Optimizing Road Safety with MobileNet-Based Classification of Over-dimensioned Trucks

Nurul Afifah Arifuddin, Hary Capri, Deni Setiawan, Rifka Dwi Amalia, Kharisma Wiati Gusti

Abstract


This study aims to automatically detect overdimension trucks using a lightweight and efficient deep learning model based on MobileNet. Overdimension trucks pose serious threats to road infrastructure, traffic safety, and contribute to increased economic costs due to road damage and congestion. The developed model utilizes MobileNet as a feature extractor without the standard fully connected layers, and is equipped with additional layers including Flatten, Batch Normalization, Dense with Leaky ReLU activation, and Dropout to enhance training stability and prevent overfitting. The dataset consists of two classes—normal trucks and overdimension trucks—with images sized 128×128 pixels, collected from internet sources and field photos. The training process employs binary crossentropy loss, the Adam optimizer with an initial learning rate of 0.0001, and an Early Stopping mechanism. Fine-tuning is performed by unfreezing layers from the 100th layer upward and lowering the learning rate to 0.00001. Evaluation results show an accuracy of 97.92%, with consistent loss and accuracy visualization, demonstrating the model's capability in classifying overdimension trucks to support automatic traffic monitoring systems. This model has the potential to be implemented in toll gate systems to automatically deny access to overdimension vehicles. Furthermore, integration with roadside CCTV allows real-time monitoring of vehicle dimension violations across various traffic checkpoints.


Keywords


Overdimension; Truck; Deep Learning; Mobilenet; Classification

Full Text:

References


Antara, “Jembatan Widang-Babat: Baut-bautnya Harus Sering Dikencangkan,” Tempo.co.

A. Budiharjo, T. Andika, N. Fitriani, R. Rukman, and B. Turasno, “Operational Data Analytics of Over Dimensional and Overloaded Truck in Indonesia,” RSF Conference Series: Engineering and Technology, vol. 2, no. 2, pp. 88–98, Nov. 2022, doi: 10.31098/cset.v2i2.562.

A. Baharuddin, B. Istiyanto, and T. G. N. Prasetyo, “ANALISIS PENGARUH BEBAN, TATA CARA PEMUATAN DAN KECEPATAN TERHADAP KINERJA PENGEREMAN,” Jurnal Pendidikan Vokasi Otomotif, vol. 7, no. 1, pp. 150–167, 2024.

T. R. Gunawan, A. Kurniawan, and M. Fauzi, “ANALISIS PENGARUH KENDARAAN OVER DIMENSI DAN OVER LOADING,” Jurnal Ilmiah Teknik dan Manajemen Industri, vol. 3, no. 1, pp. 112–117, Jul. 2023, doi: 10.46306/tgc.v3i1.61.

J. A. Pratama and A. Susanti, “Analisis Penanganan Permasalahan Overdimension Overloading di Ruas Jalan Mojokerto–Surabaya,” Jurnal Media Publikasi Terapan Transportasi, vol. 1, no. 2 (Agustus), pp. 203–212, 2023.

I. Cil, K. Soğukçeşme, S. Zereyalp, and M. Erginli, “Detection of Defects on Highways Followed by Logistics Transport Trucks with Deep Learning and Providing Safe Driving Support for Drivers,” in 2023 Innovations in Intelligent Systems and Applications Conference (ASYU), IEEE, 2023, pp. 1–5.

H. Wang, D. Zhang, and Z. Huang, “Research on heavy truck recognition algorithm based on deep learning,” International Journal of Wireless and Mobile Computing, vol. 23, no. 2, pp. 132–138, 2022.

W. Sun, W. Jiang, H. Liu, L. Han, Y. Liu, and S. Li, “Research on the Method and Application of Truck Type Recognition Based on Deep Learning from the Perspective of Unmanned Aerial Vehicles,” Transp Res Rec, p. 03611981241254109, 2024.

N. A. Arifuddin, K. W. Gusti, and R. D. Amalia, “A CNN Model for ODOL Truck Detection,” Sinkron: jurnal dan penelitian teknik informatika, vol. 8, no. 3, pp. 1697–1705, 2024.

MUHAMMAD FAJAR B, “DETEKSI DAN PELACAKAN OBJEK MENGGUNAKAN METODE CNN MOBILENET SSD DAN CENTROID TRACKING,” Universitas Gadjah Mada, Yogyakarta, 2021.

D. Picard, “Torch.manual_seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision,” CoRR, vol. abs/2109.08203, 2021, [Online]. Available: https://arxiv.org/abs/2109.08203

R. Santoso, “Augmentasi Data pada Prasasti Logam untuk Deteksi Aksara Kawi,” JURNAL FASILKOM, vol. 14, no. 1, pp. 234–241, 2024.

R. Anggarkusuma and I. A. Musdar, “KLASIFIKASI CITRA KOMPONEN SEPEDA MOTOR MENGGUNAKAN ALGORITMA CNN DENGAN ARSITEKTUR MOBILENET,” KHARISMA Tech, vol. 19, no. 2, pp. 1–12, 2024.

J. Lee, R. Tang, and J. Lin, “What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning,” CoRR, vol. abs/1911.03090, 2019, [Online]. Available: http://arxiv.org/abs/1911.03090

D. Sinha and M. El-Sharkawy, “Thin MobileNet: An Enhanced MobileNet Architecture,” in 2019 IEEE 10th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), 2019, pp. 280–285. doi: 10.1109/UEMCON47517.2019.8993089.

K. G. Panchbhai, M. G. Lanjewar, and A. V. Naik, “Modified MobileNet with leaky ReLU and LSTM with balancing technique to classify the soil types,” Earth Sci Inform, vol. 18, no. 1, p. 77, 2024, doi: 10.1007/s12145-024-01521-1.

S. Cai, Y. Shu, G. Chen, B. C. Ooi, W. Wang, and M. Zhang, “Effective and efficient dropout for deep convolutional neural networks,” arXiv preprint arXiv:1904.03392, 2019.

C. Zhang et al., “UNDERSTANDING DEEP LEARNING REQUIRES RE-THINKING GENERALIZATION,” 2017.




DOI: https://doi.org/10.30591/jpit.v10i2.8239

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

JPIT INDEXED BY

  
  

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.