Fusion of complementary detectors for improving blotch detection in digitized films

Authors:
Sorin Tilie;Isabelle Bloch;Louis Laborelli
Affiliations:
Département Recherche et Expérimentation, Institut National de l'Audiovisuel, 4 av. de l'Europe, 94366 Bry sur Marne, France and GET-ENST, TSI CNRS UMR 5141, 46 rue Barrault, 75013 Paris ...;GET-ENST, TSI CNRS UMR 5141, 46 rue Barrault, 75013 Paris, France;Département Recherche et Expérimentation, Institut National de l'Audiovisuel, 4 av. de l'Europe, 94366 Bry sur Marne, France
Venue:
Pattern Recognition Letters
Year:
2007

Citing 10
Cited 1

A New Class of Detail-Preserving Filters for Image Processing

IEEE Transactions on Pattern Analysis and Machine Intelligence
On the Dempster-Shafer framework and new combination rules

Information Sciences: an International Journal
The Combination of Evidence in the Transferable Belief Model

IEEE Transactions on Pattern Analysis and Machine Intelligence
Motion Picture Restoration: Digital Algorithms for Artefact Suppression in Degraded Motion Picture Film and Video

Motion Picture Restoration: Digital Algorithms for Artefact Suppression in Degraded Motion Picture Film and Video
Convergence Properties of the Nelder--Mead Simplex Method in Low Dimensions

SIAM Journal on Optimization
Deterioration detection for digital film restoration

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Efficient detection of temporally impulsive dirt impairments in archived films

Signal Processing
LUM filters: a class of rank-order-based filters for smoothing andsharpening

IEEE Transactions on Signal Processing
Multistage order statistic filters for image sequence processing

IEEE Transactions on Signal Processing
Genetic algorithm optimization of multidimensional grayscale soft morphological filters with applications in film archive restoration

IEEE Transactions on Circuits and Systems for Video Technology

A fusion-based approach to digital movie restoration

Pattern Recognition

Quantified Score

Hi-index	0.10

Visualization

Abstract

This paper proposes a novel method based on Dempster-Shafer (belief function) theory for the detection of blotches in digitized archive film sequences. The detection scheme relies on the fusion of two uncorrelated blotch detectors, one working in the spatial domain and the other one in the temporal domain. The imprecision and uncertainty of both detectors have been modeled using belief function theory, and their combination improves the decision, by taking into account the ignorance and the conflict between detectors. Quantitative evaluation using real blotches ground truth shows that this combination scheme improves the global performance, and compares favorably with two classical blotch detectors.