Performance Evaluation for Vision-Based Vehicle Classification Using Convolutional Neural Network

  • Authors

    • Raja Durratun Safiyah
    • Zaid Abdul Rahim
    • Syamsul Syafiq
    • Zaidah Ibrahim
    • Nurbaity Sabri
    https://doi.org/10.14419/ijet.v7i3.15.17507

    Received date: August 14, 2018

    Accepted date: August 14, 2018

    Published date: August 13, 2018

  • Vision Based Vehicle Classification, Convolutional Neural Network (CNN), Deep Learning Training from Scratch, AlexNet, GoogleNet

  • Abstract

    Vision-based vehicle classification is a very challenging task due to vehicle pose and angle variations, weather conditions, lighting quality, and limited number of available datasets for training.  It can be applied for driver assistance system and autonomous vehicles.  This paper conducted a performance evaluation for this task based on three Convolutional Neural Network (CNN) models, which are simple CNN, and pre-trained CNN models that are AlexNet and GoogleNet.  A dataset of more than 7000 images from the Image Processing Group (IPG) has been used for training and testing and the results indicate that AlexNet achieves the best classification result that is 65.09%. This result is obtained because of the variations of the quality of the images.  

  • References

    1. H. Salman, J. Grover, and T. Shankar, “Hierarchical Reinforcement Learning for Sequencing Behaviors,” vol. 2449, pp. 2352–2449, 2018.
    2. C. H. Samer, K. Rishi, and Rowen, “Image Recognition Using Convolutional Neural Networks,” Cadence Whitepaper, pp. 1–12, 2015.
    3. J. T. Lee and Y. Chung, “Deep Learning-Based Vehicle Classifica-tion Using an Ensemble of Local Expert and Global Networks,” IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit. Work., vol. 2017–July, pp. 920–925, 2017.
    4. E. T. S. I. T. Grupo de Tratamiento de Imágenes (GTI), “Vehicle Images Database,” 2011. [Online]. Available: https://www.gti.ssr.upm.es/data/Vehicle_database.html. [Accessed: 17-Apr-2018].
    5. W. Ong Vui Jiunn, N. Sabri and Z. Ibrahim, “Image-based Human Fall Recognition using Gaussian Mixture Model and Support Vec-tor Machine”, International Journal of Control Theory and Applica-tions, vol. 9,number 44, 2016
    6. Z. Ibrahim, N. Sabri and N. N. Mohd Manghor, “Leaf Recognition
    7. Using Texture Features for Herbal Plant Identification’, Interna-tional Journal of Electrical Engineering and Computer Science (IJEECS),Vol.9, No. 1 2018, pp.152-156.
    8. N. Sabri and Z. Ibrahim, “Palm Oil Fresh Fruit Bunch Ripeness Grading Identification using Color Features”, Journal of Fundamen-tal and Applied Science, 2017, 9(4S), pp. 563-579.
    9. A. Krizhevsky and G. E. Hinton, “ImageNet Classification with Deep Convolutional Neural Networks,” pp. 1–9.
    10. R.P. Avery, Y. Wang, G.S. Rutherford, “Length-Based Vehicle-Classification Using Images from Uncalibrated Video Cameras,” in:Proceedings of the 7th International IEEE Conference on Intelli-gentTransportation System, pp.737-742, 2004.
    11. G. Zhang, R.P. Avery, Y. Wang, “A Video-Based Vehicle Detec-tion andClassification System for Real-Time Traffic Data Collection UsingUncalibrated Video Cameras,” Transportation Research Rec-ord: Journalof the Transportation Research Board, 1993: 138-147, 2007.
    12. G. Moussa and K. Hussain, “Laser Intensity Automatic Vehicle-Classification System,” North American Travel Monitoring Exposi-tionand Conference (NATMEC)”, Washington, DC, USA, August 6-8,2008. G. Zhang, R.P. Avery, Y. Wang, “A Video-Based Vehi-cle Detection andClassification System for Real-Time Traffic Data Collection UsingUncalibrated Video Cameras,” Transportation Re-search Record: Journalof the Transportation Research Board, 1993: 138-147, 2007.
    13. X. Ma, W. Eric, and L. Grimson, “Edge-Based Rich Representation forVehicle Classification”, Proc. Int. Conf. Computer Vision, vol. 2, pp.1185- 1192, 2005
    14. L. Zhang, S.Z. Li, X. Yuan, S. Xiang, “Real-Time Object Classifi-cationin Video Surveillance Based On Appearance Learning,” in: Proceedingsof IEEE Conference on Computer Vision and Pattern Recognition, 1-8,2007.
    15. E. Okafor, M. A. Wiering, E. Okafor, P. Pawara, F. Karaaba, and O. Surinta, “Comparative Study Between Deep Learning and Bag of Visual Words for Wild-Animal Recognition Comparative Study Be-tween Deep Learning and Bag of Visual Words for Wild-Animal Recognition,” no. December, 2016.
    16. L. F. Rodrigues, M. C. Naldi, and J. F. Mari, “HEp-2 Cell Image Classification Based on Convolutional Neural Networks,” 2017 Work. Comput. Vis., pp. 13–18, 2017.

  • Downloads

  • How to Cite

    Durratun Safiyah, R., Abdul Rahim, Z., Syafiq, S., Ibrahim, Z., & Sabri, N. (2018). Performance Evaluation for Vision-Based Vehicle Classification Using Convolutional Neural Network. International Journal of Engineering and Technology, 7(3.15), 86-90. https://doi.org/10.14419/ijet.v7i3.15.17507