The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

An EM Algorithm-Based Approach for Imputation of Pixel Values in Color Image

색조영상에서 랜덤결측화소값 대체를 위한 EM 알고리즘 기반 기법

  • Kim, Seung-Gu (Department of Computer & Data Information, Sangji University)
  • 김승구 (상지대학교 컴퓨터데이터정보학과)
  • Received : 20100200
  • Accepted : 20100300
  • Published : 2010.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, a frequentistic approach to impute the values of R, G, B-components in random missing pixels of color image is provided. Under assumption that the given image is a realization of Gaussian Markov random field, its model is designed such that each neighbor pixel values for a given pixel follows (independently) the normal distribution with covariance matrix scaled by an evaluates of the similarity between two pixel values, so that the imputation is not to be affected by the neighbors with different color. An approximate EM-based algorithm maximizing the underlying likelihood is implemented to estimate the parameters and to impute the missing pixel values. Some experiments are presented to show its effectiveness through performance comparison with a popular interpolation method.

본 논문에서는 색조영상의 R-, G-, B-성분에서 랜덤결측된 화소값들의 대체를 위한 프리퀀티스틱(frequentictic) 기법을 제공한다. 이 기법은 관측영상을 가우시안 마코프 랜덤필드 상의 실현치로서 가정하고, 주어진 화소 내의 근방 화소들이 에지 강도에 따른 서로 다른 분산을 가지는 정규분포를 따른다고 설계함으로써 에지에서 결측화소 대체값이 이질적 색상에 영향 받지 않도록 한다. 이러한 모형하에서 우도가 최대화하도록 결측화소값들을 근사 EM 알고리즘에 기반 한 방법으로 모수들을 추정하고 결측화소를 대체한다. 제안된 방법의 결과들은 보간법에 기초한 대체법과 비교하여 그 유효성을 보인다.

Keywords

Acknowledgement

Supported by : 상지대학교

References

  1. 김승구 (2009). 마코프 랜덤 필드 하에서 정규혼합모형에 의한 다중 결측값 대체기법: 색조영상 결측 화소값 대체에 응용, <한국통계학회논문집>, 16, 914-925. https://doi.org/10.5351/CKSS.2009.16.6.925
  2. Besag, J. (1975). Statistical analysis of non-lattice data, The Statistician, 24, 179-195. https://doi.org/10.2307/2987782
  3. Blanchet, J. and Vignes, M. (2009). A model-based approach to gene clustering with missing observation reconstruction in a Markov random field framework, Journal of Computational Biology, 16, 475-486. https://doi.org/10.1089/cmb.2008.0078
  4. Dass, S. C. and Nair, V. N. (2003). Edge detection, spatial smoothing, and image reconstruction with partially observed multivariate data, Journal of the American Statistical Association, 98, 77-89. https://doi.org/10.1198/01621450338861911
  5. Dempster, A. P., Laird, N. M. and Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm (with discussion), Journal of Royal Statistical Society B, 39, 1-38.
  6. Kim, D., Lee, Y. and Oh, H. S. (2006). Hierarchical likelihood-based wavelet method for denoising signals with missing data, IEEE Signal Processing Letters, 13, 361-364. https://doi.org/10.1109/LSP.2006.871713
  7. Ogawa, T., Haseyama, M. and Kitajima, H. (2006). Restoration of missing intensity of still images by using optical flows, System and Computers in Japan, 37, 1786-1795. https://doi.org/10.1002/scj.20376
  8. Qian, W. and Titterington, D. M. (1992). Stochastic relaxations and EM algorithms for Markov random fields, Journal of Statistical Computation and Simulation, 40, 55-69. https://doi.org/10.1080/00949659208811365
  9. Wei, G. C. G. and Tanner, M. A. (1990). A Monte Calro implementation of the EM algorithm and the poor man's data augmentation algorithms, Journal of the American Statistical Association, 85, 699-704. https://doi.org/10.2307/2290005