Automatic Film Restoration Using Distributed Genetic Algorithm

분산 유전자 알고리즘을 이용한 자동 필름 복원

  • Kim, Byung-Geun (Dept. of advanced technology fusion, school of Internet and multimedia Eng. Konkuk Univ) ;
  • Kim, Kyung-Tai (Dept. of advanced technology fusion, school of Internet and multimedia Eng. Konkuk Univ) ;
  • Kim, Eun-Yi (Dept. of advanced technology fusion, school of Internet and multimedia Eng. Konkuk Univ)
  • 김병근 (건국대학교 신기술융합학과) ;
  • 김경태 (건국대학교 신기술융합학과) ;
  • 김은이 (건국대학교 신기술융합학과)
  • Published : 2009.03.25

Abstract

In recent years, a film restoration has gained increasing attention by many researchers, to support multimedia service of high quality. In general, an old film is degraded by dust, scratch, flick, and so on. Among these, the common factors are scratch and blotch, so that many researchers have been investigated to restore these degradations. However, the methods in literature have one major limitation: A method is working well in dealing with scratches, however it is poorly working in processing the blotches. The goal of this work is to develop a robust technique to restore images degraded by both scratches and blotches. For this, we use MRF-MAP (Markov random field - maximum a posteriori) framework, so that the restoration problem is considered as the minimization problem of the posteriori energy function. As the minimization is one of complex combinatorial problem, we use distributed genetic algorithms (DGAs) that effectively deal with combinatorial problems. To asses the validity of the proposed method, it was tested on natural old films and artificially degraded films, and the results were compared with other methods. Then, the results show that the proposed method is superior to other methods.

최근 고화질의 멀티미디어 서비스에 대한 수요가 증파됨에 따라 필름 복원은 많은 연구자들에 대한 관심이 증가하고 있다. 일반적으로 오래된 필름은 dust, 스크래치, flick 등에 의해 손상된다. 이들 중 대부분에 손상요인들은 스크래치와 블로치이며, 많은 연구자들이 이러한 손상요인을 복원하는 연구를 하고 있다. 그러나 기존의 방법은 대부분 한계점이 있다: 스크래치에 대해서는 잘하지만, 블로치를 처리하기에는 불안전하다. 본 논문에서는 스크래치와 블로치 모두에 의해 손상된 이미지를 복원하는 강건한 방법을 개발하는 것이다. 이를 위해, 우리는 MRF-MAP 프레임워크를 사용하고, 복원문제는 사후 에너지 함수의 최소화 문제로 간주된다. 최소화는 복잡한 결합 문제에 하나이고, 우리는 효과적인 분배와 결합 문제를 위해 분산 유전자 알고리즘을 사용한다. 제안된 방법의 효율성을 증명하기위해, 오래된 필름과 인위적으로 손상된 필름에 실험하였으며, 그 결과를 다른 방법과 비교하였다. 그 결과는 제안된 방법이 더 우수함을 보여주었다.

Keywords

References

  1. S. Geman et al., "A nonlinear filter for the film restoration and other problems in image processing," Graphical Models and Image Processing, vol. 54, pp. 281 - 289, July 1992 https://doi.org/10.1016/1049-9652(92)90075-9
  2. P. Andrey and P. Tarroux, "Unsupervised Image Segmentation using a Distributed Genetic Algorithm," Pattern Recognition, Vo1.27, pp.659-673, May 1994 https://doi.org/10.1016/0031-3203(94)90045-0
  3. L. Joyeux, S. Boukir and B. Besserer, "Film line scratch removal using Kalman filtering and Bayesian restoration," in Proc. of IEEE WACV2000, pp. 8-13, Palm Springs, USA, Dec 2000 https://doi.org/10.1109/WACV.2000.895396
  4. 김경태, 고은정, 김은이. "공간적인 정보 기반의 디지털 필름 스크래치 복원," 한국컴퓨터종합학술대회 논문집, 제34권, 제1호(c), 454-459, 2007년 3월
  5. P. Schallauer, A. Pinz and W. Hass, "Automatic restoration algorithms for 35mm film," Journal of Computer Vision Research, Vol. 1, no. 3, Summer, 1999
  6. A. C. Kokaram, "Detection and removal of line scratches in degraded motion picture sequences," Signal Processing, Vol. 1. pp. 5-8, September 1996
  7. V. Bruni, D. Vitulano, "A generalized model for scratch detection," IEEE Transactions on Image Processing, Vol. 13, no.1, pp. 44-50, Jan 2004 https://doi.org/10.1109/TIP.2003.817231
  8. L. D'morea, L. Marcellinoa, A. Murli. "Image sequence inpainting: Towards numerical software for detection and removal of local missing data via motion estimation," Journal of Computational and Applied Mathematics, Vol. 198, no. 2, pp. 396–413, January 2007 https://doi.org/10.1016/j.cam.2005.09.030
  9. Yang-Ta Kao, Shih, T.K., Hsing-Ying Zhong, Liang-Kuang Dai, "Scratch Line Removal on Aged Films," International Symposium on Multimedia, pp. 147-151, 2007
  10. Seong-Whan Kim and Ki-Hong Ko. "Efficient Optimization of Inpainting Scheme and Line Scratch Detection for Old Film Restoration," Lecture Notes in Computer Science, Vol 4099, pp. 623-631, 2006 https://doi.org/10.1007/978-3-540-36668-3_66