DOI QR코드

DOI QR Code

엣지 디바이스에서의 딥러닝 기반 차량 인식 및 속도 추정을 통한 스마트 횡단보도 시스템의 설계 및 구현

Design and Implementation of A Smart Crosswalk System based on Vehicle Detection and Speed Estimation using Deep Learning on Edge Devices

  • Jang, Sun-Hye (Department of Computer Engineering, Kumoh National Institute of Technology) ;
  • Cho, Hee-Eun (Department of Computer Engineering, Kumoh National Institute of Technology) ;
  • Jeong, Jin-Woo (Department of Computer Engineering, Kumoh National Institute of Technology)
  • 투고 : 2020.02.10
  • 심사 : 2020.02.28
  • 발행 : 2020.04.30

초록

최근 우리나라의 자동차 보급률이 증가함에 따라 교통사고 발생 건수 또한 증가하고 있다. 특히, 차량간 사고뿐만 아니라 횡단보도 근처에서의 인명 사고 또한 증가하고 있어 횡단보도 교통안전에 대한 주의가 더욱 요구되고 있다. 본 논문에서는 NVIDIA Jetson Nano급의 엣지 디바이스를 이용하여 횡단보도에 접근하는 차량을 인식하고 속도를 추정함으로써 횡단보도 주위 안전 상태를 예측하는 시스템을 제안한다. 딥러닝 기반 차량 위치 인식을 통하여 얻은 정보들을 바탕으로 다양한 기계 학습 기법을 학습시켜 차량 속도에 따른 위험 정도를 예측한다. 마지막으로, 실제 주행 영상을 이용한 실험 및 웹 시뮬레이션을 통해 제안하는 시스템의 성능과 활용 가능성을 검증하였다.

Recently, the number of traffic accidents has also increased with the increase in the penetration rate of cars in Korea. In particular, not only inter-vehicle accidents but also human accidents near crosswalks are increasing, so that more attention to traffic safety around crosswalks are required. In this paper, we propose a system for predicting the safety level around the crosswalk by recognizing an approaching vehicle and estimating the speed of the vehicle using NVIDIA Jetson Nano-class edge devices. To this end, various machine learning models are trained with the information obtained from deep learning-based vehicle detection to predict the degree of risk according to the speed of an approaching vehicle. Finally, based on experiments using actual driving images and web simulation, the performance and the feasibility of the proposed system are validated.

키워드

참고문헌

  1. Yonhap News Agency. 157 pedestrian traffic accident patients per day: seven out of ten crosswalks [Internet]. Available : https://www.yna.co.kr/view/AKR20180227095200017
  2. Dong-A ILBO. 62% of traffic fatalities on foot are 'smombie' [Internet]. Available : http://www.donga.com/news/article/all/20180517/90136145
  3. A. Kurniawan, A. Ramadlan and E. M. Yuniarno, "Speed Monitoring for Multiple Vehicle Using Closed Circuit Television (CCTV) Camera," in Proceeding of the IEEE International Conference on Computer Engineering, Network and Intelligent Multimedia (CENIM), pp.88-93, 2018.
  4. J. H. Kim, D. K. Shin, J. K. Kim, C. H. Kwon, and H. R. Byun, "Optical Flow Based Vehicle Counting and Speed Estimation in CCTV Videos," Journal of Broadcast Engineering, vol.22, no.4, pp.448-461, Jul. 2017 https://doi.org/10.5909/JBE.2017.22.4.448
  5. J. H. Kim, D. H. Kim, and W. S. Lee, "Real-time Measurement of Traffic Speed and Volume Based on CCTV Videos," in Proceeding of the KIIT Summer Conference, pp. 107-110, 2018
  6. N. H. Cho, S. Z. Byun, M. K. Lee, and Y. H. Kim, "Implementation of Pedestrian Crosswalk Monitoring System for Accident Prevention," The Journal of Korean Institute of Information Technology, vol. 14, no.4, pp. 45-50, Apr. 2016. https://doi.org/10.14801/jkiit.2016.14.4.45
  7. K. P. Kim, E. J. Do, and Y. S. Bae, "AI crosswalk system for safety of pedestrian", in Proceeding of Korean Institute of Information Technology Conference, pp. 1683-1685, 2016.
  8. W. Czajewski and M. Iwanowski, "Vision-based vehicle speed measurement method," in Proceeding of the International Conference on Computer Vision and Graphics pp. 308-315, 2010.
  9. A. Dehghani and A. Pourmohammad, "Single camera vehicles speed measurement," in Proceeding of the 8th Iranian Conference on Machine Vision and Image Processing (MVIP), pp.190-193, 2013.
  10. C. H. Xiao and N. H. C. Yung, "A novel algorithm for estimating vehicle speed from two consecutive images," in Proceeding of the IEEE Workshop on Applications of Computer Vision (WACV), pp. 12-12, 2007.
  11. H. Zhiwei, L. Yuanyuan, and Y. Xueyi, "Models of vehicle speeds measurement with a single camera", in Proceeding of the International Conference on Computational Intelligence and Security, pp. 283-286, 2007.
  12. D. C. Luvizon, B. T. Nassu, and R. Minetto, "A VideoBased System for Vehicle Speed Measurement in Urban Roadways," IEEE Transactions on Intelligent Transportation Systems, vol.18, no.6, pp. 1393-1404, Sep. 2016. https://doi.org/10.1109/TITS.2016.2606369
  13. C. Y. Cho, H. K. Yim, and M.. J. Lee, "Development of ICT-based road safety integrated facilities for pedestrian crossing," Journal of the Korea Academica-Industrial cooperatino Society, vol.18, no.12, pp. 93-99, Jun. 2017.
  14. S. H. Ku, J. Y. Jang, E. H. Jung, E. S. Choi, and D. H. Lee, "Integrated Crosswalk Pedestrian Safety System with Smartphone Application," in Proceedings of the Korean Institute of Information Technology Conference, pp. 349-350. 2019.
  15. G. B. Kim and B. K. Kim, "Using Arduino crossing the sidewalk safety systems," in Proceeding of the Korean Institute of Information Technology Conference, pp. 361-362, 2015.
  16. S. J. Bae, H. J. Choi, and G. M. Jeong, "YOLO Model FPS Enhancement Method for Determining Human Facial Expression based on NVIDIA Jetson TX1," Journal of Korea Institute of Information, Electronics, and Communication Technology, vol.12, no.5, pp. 467-474. 2019.
  17. D. Y. Chun, J. W. Choi, and H. Kim, "Two-Context CNN Inference with TensorRT," in Proceeding of the Institute of Electronics and Information Engineers Fall Conference, pp. 460-461, 2019.