Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회지)

Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회)
권호 표지
DOI QR코드

DOI QR Code

Application of Deep Learning Method for Real-Time Traffic Analysis using UAV

UAV를 활용한 실시간 교통량 분석을 위한 딥러닝 기법의 적용

  • Park, Honglyun (School of Drone & Transportation Engineering, Youngsan University) ;
  • Byun, Sunghoon (School of Computer Engineering, Youngsan University) ;
  • Lee, Hansung (School of Computer Engineering, Youngsan University)
  • Received : 2020.07.21
  • Accepted : 2020.08.19
  • Published : 2020.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Due to the rapid urbanization, various traffic problems such as traffic jams during commute and regular traffic jams are occurring. In order to solve these traffic problems, it is necessary to quickly and accurately estimate and analyze traffic volume. ITS (Intelligent Transportation System) is a system that performs optimal traffic management by utilizing the latest ICT (Information and Communications Technology) technologies, and research has been conducted to analyze fast and accurate traffic volume through various techniques. In this study, we proposed a deep learning-based vehicle detection method using UAV (Unmanned Aerial Vehicle) video for real-time traffic analysis with high accuracy. The UAV was used to photograph orthogonal videos necessary for training and verification at intersections where various vehicles pass and trained vehicles by classifying them into sedan, truck, and bus. The experiment on UAV dataset was carried out using YOLOv3 (You Only Look Once V3), a deep learning-based object detection technique, and the experiments achieved the overall object detection rate of 90.21%, precision of 95.10% and the recall of 85.79%.

급격한 도시화로 인해 출퇴근 시간의 차량 정체, 상시 정체지역 발생 등 다양한 교통문제들이 발생하고 있다. 이러한 교통문제들을 해결하기 위해서는 신속·정확한 교통량 예측 및 분석이 필요하다. ITS (Intelligent Transportation System)는 최신 ICT (Information and Communications Technology) 기술들을 활용하여 최적의 교통관리를 수행하는 시스템이며, 다양한 기법을 통해 신속·정확한 교통량을 분석하기 위한 많은 연구가 수행 되었다. 본 연구에서는 높은 정확도로 실시간 교통량 분석을 위해 UAV (Unmanned Aerial Vehicle) 동영상을 활용한 딥러닝(deep learning) 기반의 차량탐지기법을 제안하고자 한다. 이를 위해, UAV를 활용하여 다양한 차량이 통행하는 교차로에서 학습 및 검증에 필요한 정사 동영상 촬영을 수행하였으며, 승용차(sedan), 트럭(truck), 버스(bus)로 분류하여 차량을 학습시켰다. 딥러닝 알고리즘은 대표적인 객체탐지 알고리즘 중의 하나인 YOLOv3 (You Only Look Once V3)를 이용하였으며, 실험결과 전체 차량 검출율은 90.21%이며, 정확도와 재현율은 각각 95.10%와 85.79%이다. 본 연구를 통하여, 드론을 이용한 영상으로부터 차량 탐지를 통한 실시간 교통량 분석이 가능함을 확인하였다.

Keywords

References

  1. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014), Rich feature hierarchies for accurate object detection and semantic segmentation, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 23-28 June, Columbus, USA, pp. 580-587.
  2. Girshick, R. (2015), Fast R-CNN, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 7-13 December, Santiago, Chile, pp. 1440-1448.
  3. Ha, D.M., Lee, J.M., and Kim, Y.D. (2001), A study on the image based traffic information extraction algorithm, Journal of Korean of Transformation, Vol. 19, No. 6, pp. 161-170.
  4. Hong, G.S., Eom, T.J., and Kim, B.G. (2011), Development of vision-based monitoring system technology for traffic analysis and surveillance, Journal of Information and Security, Vol. 11, No. 4, pp. 59-66. (in Korean with English abstract)
  5. Jo, S.H., Kim, C.G., Lim, H.Y., and Shin, Y.T. (2018), A study on the traffic flow analysis method based on change detection for traffic video data, Jounal of Information Technology and Architecture, Vol. 15, No. 13, pp. 373-382. (in Korean with English abstract)
  6. Ke, R., Li, Z., Kim, S., Ash, J., Cui, Z., and Wang, Y. (2016), Real-time bidirectional traffic flow parameter estimation from aerial videos, IEEE Transaction of Intelligent Transportation System, Vol. 18, No. 4, pp. 890-901. https://doi.org/10.1109/TITS.2016.2595526
  7. Lee, D.H. and Park, Y.T. (2004), Robust traffic monitoring system by spatio-temporal image analysis, Communications of the Korea Information Science Society, Vol. 31, No. 11, pp. 1534-1542. (in Korean with English abstract)
  8. Lee, H.S., Moon, D.S., Kim, I.K., Jung, H.S., and Park, D.H. (2015), Anomaly intrusion detection based on hyperellipsoid in the kernel feature space, KSII Transactions on Internet and Information Systems, Vol. 9, No. 3, pp. 1173-1192. https://doi.org/10.3837/tiis.2015.03.019
  9. Lee, K.Y. and Jang, M.S. (2005), Development of visionbased vehicle tracking for extracting microscopic traffic information, Journal of Korean of Transformation, Vol. 23, No. 7, pp. 137-148. (in Korean with English abstract)
  10. Nizar, T.N., Anbarsanti, N., and Prihatmanto, A.S. (2014), Multi-object tracking and detection system based on feature detection of the intelligent transportation system, 2014 IEEE 4th International Conference of System Engineering and Technology (ICSET), 24-25 November, Bandung, Indonesia, pp. 1-6.
  11. Peppa, M.V., Bell, D., Komar, T., and Xiao, W. (2018), Urban traffic flow analysis based on deep learning car detection from CCTV image series, The international Archives of the Photogrammetry, Remote Sensing and Spatial Information Science, Vol. XLII-4, pp. 499-506. https://doi.org/10.5194/isprs-archives-XLII-4-499-2018
  12. Redmon, J., Divvala, S., Girshick, R., and Farhadi, A. (2016) You only look once: unified, real-time object detection, IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 26 June - 1 July, Las Vegas, USA, pp. 779-788.
  13. Ren, S., He, K., Girshick, R., and Sun, J. (2016) Faster R-CNN: towards real-time object detection with region proposal networks, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 39, No. 6, pp. 1137-1149. https://doi.org/10.1109/TPAMI.2016.2577031
  14. Seo, S.H. and Lee, S.B. (2018), A study on traffic data collection and analysis for uninterrupted flow using Drones, Journal of The Korean Institute of Intelligent Transportation Systems, Vol. 17, No. 6, pp. 144-152. (in Korean with English abstract) https://doi.org/10.12815/kits.2018.17.6.144
  15. Seong, S.K., Song, J.H., Yoon D.H., Kim J.Y., and Choi, J.W. (2019), Determination of vehicle trajectory through optimization of vehicle bounding boxes using a convolution neural network, Sensors, Vol. 19, No. 19, pp. 4263-4281. https://doi.org/10.3390/s19194263
  16. Zhao, Z.Q., Zheng, P., Xu, S.T., and Wu, X. (2019), Object detection with deep learning: a review, IEEE Transactions on Neural Networks and Learning Systems, Vol. 30, No. 11, pp. 3212-3232. https://doi.org/10.1109/tnnls.2018.2876865