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

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

Extraction of Corresponding Points Using EMSAC Algorithm

EMSAC 알고리듬을 이용한 대응점 추출에 관한 연구

  • Ye, Soo-Young (Pusan National University School of Medicine) ;
  • Jeon, Ah-Young (Dept. of Interdisciplinary program in Biomedical Engineering, Pusan National University) ;
  • Jeon, Gye-Rok (Dept. of Interdisciplinary program in Biomedical Engineering, Pusan National University) ;
  • Nam, Ki-Gon (Dept. of Electronics Engr., Pusan National University)
  • 예수영 (부산대학교 의학전문대학원 BK21 고급의료양성사업단) ;
  • 전아영 (부산대학교 의공학협동과정) ;
  • 전계록 (부산대학교 의공학협동과정) ;
  • 남기곤 (부산대학교 전자공학과)
  • Published : 2007.07.25
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we proposed the algorithm for the extraction of the corresponding points from images. The proposed algorithm EMSAC is based on RANSAC and EM algorithms. In the RANSAC procedure, the N corresponding points are randomly selected from the observed total corresponding points to estimate the homography matrix, H. This procedure continues on its repetition until the optimum H are estimated within number of repetition maximum. Therefore, it takes much time and does not converge sometimes. To overcome the drawbacks, the EM algorithm was used for the selection of N corresponding points. The EM algorithm extracts the corresponding points with the highest probability density to estimate the optimum H. By the experiments, it is demonstrated that the proposed method has exact and fast performance on extraction of corresponding points by combining RANSAC with EM.

본 논문에서는 영상으로부터 획득된 대응점을 추출하기 위한 새로운 알고리듬을 제안한다. 제안하는 EMSAC 알고리듬은 EM과 RANSAC에 기반을 두고 있다. RANSAC 과정에서는 N개의 대응점들이 랜덤하게 선택되어진다. 랜덤으로 N개의 대응점을 선택하는 과정은 최대 반복횟수 내에서 적절한 파라미터가 추정될 때까지 반복된다. 이는 시간이 오래 걸리고 때로는 적절한 파라미터에 수렴하지 않는 경우도 발생한다. 그러므로 본 연구에서는 RANSAC 알고리듬에서 N개 대응점을 임의로 선택하는 대신 최적의 해가 존재할 확률이 높은 영역에서 대응점을 선택하는 EMSAC 알고리듬을 사용하였다. EMSAC 알고리듬은 반복적인 선택을 줄여 안정적이고 처리 속도가 빠른 대응점들을 추출할 수 있다.

Keywords

References

  1. P. H. S Torr and D.W Murray, 'Statistical detection of independent movement from a moving camera,' Image and Vision Computing, vol. 11, no. 4, pp. 180-187, 1993 https://doi.org/10.1016/0262-8856(93)90034-E
  2. G. Xu and Z. Zhang, Epipolar geometry in stereo, motion and object recognition, Kluwer Academic Pulishers, 1996
  3. M. A. Fischler and R. C. Bolles, 'Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography,' Comm. of the ACM, vol. 24, no. 6, pp. 381-395, Jun. 1981 https://doi.org/10.1145/358669.358692
  4. H. Akaike. 'Information theory and an extension of the maximum likelihood principle,' 2nd International Symposium on Information Theory, Armenia, 1971
  5. E. Malis and R. Cipolla, 'Multi-view constraints between collineations: application to selfcalibration from unknown planar structures,' In European Conference on Computer Vision, June 2000
  6. Perez, M. M.; Dennis, T. J., 'An adaptive implementation of the SUSAN method for image edge and feature detection,' International Conference on Image Processing, vol. 2, pp. 394 - 397, Oct. 1997
  7. G. Borgefors, 'Distance Transformations in Digital Images,' Computer Vision, Graphics, and Image Processing, vol. 34, pp. 344-371, 1986 https://doi.org/10.1016/S0734-189X(86)80047-0
  8. D. W. Paglieroni, 'Distance Transforms: Properties and Machine Vision Applications,' CVGIP: Graphical Models and Image Processing, vol. 54, no. 1, pp. 56-74, Jan. 1992 https://doi.org/10.1016/1049-9652(92)90034-U
  9. Yuille, A., Ferraro, M., Zhang, T., 'Surface shape from warping,' IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 846 - 851, June 1997
  10. DeCarlo, D., Metaxas, D., 'Adaptive shape evolution using blending,' Fifth International Conference on Computer Vision, pp. 834 - 839, June 1995