The Journal of The Korea Institute of Intelligent Transport Systems (한국ITS학회 논문지)

The Korea Institute of Intelligent Transport Systems (한국ITS학회)
권호 표지
DOI QR코드

DOI QR Code

GAP Estimation on Arterial Road via Vehicle Labeling of Drone Image

드론 영상의 차량 레이블링을 통한 간선도로 차간간격(GAP) 산정

  • Jin, Yu-Jin (Dept. of Earth and Environmental System Sciences National Univ. of Pukyong) ;
  • Bae, Sang-Hoon (Dept. of Spatial Information Engineering System, National Univ. of Pukyong)
  • 진유진 (부경대학교 지구환경시스템과학부) ;
  • 배상훈 (부경대학교 공간정보시스템공학과)
  • Received : 2017.10.23
  • Accepted : 2017.12.19
  • Published : 2017.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

The purpose of this study is to detect and label the vehicles using the drone images as a way to overcome the limitation of the existing point and section detection system and vehicle gap estimation on Arterial road. In order to select the appropriate time zone, position, and altitude for the acquisition of the drone image data, the final image data was acquired by shooting under various conditions. The vehicle was detected by applying mixed Gaussian, image binarization and morphology among various image analysis techniques, and the vehicle was labeled by applying Kalman filter. As a result of the labeling rate analysis, it was confirmed that the vehicle labeling rate is 65% by detecting 185 out of 285 vehicles. The gap was calculated by pixel unitization, and the results were verified through comparison and analysis with Daum maps. As a result, the gap error was less than 5m and the mean error was 1.67m with the preceding vehicle and 1.1m with the following vehicle. The gaps estimated in this study can be used as the density of the urban roads and the criteria for judging the service level.

본 연구에서는 기존 지점 및 구간 검지체계의 한계를 극복하기 위한 방편으로 드론 촬영영상을 활용하여 차량을 검지 및 레이블링 하고 이를 기반으로 도심부 간선도로상 차간간격을 산정하는 것을 목적으로 한다. 드론 영상 데이터 획득 시 적정 시간대, 위치, 고도를 선정하기 위하여 여러 조건하에서 촬영을 실시하여 최종 영상 데이터를 획득하였다. 다양한 영상분석기법 중 혼합 Gaussian, 영상 이진화, 모폴로지 기법을 적용시켜 차량을 검지하였고 칼만 필터를 적용하여 차량을 레이블링 하였다. 레이블링율 분석 결과 실제 차량 수 285대 중 185대를 검지함으로써 차량 레이블링율은 65%로 나타나는 것을 확인하였다. 차간간격은 픽셀 단위화를 통해 산정하였으며, 결과는 다음 지도와의 비교 분석을 통해 검증을 수행하였다. 검증 결과 차간간격 오차가 모두 5m 미만으로 나타났으며 평균 오차는 선행차량과의 차간간격은 1.67m, 후행차량과의 차간간격은 1.1m로 분석되었다. 본 연구에서 산출된 차간간격은 도심부 도로의 밀도, 서비스 수준 판단 기준 설정 등으로 활용될 수 있을 것이다.

Keywords

References

  1. Bhaskar P. K. and Youg S.(2014), "Image Processing Based Vehicle Detection and Tracking Method," International Conference on Computer and Information Sciences, pp.1-5.
  2. Cao L., Wang C. and Li J.(2016), "Vehicle detection from highway satellite images via transfer learning," Information sciences, vol. 366, pp.177-187. https://doi.org/10.1016/j.ins.2016.01.004
  3. Chen Z., Ellis T., Velastin S.(2012), "Vehicle detection, tracking and classification in urban traffic," Conference Record - IEEE Conference on Intelligent Transportation Systems, pp.951-956.
  4. Han S. H. and Lee K. H.(2003), "A Study on The Detection of Multiple Vehicles Using Sequence Image Analysis," The Korea Society of Computer and Information, vol. 8, no. 2, pp.37-43.
  5. Kim G. S. and Cho J. S.(2010), "Vehicle Detection and Distance Estimation Based on Camera Vision System," The Institute of Electronics and Information Engineers, pp.1207-1210.
  6. Lee A. R.(2017), Convergence Weekly TIP, Convergence Research Policy Center, vol. 53, pp.2-11.
  7. Li L. D., Xu Q. W. and Jiang H.(2017), "Detection Method for Vehicles in Tunnels Based on Surveillance Images," IEEE Conference Publications - 2017 4th International Conference on Transportation Information and Safety (ICTIS), pp.836-843.
  8. National Law Information Center(2017), Road Traffic Act, asterisk 6.
  9. R C., Gang Q. and Qinfen Z.(2004) "Vehicle detection and tracking using acoustic and video sensors," IEEE International Conference on Acoustics, Speech and Signal Processing, vol. 5, pp.793-796.
  10. Shin M. S. and Bae S. H.(2010), "Use of Aerial Photography for Estimating Traffic Information by Vehicle Gap Distances," Transport Institute, vol. 17, no. 2, pp.73-82.