한국통신학회논문지 (The Journal of Korean Institute of Communications and Information Sciences)

한국통신학회 (The Korean Institute of Commucations and Information Sciences)
권호 표지

복수 특징의 사전 검사에 의한 영상 벡터양자화의 고속 부호화 기법

A Fast Encoding Algorithm for Image Vector Quantization Based on Prior Test of Multiple Features

  • 류철형 (국방과학연구소 1체계-1부-2팀) ;
  • 나성웅 (충남대학교 정보통신공학부 영상통신 연구실)
  • 발행 : 2005.12.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 영상 백터 양자화를 위한 새로운 고속 부호화 기법을 제안하는데, 제안 기법은 다차원의 참조 표로 복수 특징의 부분 거리를 사용한다. 복수 특징을 사용하는 기존 기법은 탐색 순서와 연산 과정을 고려할 때 복수 특징을 단계적으로 처리한다. 반면에 제안 기법은 참조 표를 사용하여 복수 특징들을 동시에 활용한다. 본 논문에서는 가용한 수준의 메모리를 위해 테두리 효과를 고려하는 참조 표의 구성 방법과 참조 표의 부분 거리를 활용하며 현재의 탐색을 중지하는 방법을 상세하게 기술한다. 시뮬레이션 결과는 제안 기법의 효율성을 확인시켜 주는데, 부호책 크기가 256일 때 제안 기법은 OHTPDS 기법이나 $M-L_2NP$ 기법 등과 같이 최근에 제안된 기법들이 요구하는 연산량의 $70\%$ 수준까지 연산량을 감소시킨다. 가용한 수준의 전처리와 메모리를 사용함으로써 제안 기법은 전체탐색 기법과 통일한 화질을 유지하면서 전체 탐색 기법이 요구하는 연산량의 $2.2\%$ 이하로 연산량을 감소시킨다.

This paper presents a new fast encoding algorithm for image vector quantization that incorporates the partial distances of multiple features with a multidimensional look-up table (LUT). Although the methods which were proposed earlier use the multiple features, they handles the multiple features step by step in terms of searching order and calculating process. On the other hand, the proposed algorithm utilizes these features simultaneously with the LUT. This paper completely describes how to build the LUT with considering the boundary effect for feasible memory cost and how to terminate the current search by utilizing partial distances of the LUT Simulation results confirm the effectiveness of the proposed algorithm. When the codebook size is 256, the computational complexity of the proposed algorithm can be reduced by up to the $70\%$ of the operations required by the recently proposed alternatives such as the ordered Hadamard transform partial distance search (OHTPDS), the modified $L_2-norm$ pyramid ($M-L_2NP$), etc. With feasible preprocessing time and memory cost, the proposed algorithm reduces the computational complexity to below the $2.2\%$ of those required for the exhaustive full search (EFS) algorithm while preserving the same encoding quality as that of the EFS algorithm.

키워드

참고문헌

  1. Y. Linde, A. Buzo, R. Gray, 'An Algorithm for Vector Quantizer Design,' IEEE Trans. Commun., 28, pp. 84-95, Jan. 1980 https://doi.org/10.1109/TCOM.1980.1094577
  2. C. Bei, R. Gray, 'An Improvement of the Minimum Distortion Encoding Algorithm for Vector Quantization,' IEEE Trans. Commun., 33, pp. 1132-1133, Oct. 1985 https://doi.org/10.1109/TCOM.1985.1096214
  3. S. Chen, J. Pan, 'Fast Search Algorithm for VQ-based Recognition of Isolated Words,' IEE Proceedings, 136, pp. 391-396, Dec. 1989
  4. S. Choi, S. Chae, 'Incremental-search Fast Vector Quantiser using Triangular Inequalities for Multiple Anchors,' IEE Electron. Lett., 34, pp. 1192-1193, June 1998 https://doi.org/10.1049/el:19980897
  5. S. Ra, J. Kim, 'A Fast Mean-distance- ordered Partial Codebook Search Algorithm for Image Vector Quantization,' IEEE Trans. Circuits and Systems II: Analog and Digital Signal Processing., 40, pp. 576-579, Sept. 1993 https://doi.org/10.1109/82.257335
  6. C. Lee, L. Chen, 'Fast Closest Codeword Search Algorithm for Vector Quantization,' IEE Proc. Vision, Image and Signal Processing, 141, pp. 143-148, June 1994 https://doi.org/10.1049/ip-vis:19941140
  7. S. Baek, B. Jeon, K. Sung, 'A Fast Encoding Algorithm for Vector Quantization,' IEEE Signal Processing Lett., 4, pp. 325-327, Dec. 1997 https://doi.org/10.1109/97.650035
  8. Z. Lu, D. Xu, S. Sun, 'Fast Codeword Search Algorithm for Image Vector Quantization based on Ordered Hadamard Transform,' IECE Trans. Inf. & Syst., E86-D, pp. 1318-1320, July 2003
  9. S. Choi, S. Chae, 'Extended Mean-distance-ordered Search using Multiple L1 and L2 Inequalities for Fast Vector Quantization,' IEEE Trans. C. & Systems II: Analog and Digital Signal Processing, 47, pp. 349-352, Apr. 2000 https://doi.org/10.1109/82.839670
  10. J. Pan, Z. Lu, S. Sun, 'An Efficient Encoding Algorithm for Vector Quantization based on Sub-vector Technique,' IEEE Trans. Image Processing, 12, pp. 265-270, March 2003 https://doi.org/10.1109/TIP.2003.810587
  11. Z. Pan, K. Kotani, T. Ohmi, 'Improved Fast Encoding Method for Vector Quantization based on Subvector Technique,' IEEE International Symposium on Circuits and Systems(ISCAS), 2005, pp. 6332-6335, May 2005
  12. B. Song, J. Ra, 'A Fast Search Algorithm for Vector Quantization using L2-norm Pyramid of Codewords,' IEEE Trans. Image Process., 11, pp. 10-15, Jan. 2002 https://doi.org/10.1109/83.977878
  13. Z. Pan, K. Kotani, T. Ohmi, 'Fast Encoding Method for Vector Quantization using Modified $L_2-Norm$ Pyramid,' IEEE Signal Process. Lett., 12, pp. 609-612, Sept. 2005 https://doi.org/10.1109/LSP.2005.851263
  14. S. Tai, C. Lai, Y. Lin, 'Two Fast Nearest Neighbor Searching Algorithms for Image Vector Quantization,' IEEE Trans. Commun., 44, pp. 1623-1628, Dec. 1996 https://doi.org/10.1109/26.545888
  15. Z. Pan, K. Kotani, T. Ohmi, 'Improved Fast Search Method for Vector Quantization using Discrete Walsh Transform,' International Conference on Image Process. (ICIP), 2004, 5, pp. 3177-3180, Oct. 2004
  16. 최지웅, 나성웅, '동영상의 차분이미지 부호화를 위한 고속 벡터양자화 알고리즘', 한국통신학회 학술발표 논문집, 17(3), pp.467-470, 1998
  17. 정일화, 최인호, 이대영, '벡터양자화의 고속 인코딩과 인덱스 무손실 압축', 한국통신학회 논문지, 27(9A), pp.925-934, 2002
  18. C. Ryu, S. Ra, 'A Fast Full Search Equivalent Encoding Algorithm for Image Vector Quantization based on the Walsh-Hadamard Transform and a Look-up Table,' IEEE Nonlinear Signal and Image Processing, Abstract, pp.23-23, May 2005
  19. C. Ryu, S. Ra, 'A Fast Full Search Equivalent Encoding Algorithm for Image Vector Quantization based on the WHT and a LUT,' The 5th IEEE International Workshop on System-on-chip for Real-Time Applications, 1, pp.405-409, July 2005