Open AccessEFFICIENT RESTORATION OF VARIABLE AREA SOUNDTRACKS
Author(s) -
Abdelâali Hassaïne,
Étienne Decencière,
Bernard Besserer
Publication year - 2011
Publication title -
image analysis and stereology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.237
H-Index - 27
eISSN - 1854-5165
pISSN - 1580-3139
DOI - 10.5566/ias.v28.p113-119
Subject(s) - image restoration , computer vision , artificial intelligence , computer science , image (mathematics) , segmentation , domain (mathematical analysis) , variable (mathematics) , field (mathematics) , image segmentation , signal (programming language) , image processing , mathematics , pure mathematics , programming language , mathematical analysis
The restoration of motion picture films using digital image processing has been an active research field for many years. The restoration of the soundtrack however, has mainly been performed in the sound domain, using signal processing methods, in spite of the fact that it is recorded as a continuous image between the images of the film and the perforations. In this paper a restoration method for variable area soundtrack restoration at the image level is presented. First, a novel method is proposed for the detection of the symmetry axis of the scanned soundtrack. Then, a comparison between the watershed and the region growing segmentation of the soundtrack is developed. Another algorithm aiming to enforce the symmetry and to correct the edges of the segmented image is presented. A last step aiming to smooth the edges of the obtained image is performed. Finally, experimental results are reported and possible future improvements are discussed