Klasifikasi Kelayakan Ban Sepeda Motor Menggunakan Metode Convolutional Neural Network
Abstract
Tires is a primary component in motorcycle that plays a crucial role in driving safety and comfort. Damage to tires, such as wear, cuts, or cracks, can reduce traction, disrupt stability, and increase the risk of traffic accidents. Generally, tire condition inspections are conducted conventionally by technicians who may have limitations in accurately detecting damage. This research aims to develop a tire damage classification system using the Convolutional Neural Network (CNN) method with the MobileNetV2 architecture and with transfer learning approach. The dataset used consists of motorcycle tire images categorized into four classes: normal, bald, cutburst, and spotwear. The training process was conducted using a grid search technique to determine the optimal hyperparameter configuration. The best results obtained with a combination of batch size 16, learning rate 0.001, and 43 epochs, yielding a test accuracy of 96.67%, precision of 95%, recall of 95%, and an F1-score of 95%.
Keywords: Tire; Convolutional Neural Network; MobileNetV2
Abstrak
Ban merupakan komponen utama pada kendaraan sepeda motor yang berperan penting dalam keselamatan dan kenyamanan berkendara. Kerusakan pada ban, seperti keausan, sobekan, atau retakan, dapat mengurangi traksi, mengganggu stabilitas, dan meningkatkan risiko kecelakaan lalu lintas. Pemeriksaan kondisi ban secara umum masih dilakukan secara manual oleh teknisi, yang memiliki keterbatasan dalam hal objektivitas dan akurasi. Penelitian ini bertujuan untuk mengembangkan sistem klasifikasi kerusakan ban sepeda motor secara otomatis menggunakan metode Convolutional Neural Network (CNN) dengan arsitektur MobileNetV2 dengan pendekatan transfer learning. Dataset terdiri dari citra ban sepeda motor yang diklasifikasikan ke dalam empat kelas, yaitu normal, bald, cutburst, dan spotwear. Proses pelatihan dilakukan melalui metode grid search untuk menentukan konfigurasi parameter terbaik. Hasil terbaik diperoleh pada kombinasi hyperparameter dengan batch size 16, learning rate 0.001, dan jumlah epoch 43, menghasilkan akurasi uji sebesar 96,67%, precision 95%, recall 95% dan F1-score 95%.
Kata kunci: Ban; Convolutional Neural Network; MobileNetV2
References
H. Ponda, N. F. Fatma, dan I. Siswantoro, “Usulan Penerapan Lean Manufacturing Dengan Metode Value Stream Mapping (VSM) Dalam Meminimalkan Waste Pada Proses Produksi Ban Motor Pada Industri Pembuat Ban,” Heuristic, vol. 19, no. 1, pp. 23–42, 2022, doi: 10.30996/heuristic.v19i1.6568.
K. P. RI, “Tekan Angka Kecelakaan Lalu Lintas, Kemenhub Ajak Masyarakat Beralih ke Transportasi Umum dan Utamakan Keselamatan Berkendara,” Kementrian Perhubungan Republik Indonesia. Accessed: Oct. 07, 2024. [Online]. Available: https://dephub.go.id/post/read/tekan-angka-kecelakaan-lalu-lintas,-kemenhub-ajak-masyarakat-beralih-ke-transportasi-umum-dan-utamakan-keselamatan-berkendara# [Diakses: 07 Oktober 2024]
I. E. Hendrawan, R. I. Adam, dan C. Rozikin, “Klasifikasi Retak Ban Kendaraan Menggunakan Arsitektur ResNet50,” SATIN - Sains dan Teknologi Informasi, vol. 9, no. 1, pp. 23–32, 2023, doi: 10.33372/stn.v9i1.902.
Z. M. Jawi dan A. Hamzah, Car Users’ Knowledge and Practices on Tyre Maintenance in Malaysia, MRR No. 337, Malaysian Institute of Road Safety Research (MIROS),pp. 1–34 Jan. 2020.
D. Ekambaram and V. Ponnusamy, “Identification of Defects in Casting Products by using a Convolutional Neural Network,” IEIE Trans. Smart Process. Comput., vol. 11, no. 3, pp. 149–155, 2022, doi: 10.5573/IEIESPC.2022.11.3.149.
X. Liu, L. Faes, A. U. Kale, et al., “A comparison of deep learning performance against health-care professionals in detecting diseases from medical imaging: a systematic review and meta-analysis,” Lancet Digital Health, vol. 1, no. 6, pp. e271–e297, 2019, doi: 10.1016/S2589-7500(19)30123-2.
S. E. Sukmana, C. Rahmad, M. S. Khairy, R. Ariyanto, dan R. Andhani, “Classification of Damage to Car Tires Using Mobile-Based Deep Learning,” Prosiding IEIT 2024 - Int. Conf. on Electrical and Information Technology, Surabaya, pp. 176–181, 10 Januari 2024, doi: 10.1109/IEIT64341.2024.10763133.
N. Sharma, R. Sharma, dan N. Jindal, “Machine Learning and Deep Learning Applications-A Vision,” Global Transitions Proceedings, vol. 2, no. 1, pp. 24–28, 2021, doi: 10.1016/j.gltp.2021.01.004.
N. Febriyanto, C. Rahmad, dan C. B. Vista, “Klasifikasi Kelayakan Ban Mobil dengan Deep Learning,” Jurnal Informatika Polinema, vol. 7, no. 4, pp. 58–64, 2021.
K. Prayoga, R. Magdalena, dan S. Saidah, “Sistem Deteksi Kecacatan Ban Dengan Convolutional Neural Network,” e-Proceeding of Engineering, vol. 10, no. 3, pp. 2229–2234, 2023.
H. C. Mayana dan D. Leni, “Deteksi Kerusakan Ban Mobil Menggunakan Convolutional Neural Network dengan Arsitektur ResNet-34,” Surya Teknik, vol. 10, pp. 45–50, 2023.
J. G. Ebron, K. M. Paliso, G. A. C. Reyes, S. P. Tongol, J. R. Clavillas, dan D. Ortega, “Predicting Motorcycle Tire Failure with Deep Learning,” Proc. Int. Conf. on ICT and Knowledge Engineering (ICTKE), Bangkok, pp. 1–6, 15 Maret 2024, doi: 10.1109/ICTKE62841.2024.10787193.
C. Luo, X. He, J. Zhan, L. Wang, W. Gao, dan J. Dai, “Comparison and Benchmarking of AI Models and Frameworks on Mobile Devices,” arXiv preprint, arXiv:2005.05085, 2020. [Online]. Available: http://arxiv.org/abs/2005.05085 [Diakses: 01 Desember 2024].
D. Nguyen, C. Nguyen, T. Duong-Ba, H. Nguyen, A. Nguyen, dan T. Tran, “Joint network coding and machine learning for error-prone wireless broadcast,” Proc. IEEE 7th Annu. Comput. Commun. Work. Conf. (CCWC), Las Vegas, pp. 1–6, 9 Januari 2017, doi: 10.1109/CCWC.2017.7868415.
M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, dan L. Chen, “MobileNetV2: Inverted Residuals and Linear Bottlenecks,” Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, pp. 4510–4520, 2018.
B. Khasoggi, Ermatita, dan Samsuryadi, “Efficient MobileNet Architecture as Image Recognition on Mobile and Embedded Devices,” Indonesian Journal of Electrical Engineering and Computer Science, vol. 16, no. 1, pp. 389–394, 2019, doi: 10.11591/ijeecs.v16.i1.pp389-394.
B. H. Shekar, “Optimizing CNN Training using Batch Normalization and Dropout: A Case Study,” Lect. Notes Networks Syst., vol. 535, pp. 1–8, 2019.
D. A. Anggoro dan S. S. Mukti, “Performance Comparison of Grid Search and Random Search Methods for Hyperparameter Tuning in Extreme Gradient Boosting Algorithm to Predict Chronic Kidney Failure,” Int. J. Intell. Eng. Syst., vol. 14, no. 6, pp. 198–207, 2021, doi: 10.22266/ijies2021.1231.19.
J. Turihohabwe, S. Richard, dan W. William, “Hyperparameter Optimization Through Transfer Learning for Classification Tasks,” Indonesian Journal of Computer Science, vol. 14, no. 1, pp. 1–13, 2025.
S. Sharma dan K. Guleria, “A Deep Learning based Model for the Detection of Pneumonia from Chest X-Ray Images using VGG-16 and Neural Networks,” Procedia Computer Science, vol. 218, pp. 357–366, 2022, doi: 10.1016/j.procs.2023.01.018.
How To Cite This :
Refbacks
- There are currently no refbacks.
