Deep Learning-Based Classification of Indian Road Vehicles Using a Custom Dataset and Pretrained Models
Received: 27 June 2025 | Revised: 14 August 2025 | Accepted: 26 August 2025 | Online: 8 December 2025
Corresponding author: Bhakti Paranjape
Abstract
Vehicle classification is a vital component of intelligent transportation systems, enabling applications such as traffic monitoring, automated parking, vehicle restrictions, autonomous driving, toll collection, and expressway traffic analysis. On Indian roads, a diverse range of uniquely designed vehicles—such as trucks, rickshaws, and certain bike types—poses distinct classification challenges. Traditional image processing and pattern recognition approaches, relying on handcrafted features and limited datasets, often struggled under real-world conditions affected by lighting, weather, and environmental variability. Deep learning methods have overcome many of these limitations, yet a region-specific benchmark dataset for Indian vehicles has been lacking. To address this gap, a custom dataset, the Bharatiya Vehicle Dataset (BhVD), was developed, containing four prominent vehicle categories: cars, bikes, trucks, and rickshaws. In this work, thirteen pretrained Keras models, originally trained on ImageNet, were fine-tuned and evaluated on the BhVD and two other Indian traffic datasets. The models were compared across accuracy, inference speed, and real-time applicability. The InceptionResNetV2 model achieved the highest accuracy of 94.34% on the BhVD, whereas MobileNet proved the fastest, with an inference time of 60 ms. The results demonstrate the effectiveness of transfer learning for region-specific vehicle classification and provide insights into selecting models that balance speed and accuracy for real-world deployment in Indian traffic environments.
Keywords:
classification, object detection, Indian vehicles, Convolutional Neural Networks (CNNs)Downloads
References
N. U. A. Tahir, Z. Zhang, M. Asim, J. Chen, and M. ELAffendi, "Object Detection in Autonomous Vehicles under Adverse Weather: A Review of Traditional and Deep Learning Approaches," Algorithms, vol. 17, no. 3, Mar. 2024, Art. no. 103. DOI: https://doi.org/10.3390/a17030103
S. Cao, "Review of Object Detection Challenges in Autonomous Driving," Applied and Computational Engineering, vol. 8, no. 1, pp. 707–713, Aug. 2023. DOI: https://doi.org/10.54254/2755-2721/8/20230306
M. Won, "Intelligent Traffic Monitoring Systems for Vehicle Classification: A Survey," IEEE Access, vol. 8, pp. 73340–73358, 2020. DOI: https://doi.org/10.1109/ACCESS.2020.2987634
M. A. Butt et al., "Convolutional Neural Network Based Vehicle Classification in Adverse Illuminous Conditions for Intelligent Transportation Systems," Complexity, vol. 2021, no. 1, Feb. 2021, Art. no. 6644861. DOI: https://doi.org/10.1155/2021/6644861
"Vehicle Classification - Dataset." Kaggle. [Online]. Available: https://www.kaggle.com/datasets/maciejgronczynski/vehicle-classification-dataset.
M. Maher, "Vehicle Image Classification." Kaggle. [Online]. Available: https://www.kaggle.com/datasets/mohamedmaher5/vehicle-classification.
"Indian Vehicle Dataset." Kaggle. [Online]. Available: https://www.kaggle.com/datasets/dataclusterlabs/indian-vehicle-dataset.
"Vehicle Type Recognition." Kaggle. [Online]. Available: https://www.kaggle.com/datasets/kaggleashwin/vehicle-type-recognition.
Z. Luo et al., "MIO-TCD: A New Benchmark Dataset for Vehicle Classification and Localization," IEEE Transactions on Image Processing, vol. 27, no. 10, pp. 5129–5141, Oct. 2018. DOI: https://doi.org/10.1109/TIP.2018.2848705
B. Paranjape, "Bharatiya Vehicles Dataset (BhVD)." Kaggle, 2025.
G. Varma, A. Subramanian, A. Namboodiri, M. Chandraker, and C. V. Jawahar, "IDD: A Dataset for Exploring Problems of Autonomous Navigation in Unconstrained Environments," in 2019 IEEE Winter Conference on Applications of Computer Vision, Waikoloa, HI, USA, 2019, pp. 1743–1751. DOI: https://doi.org/10.1109/WACV.2019.00190
V. K. Kiran, S. Dash, and P. Parida, "Improvement on Deep Features through Various Enhancement Techniques for Vehicles Classification," Sensing and Imaging, vol. 22, no. 1, Sep. 2021, Art. no. 41. DOI: https://doi.org/10.1007/s11220-021-00363-1
B. Hicham, A. Ahmed, and M. Mohammed, "Vehicle Type Classification Using Convolutional Neural Network," in 2018 IEEE 5th International Congress on Information Science and Technology, Marrakech, Morocco, 2018, pp. 313–316. DOI: https://doi.org/10.1109/CIST.2018.8596500
S. D. Badiger and M. UttaraKumari, "Vehicle Classification Using Machine Learning Algorithms Based on the Vehicular Acoustic Signature," Science, Technology and Development, vol. 8, no. 11, pp. 369–374, Nov. 2019.
L. Suhao, L. Jinzhao, L. Guoquan, B. Tong, W. Huiqian, and P. Yu, "Vehicle type detection based on deep learning in traffic scene," Procedia Computer Science, vol. 131, pp. 564–572, Jan. 2018. DOI: https://doi.org/10.1016/j.procs.2018.04.281
H. Shokravi, H. Shokravi, N. Bakhary, M. Heidarrezaei, S. S. Rahimian Koloor, and M. Petrů, "A Review on Vehicle Classification and Potential Use of Smart Vehicle-Assisted Techniques," Sensors, vol. 20, no. 11, Jun. 2020, Art. no. 3274. DOI: https://doi.org/10.3390/s20113274
D. Sharma, Z. A. Jaffery, and N. Ahmad, "Categorical vehicle classification using Deep Neural Networks," in 2019 International Conference on Power Electronics, Control and Automation, New Delhi, India, 2019, pp. 1–6. DOI: https://doi.org/10.1109/ICPECA47973.2019.8975613
V. Saikrishnan and M. Karthikeyan, "Mayfly Optimization with Deep Learning-based Robust Object Detection and Classification on Surveillance Videos," Engineering, Technology & Applied Science Research, vol. 13, no. 5, pp. 11747–11752, Oct. 2023. DOI: https://doi.org/10.48084/etasr.6231
T. Sharma et al., "Deep Learning-Based Object Detection and Classification for Autonomous Vehicles in Different Weather Scenarios of Quebec, Canada," IEEE Access, vol. 12, pp. 13648–13662, 2024. DOI: https://doi.org/10.1109/ACCESS.2024.3354076
A. Gholamhosseinian and J. Seitz, "Vehicle Classification in Intelligent Transport Systems: An Overview, Methods and Software Perspective," IEEE Open Journal of Intelligent Transportation Systems, vol. 2, pp. 173–194, 2021. DOI: https://doi.org/10.1109/OJITS.2021.3096756
Md. A. Hossain, Md. A. Rahman, R. Khanom, S. Sultana, and A. Hossain, "Real-time Vehicle Detection and Classification on the Padma Multipurpose Bridge in Bangladesh Using a Deep Learning Model," in 2023 International Conference on Next-Generation Computing, IoT and Machine Learning, Gazipur, Bangladesh, 2023, pp. 1–6. DOI: https://doi.org/10.1109/NCIM59001.2023.10212549
"Keras documentation: Keras Applications." Keras. https://keras.io/api/applications/.
O. Russakovsky et al., "ImageNet Large Scale Visual Recognition Challenge," International Journal of Computer Vision, vol. 115, no. 3, pp. 211–252, Dec. 2015. DOI: https://doi.org/10.1007/s11263-015-0816-y
Downloads
How to Cite
License
Copyright (c) 2025 Bhakti Paranjape, Apurva Naik
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
