Object Extraction Technique using Extension Search Algorithm based on Bidirectional Stereo Matching

양방향 스테레오 정합 기반 확장탐색 알고리즘을 이용한 물체추출 기법

  • Choi, Young-Seok (School of Electrical & Computer Engineering, Chungbuk National University) ;
  • Kim, Seung-Geun (Maritime & Ocean Engineering Research Institute, Korea Ocean Research & Development Institute) ;
  • Kang, Hyun-Soo (School of Electrical & Computer Engineering, Chungbuk National University)
  • 최영석 (충북대학교 전기전자컴퓨터공학부) ;
  • 김승근 (한국해양연구원 해양시스템안전연구소) ;
  • 강현수 (충북대학교 전기전자컴퓨터공학부)
  • Published : 2008.03.25

Abstract

In this paper, to extract object regions in stereo image, we propose an enhanced algorithm that extracts objects combining both of brightness information and disparity information. The approach that extracts objects using both has been studied by Ping and Chaohui. In their algorithm, the segmentation for an input image is carried out using the brightness, and integration of segmented regions in consideration of disparity information within the previously segmented regions. In the regions where the brightness values between object regions and background regions are similar, however, the segmented regions probably include both of object regions and background regions. It may cause incorrect object extraction in the merging process executed in the unit of the segmented region. To solve this problem, in proposed method, we adopt the merging process which is performed in pixel unit. In addition, we perform the bi-directional stereo matching process to enhance reliability of the disparity information and supplement the disparity information resulted from a single directional matching process. Further searching for disparity is decided by edge information of the input image. The proposed method gives good performance in the object extraction since we find the disparity information that is not extracted in the traditional methods. Finally, we evaluate our method by experiments for the pictures acquired from a real stereoscopic camera.

본 논문은 스테레오 영상에서 물체영역을 추출하기 위하여 변위정보와 밝기정보를 함께 사용하는 개선된 물체 추출 알고리즘을 제안한다. 변위정보와 밝기정보를 이용하여 물체를 추출하는 기법은 Ping과 Chaohui에 의해 연구된 바 있으며 이들의 기법은 입력영상 밝기기반으로 분할하고 분할된 영역 내의 변위정보를 고려하여 분할영역을 병합한다. 그러나 물체와 배경의 밝기 값이 비슷한 경우, 분할영역은 물체영역과 배경영역을 동시에 포함하게 됨으로써 분할영역 단위의 병합은 물체영역 추출의 오류를 야기한다. 이 문제를 해결하기 위해서 제안된 기법에서는 화소단위의 병합을 수행한다. 또한 변위정보의 신뢰도를 높이고 단방향 정합의 부족한 변위정보를 보완하기 위하여 양방향 스테레오 정합을 구현하며, 변위 탐색의 진행 여부는 입력영상의 기울기 성분, 즉 물체의 경계정보로부터 결정한다. 제안된 물체추출 알고리즘은 기존 방법에서 추출되지 않은 변위정보를 탐색함으로써 우수한 물체추출 성능을 보인다. 최종적으로 스테레오 카메라를 이용하여 획득한 실사영상에 대한 실험으로 제안된 방법의 성능을 평가한다.

Keywords

References

  1. D. Schastein and R. Szeliski, 'A taxonomy and evaluation of dense two-frame stereo correspondence algorithm', International Journal of Computer Vision, Vol. 47, no. 1, pp. 7-42, 2002 https://doi.org/10.1023/A:1014573219977
  2. J. Sun, N. Zheng, and H. Shum, 'Stereo matching using belief propagation', IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 25, pp. 787-800, 2003 https://doi.org/10.1109/TPAMI.2003.1206509
  3. Y. Boykov, O. Veksler, and R. Zabih, 'Fast approximate energy minimization via raph cuts', IEEE Trans. Pattern Analysis and Machine Intelligence, Vol.23, pp. 1222-1239, 2001 https://doi.org/10.1109/34.969114
  4. P. F. Felzenszwalb and D. P. Huttenlocher, 'Efficient belief propagation for early vision', in Proc. Computer Vision and Pattern Recognition, 2004
  5. L. Zitnick and T. Kanade, 'A cooperative algorithm for stereo matching and occlusion detection', IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, no. 7, pp. 675-154, 2000 https://doi.org/10.1109/34.865184
  6. P. An, C. Lu and Z. Zhang, 'Object segmentation using stereo images', International Conference of Communications, Circuits and System, Vol. 1, pp. 534-538. 2004
  7. M. F. Tappen and W. T. Freeman, 'Comparison of graph cuts with belief propagation for stereo using identical MRF paramater', in Proc. International Conference on Computer Vision, 2003