연속영상에서 motion 기반의 새로운 분할 알고리즘

A new motion-based segmentation algorithm in image sequences

  • 정철곤 (성균관대학교 전기전자 및 컴퓨터공학부 디지털신호처리연구실) ;
  • 김중규 (성균관대학교 전기전자 및 컴퓨터공학부 디지털신호처리연구실)
  • 발행 : 2002.03.01

초록

본 논문에서는 연속영상에서 움직이는 객체의 motion에 기반하여 영상을 분할하는 새로운 알고리즘을 제안하였다. 전체적인 분할 과정은 2단계로 구성되어진다. 첫 단계는 '픽셀 레이블링' 단계이며, 두 번째 단계는 'motion 분할' 단계이다. '픽셀 레이블링' 단계에서는 optical flow에 의해 발생하는 속도 벡터들의 크기에 따라 영상의 각 픽셀에 레이블을 부여한다. 'Motion 분할' 단계에서는 첫 단계에서 생겨난 불필요한 잡음을 제거하기 위해 motion 필드를 마코프 랜덤 필드로 모델링하여 에너지 최소화를 통해 motion을 분할한다. 실험결과, 제안된 알고리즘이 연속영상에서 움직이는 객체의 motion을 효율적으로 분할함을 확인할 수 있었다.

This paper presents a new motion-based segmentation algorithm of moving objects in image sequences. The procedure toward complete segmentation consists of two steps: pixel labeling and motion segmentation. In the first step, we assign a label to each pixel according to magnitude of velocity vector. And velocity vector is generated by optical flow. And, in the second step, we have modeled motion field as a markov random field for noise canceling and make a segmentation of motion through energy minimization. We have demonstrated the efficiency of the presented method through experimental results.

키워드

참고문헌

  1. IEEE trans. on image processing v.6 Simultaneous motion estimation and segmentation M. M. Chang;A. M. Tekalp;M. I. Sezan
  2. Image and Vision Computing v.19 Moving object segmentation in video sequences by user interaction and automatic object tracking M. Kim;J. G. Jeon;J. S. Kwak;M. H. Lee;C. Ahn
  3. Signal Processing v.76 Spatiotemporal MRF approach to video segmentation: Application to motion detection and lip segmentation F. Luthon;A. Caplier;M. Lievin
  4. Artificial Intelligence v.17 Determining optical flow B. K. P. Horn;B. G. Schunck
  5. Signal detection & estimation M. Barkat
  6. Ad Oculos Ditital image processing H. Bassmann;P.W. Besslich
  7. ICAPRDT'99 Determination of number clusters in K-means clustering and application in colour image segmentation S. Ray;R. H. Turi
  8. Machine vision R. Jain;R. Kasturi;B. G. Schunck
  9. Central European Seminar on Computer Graphics Reconstruction of tomographic data by markov random fields M. Zimanyi
  10. Pattern Recognition Proceedings, 10th International Conference v.1 MRF model-based algorithms for image segmentation R. C. Dubes;A. K. Jain;S. G. Nadabar;C. C. Chen
  11. IEEE trans. on circuits and systems for video technology v.9 An efficient two-pass MAP-MRF algorithm for motion estimation based on mean field theory J. Wei;Z. Li
  12. Korean Institute of Communication Sciences v.26 Automatic segmentation of non-rigid object in image sequences C. Jung;J. K. Kim;C. Ahn
  13. IEEE Trans. Circuits and Systems for Video Technology v.9 A VOP Generation Tool: Automatic Segmentation of Moving Objects in Image Sequences Based on Spatio-Temporal Information M. Kim;J. G. Choi;D. Kim;H. Lee;M. H. Lee;C. Ahn;Y. S. Ho
  14. Korean Institute of Communication Sciences v.26 Automatic segmentation of moving object using automatic region growing algorithm C. Jung;K. S. Kim;J. K. Kim
  15. Signal Processing: Image Communication v.7 Bayesian algorithms for adaptive change detection in image sequences using Markov random fields T. Aach;A. Kaup
  16. 한국통신학회 추계종합학술발표회 논문집 no.12-12 마코프 랜덤 필드를 이용한 움직이는 객체의 자동분할 정철곤;김중규