Journal of Korean Society of Transportation (대한교통학회지)

Korean Society of Transportation (대한교통학회)
권호 표지

New Method for Vehicle Detection Using Hough Transform

HOUGH 변환을 이용한 차량 검지 기술 개발을 위한 모형

  • Published : 1999.03.01
Download PDF
⟨ Previous Next ⟩

Abstract

Image Processing Technique has been used as an efficient method to collect traffic information on the road such as vehicle counts, speed, queues, congestion and incidents. Most of the current methods which have been used to detect vehicles by the image processing are based on point processing, dealing with the local gray level of each pixel in the small window. However, these methods have some drawbacks. Firstly, detection is restricted by image quality. Secondly, they can not deal with occlusion and perspective projection problems, In this research, a new method which possibly deals with occlusion and perspective problems will be proposed. It extracts spatial information such as the position, the relationship of vehicles in 3-dimensional space, as well as vehicle detection in the image. The main algorithm used in this research is based on an extension of the Hough Transform. The Hough Transform which is proposed to estimates parameters of vertices and directed edges analytically on the Hough Space, is a valuable method for the 3-dimensional analysis of static scenes, motion detection and the estimation of viewing parameters.

차량대수, 속도, 대기행렬길이, 정체 및 유고 검지 등의 실시간 교통 정보를 얻기 위하여 현재 영상처리기술(Image Processing technique)이 기존의 루프검지기(Loop Detector)가 갖는 여러 단점들을 보완하는 효과적인 대체검지기로 널리 인식되고있다. 그러나 현재 사용되는 대부분의 영상검지기는 아주 작은 영역에서의 흑백 강도값(gray level)를 사용하며, 따라서 검지기의 정확도는 취급하는 영상의 화질에 크게 영향을 밭을 뿐만 아니라 3차원 실제 공간이 2차원 영상평면으로 표출되므로 생기는 투시사영(perspective projection)문제에 효과적으로 대처할 수 없다. 이런 문제 때문에 현재 영상검지기는 가능한 한 높게 그리고 도로면에 수직으로 카메라를 설치하여 가능한 한 평면화 된 2차원의 영상을 얻어 처리한다. 그러나 이는 한대의 검지기가 포함할 수 있는 영역이 매우 작을 뿐만 아니라, 가능한 한 카메라를 높게 설치해야 하므로 현실적으로 많은 어려움이 따른다. 본 연구에서는 카메라의 설치 위치 또는 각도에 따라 인식율의 정확도에 큰 차이를 보이는 기존의 알고리즘에서 탈피하여, 낮은 위치에 설치할 때 나타나는 투시사영 문제 및 물체 영상의 일부가 가려져 다음 물체의 인식이 곤란한 문제 등을 해결하기 위한 새로운 방법을 제시하였다. 본 연구에서 제시된 방법은 차량의 검지뿐만 아니라 차량의 위치, 3차원 공간에서의 차량의 관계 등에 관한 정보를 얻을 수 있으며, 사용된 알고리즘은 3차원 공간에서의 물체 인식에 우수한 확장된 Hough Transform에 기초하고 있다.

Keywords

References

  1. A survey of the Hough Transform Computer Vision, Graphics, and Image Processing v.44 Illingworth, J.;Kittler, J.
  2. The third International Conference on Image Processing and its Applications no.307 Vehicle Detection in Open-world Scenes using a Hough Transform Technique IEE Radford, C. J.
  3. Communications of the ACM v.18 Finding Circles by an Array of Accumulators Kimme, C.;Ballard, D.;Sklansky, J.
  4. US Patent, No.3,069,654 Methid and Means for Recognizing Complex Patterns Hough, P. V. C.;Arbor, A.
  5. Pattern Recognition v.13 Generalizing the Hough Transform to Detect Arbitrary Shapes Ballard, D. H.
  6. Pattern Recognition v.21 no.2 Polyhedral Object Recognition using Hough-space Features Engelbrecht, J. R.;Wahl, F. M.
  7. IEE Proceedings v.136 no.6 Hough Technique for 3D Object Recognition Tsu, H. T.;Chan, C. K.
  8. IEEE Transaction PAMI v.10 no.3 Three-Dimensional Vision Structure for Robot Application Nitzan, D.
  9. Communications of the ACM v.15 Use of the Hough Transformation to Detect Lines and Curves in Pictures Duda, R. O.;Hart, P. E.
  10. project de fin detude, Ecole Nationale Superieure de IElectronique dt de ses Applications Transformee de Hough et Algorithme detiquetage sur lordinateur parallele OPENVISION Franc, D. L.
  11. International on Pattern Recognition, No.1988 Analyzing Hough Nets for recognition of Polyhedral-like objects Wahl, F. M.
  12. Fundamentals in Computer Vision Faugeras, O. D.(Ed.)
  13. Pattern Recognition v.26 no.9 A review on image segmentation techniques Pal, N. R.;Pal, S. K.
  14. Reading Digital Image Processing Gonzalez, R. C.;Woods, R. E.
  15. Optimal filter for edge detection. KHOROS on-line user manual Castan, S.
  16. IEEE Transactions on PAMI v.8 no.6 A Computational approach to edge detection Canny, J. F.
  17. Compute vision, graphics, and image processing v.52 Hough transform and signal detection theory performance for image with additive noise Hunt, D. J.;Nolte, L. W.;Reibman, A. R.
  18. Ph.D. Thesis, University of Newcastle Acquiring parking information by image processing and neural networks Kim, D.
  19. Traffic Engineering and Control v.25 no.1 Image Processing Applied to Traffic 1. A General Review Dickinson, K. W.;Waterfall, R. C.
  20. Traffic Engineering and Control v.25 no.2 Image Processing Applied to Traffic 2. Practical Experience Dickinson, K. W.;Waterfall, R. C.
  21. 2nd IEE International Conference on Road Traffic Control Applications of video image processing for traffic control systems Ashworth, R.;Darkin, C. G.;Dickinson, K. W.; Hartley, M. G.;Wan, C. L.;Waterfall, R. C.
  22. PTRC Summer Annual Meeting. Seminar D Traffic analysis using low cost image processing Rourke, A.;Bell, M. G. H.
  23. IEE Conference on Road Traffic Data Collection The Australian road research board video based presence detector Dods, J. S.
  24. 2nd IEE International Conference on Road Traffic Monitoring Camera and computer-aided traffic sensor Versavel, J.;Lemaire, F.;van der Stede, D.
  25. IEEE Transactions on Vehicular Technology v.VT-40 no.1 Vehicle detection vedio through image processing: The Autoscope system Michalopoulos, P. G.
  26. Computer Image Processing in Traffic Engineering Hoose, N.
  27. Traffic Engineering and Control v.33 no.3 IWACTS: An Image analysis tool for motorway surveillance Hoose, N.
  28. 2nd IEE International Conference on Road Traffic Monitoring TITAN: A traffic measurement system using image processing techniques Blosseville, J. M.;Krafft, C.;Lenoir, F.;Motyka, V.;Beucher, S.
  29. Traffic Engineering and Control v.32 no.9 Queue detection and congestion monitoring using image processing Rourke, A.;Bell, M. G. H.