Detecting Dissolve Cut for Multidimensional Analysis in an MPEG compressed domain : Using DCT-R of I, P Frames

MPEG의 다차원 분석을 통한 디졸브 구간 검출 : I, P프레임의 DCT-R값을 이용

  • 허정 (고려대학교 산업시스템정보공학과) ;
  • 박상성 (고려대학교 산업시스템정보공학과) ;
  • 장동식 (고려대학교 산업시스템정보공학과)
  • Published : 2003.07.01

Abstract

The paper presents a method to detect dissolve shots of video scene change detections in an MPEG compressed domain. The proposed algorithm uses color-R DCT coefficients of Ⅰ, P-frames for a fast operation and accurate detection and a minimum decoding process in MPEG sequences. The paper presents a method to detect dissolve shot for three-dimensional visualization and analysis of Image in order to recognize easily in computer as a human detects accurately shots of scene change. First, Color-R DCT coefficients for 8*8 units are obtained and the features are summed in a row. Second, Four-step analysis are Performed for differences of the sum in the frame sequences. The experimental results showed that the algorithm has better detection performance, such as precision and recall rate, than the existing method using an average for all DC image by performing four step analysis. The algorithm has the advantage of speed, simplicity and accuracy. In addition. it requires less amount of storage.

본 논문에서는 비디오 장면전환 효과 중 디졸브(dissolve)에 의한 점진적인 장면전환 구간을 검출하는 알고리즘을 제안한다. 제안한 알고리즘은 처리의 효율성과 MPEG Sequence의 최소한의 복원과정을 위해 Ⅰ, P 프레임의 Color-R값에 대한 DCT계수를 사용하였다. 인간의 시각으로는 비디오의 장면전환점을 쉽게 구분해 낼수 있듯이 컴퓨터가 인식하기 쉽도록 영상을 3차원으로 시각화하고 분석하여 장면전환 구간을 검출하였다. 우선 각각의 영상에서 Color-R에 대한 DCT계수를 추출하고 블록단위인 8*8단위 열의 합을 구해 다시 프레임에 대한 행을 4단계로 분할하여 특징치를 분석하고 4단계의 샷 특징치를 통합하여 샷을 검출한다. 실험결과 제안한 방법이 영상의 단일 특징치를 사용한 방법보다 4단계의 특징치 분석을 사용함으로서 더 좋은 성능을 나타내었다 또한 Ⅰ, P 프레임의 Color-R값의 부분적 복원과정으로 계산시간을 절약할 수 있었다.

Keywords

References

  1. VCCL Lab.,Dep.of Elec.Eng.,Univ of Ottawa, Paper submitted to Image and Vision Computing Journal Image and video indexing in Compressed Domain : A critical review M.K.Mandal;F.IDRIS;S.Panchanathan
  2. IEEE International Conference on Image Processing v.1 Efficient matching and clustering of video shots M.M.Yeung;B.Liu
  3. Proc. IFIP TC2 / WG2.6 Second Working Conf. on Visual Database System Automatic video indexing and full-motion search for object appearances A.Nagasaka;Y.Tanaka
  4. Visual Communication and Image Processing v.SPIE-1606 Video browsing using brightness data K.Otsuji;Y.Tonomura;Y.Ohba
  5. Digital Video Compression: Algorithms and Technologies v.SPIE-2419 Scene Change Detection in a MPEG Compressed Video Sequence J.Meng;Y.Juan;S.F.Chang
  6. IEEE Trans. on Multimedia v.2 no.4 Fast Scene Change Detection Using Direct Feature Extraction from MPEG Compressed Videos S.W.Lee;Y.M.Kim;S.W.Choi
  7. Multimedia Tools and Application v.1 no.1 Video parsing and browsing using compressed data H.J.Zhang;C.Y.Low;S.W.Smoliar
  8. SPIE Proceeding: Digital Video Compression: Algorithms and Technol v.2419 Scene decomposition of MPEG compressed video H.C.Liu;G.L.Zick
  9. ICIP 95 A Fast Algorithm for Video Parsing Using Compressed Sequences K.Shen;E.J.Delp
  10. IEEE Tras. on Circuit and System for Video Technology v.5 no.6 Rapid Scene Analysis on Compressed Video B.-L.Yeo;B.Liu
  11. Proc. of IEEE Workshop on Multimedia Database Management System Efficient Shot Change Detection on Compressed Video Data Tony C.T.Kuo;Y.B lin;Arbee L.P.Chen;Shu-Chin Chen;C.Y.Ni
  12. Storage and Retrieval for Image and Video Database Ⅲ v.SPIE-2420 A Statistical Approach to Scene Change Detection I.K.Sethi;N.Patel
  13. ISCAS Detecting and Compression Dissove Regions in Videio Sequences with DVI Multimedia Image Compression Algorithm A.M.Alattar
  14. Proceedings of ACM Multimedia 93 Image Processinf on Compressed Data for Large Video Databases F.Arman.A.Hsu;M.Y.Chui