Performance optimization VGG16 architecture a lightweight hybrid transformer integrated with convolutional neural network for disease classification of porang plant leaves
Keywords:
VGG16, CNN, Hybrid transformer, Plant disease, Amorphophallus muellerAbstract
The porang plant (Amorphophallus Muelleri Blume) has high economic value with export demand reaching 6,064,947 kg worth IDR 297 billion in 2019. However, the productivity of porang plants in Indonesia is seriously threatened by disease attacks that can reduce harvest yields by 23-48%. Conventional disease detection methods require a long time and specialized expertise, while economic losses due to plant diseases reach trillions of rupiah annually. The development of an artificial intelligence-based automatic detection system is a critical solution to support national food security and increase the export competitiveness of Indonesian porang commodities. This study aims to develop a classification system for porang plant leaf diseases using the VGG16 architecture on a convolutional neural network to achieve detection accuracy above 95%. The developed system will integrate transfer learning technology to optimize model performance with limited datasets. This research also aims to design an algorithm capable of identifying multiple types of porang leaf diseases in real-time with a high level of precision to support farmer decision-making. The main output is a Based on the empirical results presented, the hybrid transformer model (98%) clearly outperforms the VGG16 model (95%) in the classification accuracy metric. Performance improvement, the 3% advantage of the hybrid transformer model indicates that combining the local feature extraction capabilities of CNNs. This research will produce a standardized dataset of porang leaf diseases for the development of this system, which is expected to reduce crop losses by up to 30% and increase the efficiency of plant disease diagnosis from days to minutes.
References
1. D. Rustandi, S. H. Wijaya, Mushthofa, and R. Damayanti, “Anatomy Identification of Bamboo Stems with The Convolutional Neural Networks (CNN) Method,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 8, no. 1, Art. no. 1, Jan. 2024, doi: 10.29207/resti.v8i1.5370.
2. M. Zuhan and Y. Kristian, “Detection of Porang Plant Diseases and Pests (Amorphophallus Muelleri) Based on Leaf Imagery Utilizing DCNN Transfer Learning,” J. Sist. Telekomun. Elektron. Sist. Kontrol Power Sist. Dan Komput., vol. 3, no. 2, pp. 129–140, July 2023, doi: 10.32503/jtecs.v3i2.3709.
3. H. O. Velesaca, R. Mira, P. L. Suárez, C. X. Larrea, and A. D. Sappa, “Deep Learning based Corn Kernel Classification,” in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), June 2020, pp. 294–302. doi: 10.1109/CVPRW50498.2020.00041.
4. “Analysis for Detecting Banana Leaf Disease Using the CNN Method | JUITA: Jurnal Informatika.” Accessed: June 02, 2025. [Online].
5. A. Prasetyo, F. Masykur, and A. R. Yusuf, “Deteksi Gula Aren dengan CNN Berbasis AlexNet sebagai Solusi Keamanan Pangan untuk Penderita Diabetes,” SinarFe7, vol. 7, no. 1, pp. 302–309, Aug. 2025.
6. F. Masykur, A. Prasetyo, I. A. Zulkarnain, E. Kumalasari, and P. Utomo, “Yolo-Drone: Detection Paddy Crop Infected Using Object Detection Algorithm Yolo and Drone Image,” JOIV Int. J. Inform. Vis., vol. 9, no. 5, pp. 2104–2111, Sept. 2025, doi: 10.62527/joiv.9.5.3472.
7. A. Prasetyo et al., “Analisis Deteksi Citra Mata Ikan Nila dengan Metode Convolutional Neural Network Arsitektur Alexnet,” J. Pustaka Data Pus. Akses Kaji. Database Anal. Teknol. Dan Arsit. Komput., vol. 5, no. 1, pp. 48–53, June 2025, doi: 10.55382/jurnalpustakadata.v5i1.995.
8. S. Dwianto, F. N. Mubarok, D. Satriatama, and T. Agustin, “Penerapan Convolutional Neural Network (CNN) dalam Deteksi Penyakit pada Tanaman Terong,” Pros. Semin. Nas. Amikom Surak., vol. 2, pp. 270–280, Dec. 2024.
9. L. Fitriani, D. Tresnawati, and M. B. Sukriyansah, “Image Classification On Garutan Batik Using Convolutional Neural Network with Data Augmentation,” JUITA J. Inform., pp. 107–115, May 2023, doi: 10.30595/juita.v11i1.16166.
10. N. Helmawati and E. Utami, “Analysis for Detecting Banana Leaf Disease Using the CNN Method,” JUITA J. Inform., pp. 29–36, Mar. 2025, doi: 10.30595/juita.v13i1.24514.
11. V. D. T. Wijaya, S. Novianto, and U. Rosyidah, “DETEKSI HURUF ARAB MENGGUNAKAN METODE FREEMAN CHAIN CODE,” Techno.Com, vol. 13, no. 4, Art. no. 4, 2014, doi: 10.33633/tc.v13i4.606.
12. W. N. Waluyo, R. R. Isnanto, and A. F. Rochim, “Comparison of Mycobacterium Tuberculosis Image Detection Accuracy Using CNN and Combination CNN-KNN,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 7, no. 1, Art. no. 1, Feb. 2023, doi: 10.29207/resti.v7i1.4626.
13. A. TiaraSari and E. Haryatmi, “Penerapan Convolutional Neural Network Deep Learning dalam Pendeteksian Citra Biji Jagung Kering,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 5, no. 2, Art. no. 2, Apr. 2021, doi: 10.29207/resti.v5i2.3040.
14. Y. Aufar, M. H. Abdillah, and J. Romadoni, “Web-based CNN Application for Arabica Coffee Leaf Disease Prediction in Smart Agriculture,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 7, no. 1, Art. no. 1, Feb. 2023, doi: 10.29207/resti.v7i1.4622.
15. R. Shen, T. Zhen, and Z. Li, “Segmentation of Unsound Wheat Kernels Based on Improved Mask RCNN,” Sensors, vol. 23, no. 7, Art. no. 7, Jan. 2023, doi: 10.3390/s23073379.
Downloads
Published
Conference Proceedings Volume
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.