An automatic video scratch removal based on Thiele type continued fraction
- 201 Downloads
Old age, repeat play and improper preservation always deteriorate the film, and dust and mechanical operations produce artifacts like scratches and blotches. Many researches carried out to repair the damaged digital videos and video inpainting gradually becomes an important topic in digital image process ing. Challenges in scratched video inpainting are automatic detection of scratches and restoration of damaged part. This paper presents an automatic scratch detec tion method as well as a novel scratch removal approach. Stationary wavelet transform (SWT) which shows excellent performance in keeping translation-invariant is introduced to automatically detect the scratches, this strategy makes the scratches’ detection more accurate. At the heart of our method is a new nonlinear interpolation method based on continued fraction in which Thiele-type continued fraction is used to interpolate surrounding known pixels for repairing the damaged part. Algorithm presented in this paper also utilizes both spatial and temporal information of the scratched video during the restoration stage. Experimental results show that our scheme not only obtains more accurate detection of scratches, but also gives better video quality.
KeywordsVideo inpainting Scratch detection Continued fraction Non-linear interpolation
Partial differential equation
Over-complete wavelet expansion
Stationary wavelet transform
Global bi-directional motion compensation frame interpolation
The authors would like to thank Prelinger Archives (http://www.archive.org) for archive film material. We sincerely appreciate the financial support of National Natural Science Foundation of China and we are also grateful to editor and reviewers for their constructive comments and suggestions.
- 2.Bertalmio M, Sapiro G, Ballester C, Ballester C (2000) Image inpainting. Proc. SIGGRAPH’00 pp 417–424. doi: 10.1145/344779.344972
- 6.Dubey N, Agrawal V, Mohapatra S (2009) A generalized wavelet expansion-based algorithm for line scratches detection in old colored or grey videos and static images. http://www.cse.iitk.ac.in/users/ndubey/Documents/introduction.pdf. Accessed Nov 2009
- 8.Güllü MK, Urhan O, Ertürk S (2006) Scratch detection via temporal coherency analysis and removal using edge priority based interpolation. Proc. ISCAS 2006 pp 92–96. doi: 10.1109/ISCAS.2006.1693652
- 11.Joyeux L, Buisson O, Besserer B, Boukir S (1999) Detection and removal of line scratches in motion picture films. Proc. CVPR’99 pp 548–553. doi: 10.1109/CVPR.1999.786991
- 14.Nie SL, Zhang HY, Zhang LP, Fan Y, Brost V (2010) Vertical scratches detection based on edge detection for old film. Proc. IIS 2010 pp 257–260. doi: 10.1109/INDUSI S.2010.5565861
- 20.Tegolo D, Isgro F (2001) A genetic algorithm for scratch removal in static images. Proc. ICIAP 2001 pp 507–511. doi: 10.1109/ICIAP.2001.957060
- 21.Vijaykumar V R, Jothibasu P (2010) Decision based adaptive median filter to remove blotches, scratches, streaks,stripes and impluse noise in images. Proc. ICIP 2010 pp 117–120. doi: 10.1109/ICIP.2010.5651915