Non-motion-compensated region-based dirt detection for film archive restoration

Jinchang Ren, T. Vlachos

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

A novel non-motion-compensated method is proposed for dirt detection in archived film sequences. A confidence measurement extracted from raw differences between current frame and each of the previous and next frames is used to exploit the temporally impulsive nature of dirt impairments. Further evolution of the confidence signal enables the minimization of false alarms and the fine-tuning of detector sensitivity. After morphological filtering and consistency checks candidate regions of dirt emerge, enabling the computation of binary detection masks. Our experiments show that our method compares favorably with extended spike detection index (SDIp), rank order detector (ROD), and lower-upper-middle (LUM) approaches and provides efficient and robust detection performance for a wide range of archived film material.
LanguageEnglish
Pages087004(1-6)
Number of pages7
JournalOptical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers
Volume45
Issue number8
DOIs
Publication statusPublished - 2006

Fingerprint

dirt
restoration
Restoration
Detectors
Masks
confidence
Tuning
detectors
impairment
false alarms
spikes
masks
Experiments
tuning
optimization

Keywords

  • video signal processing
  • image restoration
  • film dirt detection

Cite this

@article{48a4076f5a25489d8c3d9f6377805c63,
title = "Non-motion-compensated region-based dirt detection for film archive restoration",
abstract = "A novel non-motion-compensated method is proposed for dirt detection in archived film sequences. A confidence measurement extracted from raw differences between current frame and each of the previous and next frames is used to exploit the temporally impulsive nature of dirt impairments. Further evolution of the confidence signal enables the minimization of false alarms and the fine-tuning of detector sensitivity. After morphological filtering and consistency checks candidate regions of dirt emerge, enabling the computation of binary detection masks. Our experiments show that our method compares favorably with extended spike detection index (SDIp), rank order detector (ROD), and lower-upper-middle (LUM) approaches and provides efficient and robust detection performance for a wide range of archived film material.",
keywords = "video signal processing, image restoration, film dirt detection",
author = "Jinchang Ren and T. Vlachos",
year = "2006",
doi = "10.1117/1.2338592",
language = "English",
volume = "45",
pages = "087004(1--6)",
journal = "Optical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers",
issn = "0091-3286",
number = "8",

}

TY - JOUR

T1 - Non-motion-compensated region-based dirt detection for film archive restoration

AU - Ren, Jinchang

AU - Vlachos, T.

PY - 2006

Y1 - 2006

N2 - A novel non-motion-compensated method is proposed for dirt detection in archived film sequences. A confidence measurement extracted from raw differences between current frame and each of the previous and next frames is used to exploit the temporally impulsive nature of dirt impairments. Further evolution of the confidence signal enables the minimization of false alarms and the fine-tuning of detector sensitivity. After morphological filtering and consistency checks candidate regions of dirt emerge, enabling the computation of binary detection masks. Our experiments show that our method compares favorably with extended spike detection index (SDIp), rank order detector (ROD), and lower-upper-middle (LUM) approaches and provides efficient and robust detection performance for a wide range of archived film material.

AB - A novel non-motion-compensated method is proposed for dirt detection in archived film sequences. A confidence measurement extracted from raw differences between current frame and each of the previous and next frames is used to exploit the temporally impulsive nature of dirt impairments. Further evolution of the confidence signal enables the minimization of false alarms and the fine-tuning of detector sensitivity. After morphological filtering and consistency checks candidate regions of dirt emerge, enabling the computation of binary detection masks. Our experiments show that our method compares favorably with extended spike detection index (SDIp), rank order detector (ROD), and lower-upper-middle (LUM) approaches and provides efficient and robust detection performance for a wide range of archived film material.

KW - video signal processing

KW - image restoration

KW - film dirt detection

U2 - 10.1117/1.2338592

DO - 10.1117/1.2338592

M3 - Article

VL - 45

SP - 087004(1-6)

JO - Optical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers

T2 - Optical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers

JF - Optical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers

SN - 0091-3286

IS - 8

ER -