Shamsun Nahar; Md. Nazim Hossain; Noshin Un Noor; Mohammad Anwar Hossain; Md. Sabbir Hossain Babu; Md. Sohorab Hossen

Region-Specific Vehicle Classification in Bangladesh: A Comparative Study of CNN Architectures and Ensemble Strategies

PDF (1642KB), PP.106-120

Views: 0 Downloads: 0

Author(s)

Shamsun Nahar ^1,* Md. Nazim Hossain ¹ Noshin Un Noor ¹ Mohammad Anwar Hossain ¹ Md. Sabbir Hossain Babu ¹ Md. Sohorab Hossen ¹

1. Department of Computer Science and Engineering, World University of Bangladesh, Dhaka, 1230, Bangladesh

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2026.05.06

Received: 3 Jul. 2026 / Revised: 13 Jul. 2026 / Accepted: 13 Aug. 2026 / Published: 8 Oct. 2026

Index Terms

Convolutional Neural Networks, Deep Learning, Vehicle Recognition, Image Recognition, Intelligent Transportation Systems, Ensemble Learning, Transfer Learning

Abstract

The rapid growth of vehicles in Bangladesh has exacerbated traffic congestion, safety concerns, and administrative difficulties, creating an urgent need for Intelligent Transportation Systems (ITS). However, existing global datasets poorly represent Bangladeshi vehicles, and no region-specific detection models are currently available. To address this gap, this research constructs a balanced augmented dataset combining two public repositories, yielding images across ten vehicle classes. A harmonized dataset of manually verified images was split prior to augmentation, expanding the training set to 24,696 samples. Five deep Convolutional Neural Network architectures are then evaluated using transfer learning and extensive augmentation, and a convex ensemble of three selected models with optimized weights is developed to enhance robustness. The single ConvNeXt model achieves 98.980% ± 0.198% accuracy, while the ensemble attains 98.991% ± 0.156% accuracy. Statistical testing confirms the significance of these results; the ensemble effectively reduces misclassifications between visually similar categories. Overall, the proposed system provides a dependable platform for transportation applications in Bangladesh. Future work will address current limitations in static image and category scope through video-based analysis and lightweight deployment strategies.

Cite This Paper

Shamsun Nahar, Md.Nazim Hossain, Noshin Un Noor, Mohammad Anwar Hossain, Md. Sabbir Hossain Babu, Md. Sohorab Hossen, "Region-Specific Vehicle Classification in Bangladesh: A Comparative Study of CNN Architectures and Ensemble Strategies", International Journal of Engineering and Manufacturing (IJEM), Vol.16, No.5, pp. 106-121, 2026. DOI:10.5815/ijem.2026.05.06

Reference

[1]World Health Organization, Global Status Report on Road Safety 2023. Geneva: World Health Organization, 2023. (ISBN 978-92-4-008651-7; no DOI)
[2]F. Hermens, “Automatic object detection for behavioural research using YOLOv8,” Behav. Res. Methods, vol. 56, 2024. * https://doi.org/10.3758/s13428-024-02420-5
[3]M. Maity, S. Banerjee and S. Sinha Chaudhuri, "Faster R-CNN and YOLO based Vehicle detection: A Survey," 2021 5th International Conference on Computing Methodologies and Communication (ICCMC), Erode, India, 2021, pp. 1442-1447, doi: 10.1109/ICCMC51019.2021.9418274. https://doi.org/10.1109/ICCMC51019.2021.9418274
[4]Huang, Q. & Li, Z. (2024) "Improvement of YOLOv5s-Ghost model," Journal of Physics: Conference Series, vol. 2816, no. 1, p. 012081. https://doi.org/10.1088/1742-6596/2816/1/012081
[5]R. Gavrilescu et al., “Faster R-CNN: An approach to real-time object detection,” in Proc. IEEE Int. Conf. Elect. Power Eng. (ICEPE), Iasi, Romania, 2018, pp. 0165–0168. * https://doi.org/10.1109/ICEPE.2018.8559776
[6]S. Yacoob, D. Kumar, G. G. S. Harshitha, V. Y. Teja, C. H. Veni, and K. Reddy, “Enhancing object detection robustness in adverse weather conditions,” Procedia Comput. Sci., vol. 252, pp. 1014–1024, 2025. * https://doi.org/10.1016/j.procs.2025.01.062
[7]I. Sarker, S. Rahman, N. K. Nuha, A. A. Mofael, M. S. Islam, N. S. Sworna, and D. M. Farid, “Bangladeshi vehicle classification using transfer learning with YOLOv7,” in Proc. 6th Int. Conf. Electr. Inf. Commun. Technol. (EICT), Khulna, Bangladesh, Dec. 2023, pp. 1–6. https://doi.org/10.1109/EICT61409.2023.10427682
[8]S. U. Salekin, M. H. Ullah, A. A. A. Khan, M. S. Jalal, H. H. Nguyen, and D. M. Farid, “Bangladeshi native vehicle classification employing YOLOv8,” in Int. Conf. Intell. Syst. Data Sci. (ISDS), Singapore, 2023, pp. 185–199. * https://doi.org/10.1007/978-981-99-7649-2_14
[9]X. Dong, S. Yan, and C. Duan, “A lightweight vehicles detection network model based on YOLOv5,” Eng. Appl. Artif. Intell., vol. 113, Aug. 2022, Art. no. 104914. * https://doi.org/10.1016/j.engappai.2022.104914
[10] Wang, X., Zhang, W., Wu, X. et al. Real-time vehicle type classification with deep convolutional neural networks. J Real-Time Image Proc 16, 5–14 (2019). https://doi.org/10.1007/s11554-017-0712-5
[11]V.-T. Tran and W.-H. Tsai, “Acoustic-based emergency vehicle detection using convolutional neural networks,” IEEE Access, vol. 8, pp. 75702–75713, 2020. * https://doi.org/10.1109/ACCESS.2020.2988986
[12]Mandal, V., Adu-Gyamfi, Y. Object Detection and Tracking Algorithms for Vehicle Counting: A Comparative Analysis. J. Big Data Anal. Transp. 2, 251–261 (2020). https://doi.org/10.1007/s42421-020-00025-w
[13]S. Aqel, A. Hmimid, M. A. Sabri, and A. Aarab, “Road traffic: Vehicle detection and classification,” in Proc. Intell. Syst. Comput. Vis. (ISCV), Apr. 2017, pp. 1–5. * https://doi.org/10.1109/ISACV.2017.8054969
[14]M. Betke, E. Haritaoglu, and L. S. Davis, “Real-time multiple vehicle detection and tracking from a moving vehicle,” Mach. Vis. Appl., vol. 12, no. 2, pp. 69–83, Aug. 2000. * https://doi.org/10.1007/s001380050126
[15]A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” in Proc. Adv. Neural Inf. Process. Syst., 2012, pp. 1097–1105. * https://doi.org/10.1145/3065386
[16]K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), Las Vegas, NV, USA, 2016, pp. 770–778. * https://doi.org/10.1109/CVPR.2016.90
[17]N. Dalal and B. Triggs, “Histograms of oriented gradients for human detection,” in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), San Diego, CA, USA, 2005, pp. 886–893. * https://doi.org/10.1109/CVPR.2005.177
[18]P. K. Das, M. Islam, E. Sina, and D. M. Farid, “Bangladeshi vehicle classification and detection using deep convolutional neural networks with transfer learning,” IEEE Access, vol. 13, pp. 26429– 26455, 2025. https://doi.org/10.1109/ACCESS.2025.3539713
[19]A. A. A. Khan, S. U. Salekin, M. S. Rahman, and D. M. Farid, “Deep learning for traffic data mining,” in Proc. 6th Int. Conf. Electr. Inf. Commun. Technol. (EICT), Khulna, Bangladesh, 2023, pp. 1–6. https://doi.org/10.1109/EICT61409.2023.10427697
[20]A. S. M. Siam, M. M. Hasan, M. M. Talukdar, M. Y. Arafat, S. H. Jobayer, and D. M. Farid, “Bangla news classification employing deep learning,” in Int. Conf. Intell. Syst. Data Sci. (ISDS), Singapore, 2023, pp. 155–169. https://doi.org/10.1007/978-981-99-7649-2_12
[21]Y. Tian, Q. Ye, and D. Doermann, “YOLOv12: Attention-centric real-time object detectors,” arXiv:2502.12524, 2025. * https://doi.org/10.48550/arXiv.2502.12524
[22]M. Tan and Q. Le, “EfficientNetV2: Smaller models and faster training,” arXiv:2104.00298, 2021. * https://doi.org/10.48550/arXiv.2104.00298
[23]Z. Liu et al., “A ConvNet for the 2020s,” in Proc. IEEE/CVF Conf. Comput. Vis. Pattern Recognit. (CVPR), New Orleans, LA, USA, 2022, pp. 11966–11976. * https://doi.org/10.1109/CVPR52688.2022.01167
[24]K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv:1409.1556, 2014. * https://doi.org/10.48550/arXiv.1409.1556
[25]T. Z. Sana, S. Islam, N. Hasan, and I. Ahmad, “Sorokh-Poth: A balanced Bangladeshi road vehicle image dataset integrated with a detection system based on deep convolutional neural networks,” International Journal of Research Publications, vol. 122, no. 1, 2023. https://doi.org/10.47119/IJRP1001221420234574
[26]S. Tabassum, M. S. Ullah, N. H. Al-Nur, and S. Shatabda, “Native vehicles classification on Bangladeshi roads using CNN with transfer learning,” in Proc. IEEE Region 10 Symp. (TENSYMP), Dhaka, Bangladesh, 2020. https://doi.org/10.1109/TENSYMP50017.2020.9230991
[27]S. Tabassum, S. Ullah, N. H. Al-Nur, and S. Shatabda, “Poribohon-BD: Bangladeshi local vehicle image dataset with annotation for classification,” Data Brief, vol. 33, p. 106465, 2020. https://doi.org/10.1016/j.dib.2020.106465
[28]T. Kumar, M. Turab, K. Raj, A. Mileo, R. Brennan, and M. Bendechache, “Advanced data augmentation approaches: A comprehensive survey and future directions,” arXiv:2301.02830, 2023. * https://doi.org/10.48550/arXiv.2301.02830
[29]S. Kotsiantis, D. Kanellopoulos, and P. Pintelas, “Data preprocessing for supervised learning,” Int. J. Comput. Sci., vol. 1, pp. 111–117, 2006. * https://doi.org/10.5281/zenodo.1082415
[30]A. Hassan and M. Sabha, “Feature extraction for image analysis and detection using machine learning techniques,” Int. J. Adv. Networking Appl., vol. 14, pp. 5499–5508, 2023. * https://doi.org/10.35444/IJANA.2023.14401
[31]Alamu, Rapheal. (2023). Automating Data Annotation and Labeling with AI: A Machine Learning Perspective. Artificial Intelligence.

International Journal of Engineering and Manufacturing (IJEM)