Share Email Print

Proceedings Paper

Interframe robust image processing for restoration of heavily corrupted old movies
Author(s): Takahiro Saito; Takashi Komatsu; Toshiaki Ohuchi; Takamasa Seto
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

The old film restoration involves the development of image processing technology. We focus on interframe image processing algorithms for the correction of film misalignment, flicker correction and the removal of blotches. In the digital restoration, the film is first read with a scanner, but the scanned sequence often suffers from irregular spatial vibration due to inaccurate frame alignment. Hence, we develop a robust correction method for estimating interframe misalignment separately from camera work and compensating for it. After the correction of interframe misalignment, we perform flicker correction. Flickers are defined as undesirable brightness fluctuations. We present a hierarchy of flicker-correction models for correcting old film flickers, and develop a flicker correction method which first estimates the correction-model parameters from the input sequence and then corrects its flickers according to the estimated model. Furthermore, we present a method for blotch removal. Blotch distortions are repaired with a blending-type filter. For blotch detection we employ our previously presented approach of the spatiotemporal continuity analysis. The simulations on really corrupted sequences have demonstrated that our interframe image processing techniques can reduce film misalignment and can remove flickers and blotches almost perfectly.

Paper Details

Date Published: 30 May 2000
PDF: 12 pages
Proc. SPIE 4067, Visual Communications and Image Processing 2000, (30 May 2000); doi: 10.1117/12.386626
Show Author Affiliations
Takahiro Saito, Kanagawa Univ. (Japan)
Takashi Komatsu, Kanagawa Univ. (Japan)
Toshiaki Ohuchi, Kanagawa Univ. (Japan)
Takamasa Seto, Kanagawa Univ. (Japan)

Published in SPIE Proceedings Vol. 4067:
Visual Communications and Image Processing 2000
King N. Ngan; Thomas Sikora; Ming-Ting Sun, Editor(s)

© SPIE. Terms of Use
Back to Top