대한전자공학회논문지SP (Journal of the Institute of Electronics Engineers of Korea SP)

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

곡률을 이용한 3차원 영상의 에지 기반 표면 분할 알고리즘

Edge-based Surface Segmentation Algorithm of 3-D Image using Curvature

  • 설성욱 (부산대학교 전자공학과) ;
  • 이재출 (부산대학교 전자공학과) ;
  • 남기곤 (부산대학교 전자공학과) ;
  • 전계록 (부산대학교 의공학과) ;
  • 주재흠 (부산카톨릭대학교 정보공학부)
  • 발행 : 2001.03.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 곡률을 이용한 3차원 영상의 에지 기반 표면 분할 알고리즘을 제안한다. 제안한 방법은 표면 분할의 중요한 특징 요소인 에지 검출의 과정에서 3차원 영상의 깊이 정보를 스캔라인별로 2차 곡선으로 근사화하고, 각 곡선의 구분점에 대하여 곡률 계산을 적용하여 에지를 결정한 후, 영역을 그룹화 한다. 기존의 알고리즘은 jump 에지와 crease 에지로 구분하는 다른 처리 과정으로 에지를 검출하고, 특히 crease 에지에 대해서 면 방향의 불연속정도를 결정하는데 어려움이 많았다. 제안한 알고리즘은 기하학적인 접근방법으로 근사화된 곡률 계산을 이용한 단일 처리 과정의 적용과 곡률의 불연속정도를 결정하는데 보다 용이한 방법을 제시한다. 이러한 효율적인 에지 검출을 기반으로 여러 가지 3차원 영상에 대한 실험을 통하여 제안한 방법의 성능이 기존의 방법보다 우수함을 확인하였다.

In this paper, we suggest an edge-based surface segmentation algorithm of 3D image using curvature. For the first, in this proposed method, we approximate 3D depth data to second order curves by each scan line and decide splitting points of 3D edges by curvature of the approximated curves. And finally make a group as 3D surface with the region of input image by the 3D edges. In the conventional algorithms, there are some difficulties in detecting 3D edge with the separated processes for the jump edge and the crease edge and especially, in deciding the ambiguous discontinuity of surface directions about the crease edge. The proposed algorithm decides curvature discontinuity using curvature which is simply calculated by a geometrical approximation. Furthermore, the algorithm has a cooperated process to calculate the jump and crease edges. The results of computer simulations with several 3D images show that the proposed method yields better performance as comparing with the conventional methods.

키워드

참고문헌

  1. F. Ade, Grasping Unknown Objets, in Modelling and Planning for Sensor Based Intelligent Robot System, pp. 445-459, world Scientific, Singapore, 1995
  2. P. J. Flynn and A. K. Jain, Three-Dimensional Object Recognition, in Handbook of Pattern Recognition and Image Processing : Computer Vision, pp. 497-541, Academic Press, San Diego, 1994
  3. P. Durisch, 'Photogrammetry and Computer Graphics for Visual Impact Analysis in Architecture,' Proceedings of ISPRS Conference 1992, pp. 434-445. 1992
  4. L. Blonde, 'The MONA LISA Project: General Presentation,' Proceedings on the European Workshop in Combined Real and Synthetic Image Processing for Broadcast and Video Production, VAP Media Centre, Hamburg, Germany, Nov. 1994
  5. R. Koch, '3-D Surface Reconstruction from Stereoscopic Image Sequences,' Proceedings of the ICCV Conference 95, Cambridge, MA, USA, Jun. 1995 https://doi.org/10.1109/ICCV.1995.466799
  6. D. Zhao and X. Zhang, 'Range-Data-Based Object Surface Segmentation via Edges and Critical Points,' IEEE Trans. Image Processing, 6(6), pp. 826-830, Jun. 1997 https://doi.org/10.1109/83.585233
  7. M. A. Wani and B. G. Batchelor, 'Edge-Region-Based Segmentation of Range Images,' IEEE Trans. PAMI, 16(3), pp. 314-319, Mar. 1994 https://doi.org/10.1109/34.276131
  8. A. Hoover, 'An Experimental Comparison of Range Image Segmentation Algorithms,' IEEE Trans. PAMI, 18(7) pp. 673-689, Jul. 1996 https://doi.org/10.1109/34.506791
  9. Steven D. Cochran and Gerard Medioni, '3-D Surface Description from Binocular Stereo,' IEEE Trans. PAMI, 14(10), pp. 981-994, Oct. 1992 https://doi.org/10.1109/34.159902
  10. L.H. Chen and W.C. Lin, 'Visual Surface Segmentation from Stereo,' Image and Vision Computing, Vol. 15, pp. 95-106, 1997 https://doi.org/10.1016/S0262-8856(96)01116-X
  11. R. Koch, 'Surface Segmentation and Modeling of 3-D Polygonal Objects from Stereoscopic Image Pairs,' Int'l Conference on Pattern Recognition 96, Vienna, Austria, Aug. 1996
  12. X. Jiang and H. Bunke, 'Edge Detection in Range Images Based on Scan Line Approximation,' Computer Vision and Image Understanding, 73(2), pp. 183-199, Feb. 1999 https://doi.org/10.1006/cviu.1998.0715
  13. E. Turcco and R. B. Fisher, 'Computing Surface-based Representations from Range Images,' Proc. IEEE Int'l Symp. on Intelligent Control, Glasgow, Scotland, pp. 275-280, 1992 https://doi.org/10.1109/ISIC.1992.225103
  14. X. Jiang and H. Bunke, 'Range Image Segmentation : Adaptive grouping of edges into regions,' in Computer Vision - ACCV'98, pp. 299-306, Springer-Verlag, Berlin/New York, 1998 https://doi.org/10.1007/3-540-63931-4_230