DOI QR코드

DOI QR Code

A study on the Traffic Density Collect System using View Synthesis and Data Analysis

영상정합을 이용한 교통밀도 수집방법과 수집 데이터 비교분석

  • Park, Bumjin (Korea Institute of Civil Engineering and Building Technology, Department of Future Technology and convergence) ;
  • Roh, Chang-gyun (Korea Institute of Civil Engineering and Building Technology, Department of Future Technology and convergence)
  • 박범진 (한국건설기술연구원 미래융합연구본부) ;
  • 노창균 (한국건설기술연구원 미래융합연구본부)
  • Received : 2018.08.14
  • Accepted : 2018.09.17
  • Published : 2018.10.31

Abstract

Traffic Density is the most important of the three primary macroscopic traffic stream parameters, because it is most directly related to traffic demand(Traffic Engineering, 2004). It is defined as the number of existing vehicles within a given distance at a certain time. However, due to weather, road conditions, and cost issues, collecting density directly on the field is difficult. This makes studies of density less actively than those of traffic volume or velocity. For these reasons, there is insufficient attempts on divers collecting methods or researches on the accuracy of measured values. In this paper, we used the 'Density Measuring System' based on the synthesise technology of several camera images as a method to measure density. The collected density value by the 'Density Mesuring System' is selected as the true value based on the density define, and this value was compared with the density calculated by the traditional measurement methods. As a result of the comparison, the density value using the fundamental equation method is the closest to the true value as RMSE shows 1.8 to 2.5. In addition, we investigated some issues that can be overlooked easily such as the collecting interval to be considered on collecting density directly by calculating the moment density and the average density. Despite the actual traffic situation of the experiment site is LOS B, it is difficult to judge the real traffic situation because the moment density values per second are observed max 16.0 (veh/km) to min 2.0 (veh/km). However, the average density measured for 15 minutes at 30-second intervals was 8.3-7.9 (veh/km) and it indicates precisely LOS B.

교통밀도는 교통수요와 가장 직접적인 관계가 있으므로 거시적인 지표 중에서 가장 중요하다고 알려져 있으며(Traffic Engineering, 2004), 특정시각 주어진 거리 안의 존재하는 차량의 대수로 정의한다. 하지만, 밀도는 기상과 도로조건 및 비용 상의 문제로 인하여 현장에서 직접 수집이 어렵다. 이런 이유로 속도와 교통량보다 상대적으로 연구가 활발하게 이루어지지는 않아 수집방법에 관한 다양한 시도뿐만 아니라 측정된 값의 정확도에 관한 연구가 많이 부족하다. 이에 본 논문에서는 밀도를 측정할 수 있는 방법으로 여러 대의 카메라 영상을 정합(synthesis)하는 기술을 활용하였다. 이러한 밀도수집시스템으로 수집한 밀도를 정의에 기반한 참값으로 선정하고, 이 값을 전통적인 측정방법들로 산출한 밀도와 비교하였다. 비교결과, 관계식(fundamental equation)을 이용한 산출방법으로 산출한 밀도 값이 참값과 비교하여 RMSE(Root Mean Square Error)가 1.8-2.5로 가장 참값에 가깝다. 또한 향후 밀도를 직접 수집할 때 유의하여할 수집 간격 등의 간과하기 쉬운 이슈사항을 순간밀도와 평균밀도를 산출하여 알아보았다. 실험 사이트의 실제 교통상황은 LOS B임에도 불구하고, 초 당 순간밀도는 최대(16veh/km)에서 최소 2(veh/km)의 값이 다양하게 관측되어 교통상황의 판단이 어려웠다. 하지만 30초 간격으로 15분 평균밀도는 8.3-7.9(veh/km)로 정확하게 LOS B를 판단하였다.

Keywords

References

  1. Haberman R. (1997), Mathematical Models, Prentice Hall: Upper Saddle River, N. J., pp.265-394.
  2. Hall F. L. (1986), "The relationship between occupancy and density," Transportation Forum, 3-3, pp.46-51.
  3. Hall F. L., Wakefield S. and Al-kaisy A. (2001),"Free quality of service - what really matters to drivers and passengers?," Transportation Research Record, 1776, pp.17-23. https://doi.org/10.3141/1776-03
  4. KICT (2015), Development of Realtime Traffic Tracking Technology Based on View Synthesis (5th year).
  5. Kim M.-S., Eom K.-J. and Lee C.-W. (2009), "Density Measurement for Continuous Flow Segment Using Two Point Detectors," Journal of Korean Society of Intelligent Transportation Systems, vol. 8 no. 1, pp.37-44.
  6. Koshi M., Iwasaki M. and Ohkura I. (1983), "Some findings and an overview on vehicular flow characteristics," In Proceedings of the Eighth International Symposium on Transportation and Traffic Theory, Toronto, Canada, pp.403-426.
  7. May A. D. (1990), Traffic Flow Fundamentals, Prentice-Hall: Upper Saddle River, N. J., pp.192-226.
  8. Newell G. F. (1992), "A simplified theory of kinematic waves in highway traffic, part I: general theory," Transportation Research, vol. 27, no. 4, pp.281-287.
  9. Ozkurt C. and Camci F. (2009), "Automatic traffic density estimation and vehicle classification for surveillance systems using neural networks," Mathematical and Computational Applications, vol. 14, no. 3, pp.187-196. https://doi.org/10.3390/mca14030187
  10. Park B. J., Kim T. H., Yang I. C., Heo J. Y. and Son B. S. (2015), "A Method for Measuring accurate traffic density by aerial photography," Journal of Advanced Transportation, vol, 49, pp.568-580. https://doi.org/10.1002/atr.1288
  11. Park B. J., Roh C.-G. and Kim J. S. (2014), "A Case Study of Panoramic Section Image Collection Method for Measuring Density -with matched images in the Seoul Beltway Sapaesan Tunnel-," The Journal of The Korea Institute of Intelligent Transport Systems, vol. 13, no. 4, pp.20-29. https://doi.org/10.12815/kits.2014.13.4.020
  12. Roess R. P., Prassas E. S. and McShane W. R. (2004), Traffic Engineering, third edition, Pearson Prentice Hall: Upper Saddle River, N. J., pp.112-115.
  13. Transportation Research Board (2000), Highway Capacity Manual, Washington, DC, USA.