Trainable blotch detection on high resolution archive films minimizing the human interaction

Attila Licsár, T. Szirányi, László Czúni

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Film archives are continuously in need of automatic restoration tools to accelerate the correction of film artifacts and to decrease the costs. Blotches are a common type of film degradation and their correction needs a lot of manual interaction in traditional systems due to high false detection rates and the huge amount of data of high resolution images. Blotch detectors need reliable motion estimation to avoid the false detection of uncorrupted regions. In case of erroneous detection, usually an operator has to remove the false alarms manually, which significantly decreases the efficiency of the restoration process. To reduce manual intervention, we developed a two-step false alarm reduction technique including pixel-and object-based methods as post-processing. The proposed pixel-based algorithm compensates motion, decreasing false alarms at low computational cost, while the following object based method further reduces the residual false alarms by machine learning techniques. We introduced a new quality metric for detection methods by measuring the required amount of manual work after the automatic detection. In our novel evaluation technique, the ground truth is collected from digitized archive sequences where defective pixel positions are detected in an interactive process.

Original languageEnglish
Pages (from-to)767-777
Number of pages11
JournalMachine Vision and Applications
Volume21
Issue number5
DOIs
Publication statusPublished - Aug 2010

Fingerprint

Pixels
Restoration
Motion estimation
Image resolution
Learning systems
Costs
Detectors
Degradation
Processing

Keywords

  • Blotch detection
  • Digital film restoration
  • Motion estimation
  • Object classification

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Software
  • Computer Science Applications

Cite this

Trainable blotch detection on high resolution archive films minimizing the human interaction. / Licsár, Attila; Szirányi, T.; Czúni, László.

In: Machine Vision and Applications, Vol. 21, No. 5, 08.2010, p. 767-777.

Research output: Contribution to journalArticle

@article{fe18d022c26b4c6fb6abd3828a51d81a,
title = "Trainable blotch detection on high resolution archive films minimizing the human interaction",
abstract = "Film archives are continuously in need of automatic restoration tools to accelerate the correction of film artifacts and to decrease the costs. Blotches are a common type of film degradation and their correction needs a lot of manual interaction in traditional systems due to high false detection rates and the huge amount of data of high resolution images. Blotch detectors need reliable motion estimation to avoid the false detection of uncorrupted regions. In case of erroneous detection, usually an operator has to remove the false alarms manually, which significantly decreases the efficiency of the restoration process. To reduce manual intervention, we developed a two-step false alarm reduction technique including pixel-and object-based methods as post-processing. The proposed pixel-based algorithm compensates motion, decreasing false alarms at low computational cost, while the following object based method further reduces the residual false alarms by machine learning techniques. We introduced a new quality metric for detection methods by measuring the required amount of manual work after the automatic detection. In our novel evaluation technique, the ground truth is collected from digitized archive sequences where defective pixel positions are detected in an interactive process.",
keywords = "Blotch detection, Digital film restoration, Motion estimation, Object classification",
author = "Attila Lics{\'a}r and T. Szir{\'a}nyi and L{\'a}szl{\'o} Cz{\'u}ni",
year = "2010",
month = "8",
doi = "10.1007/s00138-007-0106-y",
language = "English",
volume = "21",
pages = "767--777",
journal = "Machine Vision and Applications",
issn = "0932-8092",
publisher = "Springer Verlag",
number = "5",

}

TY - JOUR

T1 - Trainable blotch detection on high resolution archive films minimizing the human interaction

AU - Licsár, Attila

AU - Szirányi, T.

AU - Czúni, László

PY - 2010/8

Y1 - 2010/8

N2 - Film archives are continuously in need of automatic restoration tools to accelerate the correction of film artifacts and to decrease the costs. Blotches are a common type of film degradation and their correction needs a lot of manual interaction in traditional systems due to high false detection rates and the huge amount of data of high resolution images. Blotch detectors need reliable motion estimation to avoid the false detection of uncorrupted regions. In case of erroneous detection, usually an operator has to remove the false alarms manually, which significantly decreases the efficiency of the restoration process. To reduce manual intervention, we developed a two-step false alarm reduction technique including pixel-and object-based methods as post-processing. The proposed pixel-based algorithm compensates motion, decreasing false alarms at low computational cost, while the following object based method further reduces the residual false alarms by machine learning techniques. We introduced a new quality metric for detection methods by measuring the required amount of manual work after the automatic detection. In our novel evaluation technique, the ground truth is collected from digitized archive sequences where defective pixel positions are detected in an interactive process.

AB - Film archives are continuously in need of automatic restoration tools to accelerate the correction of film artifacts and to decrease the costs. Blotches are a common type of film degradation and their correction needs a lot of manual interaction in traditional systems due to high false detection rates and the huge amount of data of high resolution images. Blotch detectors need reliable motion estimation to avoid the false detection of uncorrupted regions. In case of erroneous detection, usually an operator has to remove the false alarms manually, which significantly decreases the efficiency of the restoration process. To reduce manual intervention, we developed a two-step false alarm reduction technique including pixel-and object-based methods as post-processing. The proposed pixel-based algorithm compensates motion, decreasing false alarms at low computational cost, while the following object based method further reduces the residual false alarms by machine learning techniques. We introduced a new quality metric for detection methods by measuring the required amount of manual work after the automatic detection. In our novel evaluation technique, the ground truth is collected from digitized archive sequences where defective pixel positions are detected in an interactive process.

KW - Blotch detection

KW - Digital film restoration

KW - Motion estimation

KW - Object classification

UR - http://www.scopus.com/inward/record.url?scp=77956616613&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956616613&partnerID=8YFLogxK

U2 - 10.1007/s00138-007-0106-y

DO - 10.1007/s00138-007-0106-y

M3 - Article

VL - 21

SP - 767

EP - 777

JO - Machine Vision and Applications

JF - Machine Vision and Applications

SN - 0932-8092

IS - 5

ER -