Face Image Quality and its Improvement in a Face Detection System

Nasrollahi Kamal; Thomas B. Moeslund

Face Image Quality and its Improvement in a Face Detection System

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research

Abstract

When a person passes by a surveillance camera a sequence of images is obtained. Most of these images are redundant and usually keeping some of them which have better quality is sufficient. So before performing any analysis on the face of a person, the face at the first step needs to be detected. In the second step the quality of the different face images needs to be evaluated. Finally, after choosing the best image(s) based on this quality assessment, in the third step, if this image(s) is not satisfying a predefined set of measures for good quality images, its quality should be improved. In this work we are trying to develop a system to deal with the video sequences in these 3 steps.

Original language	English
Title of host publication	Proceedings of the 16th Danish Conference on Pattern Recognition and Image Analysis
Editors	Søren I. Olsen
Number of pages	3
Publisher	Museum Tusculanum
Publication date	2008
Pages	25-27
Publication status	Published - 2008
Event	The 16'th Danish Conference on Pattern Recognition and Image Analysis - Copenhagen, Denmark Duration: 21 Aug 2008 → …

Conference

Conference	The 16'th Danish Conference on Pattern Recognition and Image Analysis
Country/Territory	Denmark
City	Copenhagen
Period	21/08/2008 → …

Series	Technical Report
Number	08-10
ISSN	0107-8283

AUB Link

Search for the material in Aalborg University Library's search engine

Big Brother <em>is</em> watching you! - Extracting a good quality face image from a poor quality surveillance video<strong/>
Moeslund, T. B., Holte, M. B., Fihl, P., Hansen, D. M. & Nasrollahi, K.
<ingen navn>
01/10/2007 → …
Project: Research

Cite this

@inproceedings{b2a9b97063b911dd92a2000ea68e967b,

title = "Face Image Quality and its Improvement in a Face Detection System",

abstract = "When a person passes by a surveillance camera a sequence of images is obtained. Most of these images are redundant and usually keeping some of them which have better quality is sufficient. So before performing any analysis on the face of a person, the face at the first step needs to be detected. In the second step the quality of the different face images needs to be evaluated. Finally, after choosing the best image(s) based on this quality assessment, in the third step, if this image(s) is not satisfying a predefined set of measures for good quality images, its quality should be improved. In this work we are trying to develop a system to deal with the video sequences in these 3 steps.",

author = "Nasrollahi Kamal and Moeslund, {Thomas B.}",

year = "2008",

language = "English",

series = "Technical Report",

number = "08-10",

pages = "25--27",

editor = "Olsen, {S{\o}ren I.}",

booktitle = "Proceedings of the 16th Danish Conference on Pattern Recognition and Image Analysis",

publisher = "Museum Tusculanum",

note = "The 16'th Danish Conference on Pattern Recognition and Image Analysis ; Conference date: 21-08-2008",

}

Face Image Quality and its Improvement in a Face Detection System. / Kamal, Nasrollahi ; Moeslund, Thomas B.
Proceedings of the 16th Danish Conference on Pattern Recognition and Image Analysis. ed. / Søren I. Olsen. Museum Tusculanum, 2008. p. 25-27 (Technical Report; No. 08-10).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research

TY - GEN

T1 - Face Image Quality and its Improvement in a Face Detection System

AU - Kamal, Nasrollahi

AU - Moeslund, Thomas B.

PY - 2008

Y1 - 2008

N2 - When a person passes by a surveillance camera a sequence of images is obtained. Most of these images are redundant and usually keeping some of them which have better quality is sufficient. So before performing any analysis on the face of a person, the face at the first step needs to be detected. In the second step the quality of the different face images needs to be evaluated. Finally, after choosing the best image(s) based on this quality assessment, in the third step, if this image(s) is not satisfying a predefined set of measures for good quality images, its quality should be improved. In this work we are trying to develop a system to deal with the video sequences in these 3 steps.

AB - When a person passes by a surveillance camera a sequence of images is obtained. Most of these images are redundant and usually keeping some of them which have better quality is sufficient. So before performing any analysis on the face of a person, the face at the first step needs to be detected. In the second step the quality of the different face images needs to be evaluated. Finally, after choosing the best image(s) based on this quality assessment, in the third step, if this image(s) is not satisfying a predefined set of measures for good quality images, its quality should be improved. In this work we are trying to develop a system to deal with the video sequences in these 3 steps.

M3 - Article in proceeding

T3 - Technical Report

SP - 25

EP - 27

BT - Proceedings of the 16th Danish Conference on Pattern Recognition and Image Analysis

A2 - Olsen, Søren I.

PB - Museum Tusculanum

T2 - The 16'th Danish Conference on Pattern Recognition and Image Analysis

Y2 - 21 August 2008

ER -

Face Image Quality and its Improvement in a Face Detection System

Abstract

Conference

AUB Link

Projects

Big Brother <em>is</em> watching you! - Extracting a good quality face image from a poor quality surveillance video<strong/>

Cite this