Journal of the Institute of Electronics Engineers of Korea SC (전자공학회논문지SC)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Vision-based Obstacle Detection using Geometric Analysis

기하학적 해석을 이용한 비전 기반의 장애물 검출

  • Lee Jong-Shill (Dept. of Biomedical Engineering, Hanyang University) ;
  • Lee Eung-Hyuk (Dept. of Electronics Engineering, Korea Polytechnic University) ;
  • Kim In-Young (Dept. of Biomedical Engineering, Hanyang University) ;
  • Kim Sun-I. (Dept. of Biomedical Engineering, Hanyang University)
  • Published : 2006.05.01
Download PDF
⟨ Previous Next ⟩

Abstract

Obstacle detection is an important task for many mobile robot applications. The methods using stereo vision and optical flow are computationally expensive. Therefore, this paper presents a vision-based obstacle detection method using only two view images. The method uses a single passive camera and odometry, performs in real-time. The proposed method is an obstacle detection method using 3D reconstruction from taro views. Processing begins with feature extraction for each input image using Dr. Lowe's SIFT(Scale Invariant Feature Transform) and establish the correspondence of features across input images. Using extrinsic camera rotation and translation matrix which is provided by odometry, we could calculate the 3D position of these corresponding points by triangulation. The results of triangulation are partial 3D reconstruction for obstacles. The proposed method has been tested successfully on an indoor mobile robot and is able to detect obstacles at 75msec.

이동 로봇의 많은 응용분야에서 장애물을 검출하는 것은 중요한 요소이다. 스테레오 비전과 광류를 이용하여 장애물을 검출하는 방법은 복잡한 연산을 요구하므로 본 논문에서는 단지 두 장면의 영상만을 이용하여 비전 기반 장애물 검출 방법을 제시하고 단일 카메라와 주행거리계를 사용하여 실시간 처리가 가능하도록 하였다. 제안한 방법은 두 장면으로부터 3차원 복원을 수행함으로서 장애물을 검출하는 방법으로 먼저 두 장면의 입력영상 각각에 대하여 Lowe의 SIFT를 사용하여 특징점을 추출하고 이들 간의 대응점을 구한다. 그리고 주행거리계로부터 주어지는 회전과 병진행렬 값들과 삼각법을 이용하여 대응점들에 대한 3차원 위치를 구한다. 이렇게 삼각법에 의해 얻어진 결과는 장애물들에 대한 부분적인 3차원 복원을 의미한다. 제안한 방법은 실내에서 주행하는 이동 로봇에 적용하였을 때 좋은 결과를 얻을 수 있었으며, 75msec의 속도로 장애물을 검출할 수 있었다.

Keywords

References

  1. Crowley JL., 'World Modelling and Position Estimation for a Mobile Robot using Ultrasonic Ranging,' Proc. IEEE Int. Conf. on Robotics and Automation, 1989, pp.674-680 https://doi.org/10.1109/ROBOT.1989.100062
  2. Everett H.R., Sensors for Mobile Robots : Theory and Applications, Wellesley MA, A.K. Peters, 1995
  3. Araki K. et al., 'High speed and Continous Rangefinder System,' Trans IECE of Japan, Vol.E74, No.10, pp.3400-3406, 1991
  4. Ricotti M., Rarili A., Ceresa M., 'Obstacle Avoidance using a line laser,' Proc. of SPIE, Vol.2058 Mobile Robots VIII, 1993, pp.164-177 https://doi.org/10.1117/12.167492
  5. Fujii Y, Wehe K, Weymouth T.E., 'Robust Monocular Depth Perception using Pairs and Approximate Motion,' Proc. IEEE Int. Conf. on Robotics and Automation, 1992, pp.33-39
  6. Abdel-Mottaled M., Chellappa R., Rosenfeld A., 'Binocular Motion Stereo using MAP,' Proc. IEEE Int. Conf. on Computer Vision and Pattern Recognition, 1993, pp.321-327 https://doi.org/10.1109/CVPR.1993.340962
  7. Matthies L., Elfes A., 'Integration of Sonar and Stereo Range Data using a Grid-Based Representation,' Proc. IEEE Int. Conf. on Robotics and Automation, 1988, pp.727-733 https://doi.org/10.1109/ROBOT.1988.12145
  8. Huber J, Graefe V., 'Motion Stereo for Mobile Robots,' IEEE Trans. on Industrial Electronics, Vol.41, No.4, 1994, pp.378-383 https://doi.org/10.1109/41.303787
  9. Wang H., Bowman C., Harris C., 'A Parallel Implementation of a Structure-from-motion Algorithm,'Proc. Second European Conference on Computer Vision, 1992, pp.272-276
  10. Tan T.N., Sullivan G.D., Baker K.D., 'Structure from Motion using the Ground Plane Constraint,' Proc. Second European Conference on Computer Vision, 1992, pp.277-281
  11. Ho P.K., Chung R., 'Stereo-Motion that Complements Stereo and Motion Analysis,' Proc. IEEE Int. Conf. on Computer Vision and Pattern Recognition, 1997, pp.213-218 https://doi.org/10.1109/CVPR.1997.609322
  12. S. Carlsson and J.-O. Eklundh, 'Object detection using model-based prediction and motion parallax,' Proc. of ECCV, pp.297？306, 1990
  13. W. Enkelmann, 'Obstacle detection by evaluation of optical flow fields from image sequences,' Proc. of ECCV, pp.134？138, 1990
  14. P. Fornland, 'Direct obstacle detection and motion from spatio-temporal derivatives,' Proc. of Int. Conf. on Computer Analysis of Images and Patterns, pp.874？879, 1995
  15. T. C. H. Heng, Y. Kuno, and Y. Shirai, 'Active sensor fusion for collision avoidance,' Proc. of IROS, vol.3, pp.1244？1249, 1997 https://doi.org/10.1109/IROS.1997.656387
  16. M. I. A. Lourakis and S. C. Orphanoudakis, 'Visual detection of obstacles assuming locally planar ground,' Proc. of 3rd ACCV, vol.2, pp.527？534, 1998
  17. N. Pears and B. Liang, 'Ground plane segmentation for mobile robot visual navigation,' Proc. of IROS, 2001 https://doi.org/10.1109/IROS.2001.977194
  18. D. Sinclair and A. Blake, 'Quantitative planar region detection,' IJCV, vol. 18, No.1, pp.77？91, 1996 https://doi.org/10.1007/BF00126141
  19. M. Bertozzi and A. Broggi, 'Gold: A parallel real-time stereo vision system for generic obstacle and lane detection,' IEEE Trans. Image Processing, Vol.7, No.1, pp.62？81, 1998 https://doi.org/10.1109/83.650851
  20. K. Hanawa and Y. Sogawa, 'Development of stereo image recognition system for ADA,' Proc. IEEE Intelligent Vehicle Symposium, pp.177？182, 2001
  21. H. Hattori and A. Maki, 'Stereo without depth search and metric calibration,' Proc. CVPR, vol.1, pp.177？184, 2000 https://doi.org/10.1109/CVPR.2000.855817
  22. T. Kato and Y. Ninomiya, 'An approach to vehicle recognition using supervised learning,' Proc. of Machine Vision Applications, pp.77？80, 1998
  23. H. Takizawa and T. Ito, 'Method of preceding vehicles recognition using self organization map,' Technical Report of IEICE, pp.23？28, 2001
  24. O. Faugeras, Q.T. Luong, T. Papadopoulo, The geometry of multiple images, MIT press, 2001
  25. R. Hartley and A. Zisserman, Multiple View Geometry in Computer Vision, Cambridge University Press, 2000
  26. Trucco and Verri, Introductory Techniques for 3-D Computer Vision, Prentice Hall, 1998
  27. Zezhi Chen, Nick Pears, et. al, 'Epipole Estimation under Pure Camera Translation', Pro. 7th Digital Image Computing, pp.849-858, 2003
  28. D. G. Lowe, 'Distinctive image features from scale-invariant keypoints,' Int. Journal of Computer Vision, 60(2), pp.91-110, 2004 https://doi.org/10.1023/B:VISI.0000029664.99615.94
  29. 조강현, 유범재 공역, 3차원 비젼, 대영사, 2000
  30. H. Duane and L. Bruce, Mastering MATLAB 6 : a comprehensive tutorial and reference, Prentice Hall, 2001