Automation of Traffic Violation Detection and Penalization
- 17 Downloads
Violation of traffic rules in two-wheelers might risk the life of persons traveling. Finding the violation of traffic rules requires a large number of police personals at several locations. In this paper, we have proposed a system that directly finds traffic violations done by two-wheelers using trained convolutional neural networks (CNN) without any human assistance. CNN uses hidden layers to perform operations, which allows applying a wide variety of filters in each layer. CNN produces better accuracy when compared with traditional image processing techniques. Violated vehicle number plates are found using optical character recognition (OCR) techniques and penalized according to the violation.
KeywordsDeep learning Traffic violation Video processing CNN OCR
The authors gratefully acknowledge CTDT, Anna University for providing financial support to carry out this research work under the student innovative projects scheme. The authors also gratefully acknowledge the Web site: https://www.trickspagal.com/2017/08/trace-vehicle-number-with-owner-name-and-address.html for providing the license plate image to test the proposed system. The authors also thank Big Data Analytics Laboratory of MIT campus, Anna University for infrastructure and support.
- 1.Aarthy, D.K., Vandanaa, S., Varshini, M., Tijitha, K.: Automatic identification of traffic violations and theft avoidance. In: Second International Conference on Science Technology Engineering and Management, pp. 72–76 (2016)Google Scholar
- 2.Marikhu, R., Moonrinta, J., Ekpanyapong, M., Dailey, M., Siddhichai, S.: Police eyes: real world automated detection of traffic violations. In: 10th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, pp. 1–6 (2013)Google Scholar
- 4.Ren, S., He, K., Girshick, R., Sun, J.: Faster R-CNN: towards real-time object detection with region proposal networks. In: Advances in Neural Information Processing Systems, pp. 91–99 (2015)Google Scholar
- 7.Liu, R., Xu, S., Park, J., Murphey, Y.L., Kristinsson, J., McGee, R., Kuang, M., Phillips, T.: Real time vehicle speed prediction using gas-kinetic traffic modeling. In: IEEE Symposium on Computational Intelligence in Vehicles and Transportation Systems, pp. 80–86 (2011)Google Scholar
- 9.Park, J., Murphey, Y.L., Kristinsson, J., McGee, R., Kuang, M., Phillips, T.: Intelligent speed profile prediction on urban traffic networks with machine learning. In: 2013 International Joint Conference on Neural Networks, pp. 1–7 (2013)Google Scholar
- 11.Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, pp. 6517–6525 (2017)Google Scholar
- 12.Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: unified, real-time object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 779–788 (2016)Google Scholar
- 14.Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: Ssd: Single shot multibox detector. In: European Conference on Computer Vision, pp. 21–37. Springer, Cham (2016)Google Scholar