIT Press, 2016.
F. Zhu, Y. Sun, Y. Zhang, W. Zhang, and J. Qi, “An Improved MobileNetV3 Mushroom Quality Classification Model Using Images with Complex Backgrounds,” Agronomy, vol 13, no 12, 2023. doi: 10.3390/agronomy13122924.
A. R. Purnajaya and M. Masparudin, “Evaluating Resampling Methods for Imbalanced Necrosis Classification on CT Scans,” JUTEKOM (Jurnal Teknol. dan Ilmu Komputer), vol. 2, no. 2, 2026, doi: 10.65258/jutekom.v2.i2.56.
G. Chu et al., “Searching for MobileNetV3,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 1314–1324. doi: 10.1109/ICCV.2019.00140.
M. Tan and Q. V Le, “EfficientNetV2 : Smaller Models and Faster Training,” in International Conference on Machine Learning (ICML), 2021, pp. 8786–8798.
J. Chen, D. Zhang, M. Suzauddola, Y. A. Nanehkaran, and Y. Sun, “Identification of plant disease images via a squeeze-and-excitation MobileNet model and twice transfer learning,” IET Image Process., vol. 15, no. 5, pp. 1115–1127, Apr. 2021, doi: https://doi.org/10.1049/ipr2.12090.
J. Zhang, X. Yang, X. Fu, B. Wang, and H. Li, “LDL-MobileNetV3S : an enhanced lightweight MobileNetV3-small model for potato leaf disease diagnosis through multi-module fusion,” Front. Plant Sci., vol. 16, no. October, pp. 1–19, 2025, doi: 10.3389/fpls.2025.1656731.
J. Han, Z. Wu, Y. Ding, Y. Guo, and R. Fu, “A privacy-protecting eggplant disease detection framework based on the YOLOv11n-12D model.,” Front. Plant Sci., vol. 16, p. 1634408, 2025, doi: 10.3389/fpls.2025.1634408.
R. Utomo, M. Idhom, and T. Trimono, “Rice Leaf Disease Classification Using EfficientNetV2 with Hyperparameter Tuning,” bit-Tech, vol. 8, no. 2, pp. 2038–2047, Dec. 2025, doi: 10.32877/bt.v8i2.3194.
M. Masparudin, A. Abdullah, and U. Usman, “Sistem prediksi kualitas santan kelapa menggunakan nearest mean classifier (NMC),” Sist. J. Sist. Inf., vol. 9, no. 3, pp. 646–655, 2020, doi: https://doi.org/10.32520/stmsi.v9i3.1015.
W. Wang, Y. Hu, T. Zou, H. Liu, J. Wang, and X. Wang, “A New Image Classification Approach via Improved MobileNet Models with Local Receptive Field Expansion in Shallow Layers.,” Comput. Intell. Neurosci., vol. 2020, p. 8817849, 2020, doi: 10.1155/2020/8817849.
Y. Kaya and E. Gürsoy, “A MobileNet-based CNN model with a novel fine-tuning mechanism for COVID-19 infection detection,” Soft Comput., vol. 27, no. 9, pp. 5521–5535, 2023, doi: 10.1007/s00500-022-07798-y.
S. Zhu et al., “Screening of Common Retinal Diseases Using Six-Category Models Based on EfficientNet,” Front. Med., vol. 9, p. 808402, 2022, doi: 10.3389/fmed.2022.808402.
F. Chollet, Deep Learning with Python. Manning Publications, 2021.
P. Galdi and R. Tagliaferri, “Data Mining: Accuracy and Error Measures for Classification and Prediction,” Elsevier, 2018. doi: 10.1016/B978-0-12-809633-8.20474-3.
R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization,” in 2017 IEEE International Conference on Computer Vision (ICCV), 2017, pp. 618–626. doi: 10.1109/ICCV.2017.74.
DOI: http://dx.doi.org/10.61944/bids.v5i1.166
Refbacks
- There are currently no refbacks.
Copyright (c) 2026 Masparudin Masparudin, Yonky Pernando, Ihsan Verdian, Fazlul Rahman
This work is licensed under a Creative Commons Attribution 4.0 International License.
Bulletin of Informatics and Data Science
Asosiasi Peneliti Data Science Indonesia
Email: pdsi.bids@gmail.com
This work is licensed under a Creative Commons Attribution 4.0 International License.