Detection of dirt impairments from archived film sequences: survey and evaluations

Jinchang Ren, T. Vlachos

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Film dirt is the most commonly encountered artifact in archive restoration applications. Since dirt usually appears as a temporally impulsive event, motion-compensated interframe processing is widely applied for its detection. However, motion-compensated prediction requires a high degree of complexity and can be unreliable when motion estimation fails. Consequently, many techniques using spatial or spatiotemporal filtering without motion were also been proposed as alternatives. A comprehensive survey and evaluation of existing methods is presented, in which both qualitative and quantitative performances are compared in terms of accuracy, robustness, and complexity. After analyzing these algorithms and identifying their limitations, we conclude with guidance in choosing from these algorithms and promising directions for future research.
LanguageEnglish
Article number067005
JournalOptical Engineering
Volume49
Issue number6
DOIs
Publication statusPublished - 1 Jul 2010

Fingerprint

dirt
impairment
evaluation
Motion estimation
Restoration
spatial filtering
Processing
restoration
artifacts
predictions

Keywords

  • image restoration
  • image sequences
  • spatial filters
  • film dirt detection

Cite this

@article{a8544547ee0a4f61ad3bb4a759bc335c,
title = "Detection of dirt impairments from archived film sequences: survey and evaluations",
abstract = "Film dirt is the most commonly encountered artifact in archive restoration applications. Since dirt usually appears as a temporally impulsive event, motion-compensated interframe processing is widely applied for its detection. However, motion-compensated prediction requires a high degree of complexity and can be unreliable when motion estimation fails. Consequently, many techniques using spatial or spatiotemporal filtering without motion were also been proposed as alternatives. A comprehensive survey and evaluation of existing methods is presented, in which both qualitative and quantitative performances are compared in terms of accuracy, robustness, and complexity. After analyzing these algorithms and identifying their limitations, we conclude with guidance in choosing from these algorithms and promising directions for future research.",
keywords = "image restoration, image sequences, spatial filters, film dirt detection",
author = "Jinchang Ren and T. Vlachos",
year = "2010",
month = "7",
day = "1",
doi = "10.1117/1.3456633",
language = "English",
volume = "49",
journal = "Optical Engineering : Journal of the Society of Photo-Optical Instrumentation Engineers",
issn = "0091-3286",
number = "6",

}

Detection of dirt impairments from archived film sequences : survey and evaluations. / Ren, Jinchang; Vlachos, T.

In: Optical Engineering, Vol. 49, No. 6, 067005, 01.07.2010.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Detection of dirt impairments from archived film sequences

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

AU - Ren, Jinchang

AU - Vlachos, T.

PY - 2010/7/1

Y1 - 2010/7/1

N2 - Film dirt is the most commonly encountered artifact in archive restoration applications. Since dirt usually appears as a temporally impulsive event, motion-compensated interframe processing is widely applied for its detection. However, motion-compensated prediction requires a high degree of complexity and can be unreliable when motion estimation fails. Consequently, many techniques using spatial or spatiotemporal filtering without motion were also been proposed as alternatives. A comprehensive survey and evaluation of existing methods is presented, in which both qualitative and quantitative performances are compared in terms of accuracy, robustness, and complexity. After analyzing these algorithms and identifying their limitations, we conclude with guidance in choosing from these algorithms and promising directions for future research.

AB - Film dirt is the most commonly encountered artifact in archive restoration applications. Since dirt usually appears as a temporally impulsive event, motion-compensated interframe processing is widely applied for its detection. However, motion-compensated prediction requires a high degree of complexity and can be unreliable when motion estimation fails. Consequently, many techniques using spatial or spatiotemporal filtering without motion were also been proposed as alternatives. A comprehensive survey and evaluation of existing methods is presented, in which both qualitative and quantitative performances are compared in terms of accuracy, robustness, and complexity. After analyzing these algorithms and identifying their limitations, we conclude with guidance in choosing from these algorithms and promising directions for future research.

KW - image restoration

KW - image sequences

KW - spatial filters

KW - film dirt detection

U2 - 10.1117/1.3456633

DO - 10.1117/1.3456633

M3 - Article

VL - 49

JO - 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 - 6

M1 - 067005

ER -