Univariant assessment of the quality of images

To evaluate the quality of images, most methods compare a degraded image to a perfect reference. Nevertheless in many cases, a reference does not exist. We propose an original univariant (i.e., without a reference) method based on the use of artificial neural networks. The principle behind it is to...

Full description

Saved in:
Bibliographic Details
Published in:Journal of electronic imaging 2002, Vol.11 (3), p.354-364
Main Authors: Jung, Mathieu, Le´ger, Dominique, Gazalet, Marc
Format: Article
Language:English
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3
cites cdi_FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3
container_end_page 364
container_issue 3
container_start_page 354
container_title Journal of electronic imaging
container_volume 11
creator Jung, Mathieu
Le´ger, Dominique
Gazalet, Marc
description To evaluate the quality of images, most methods compare a degraded image to a perfect reference. Nevertheless in many cases, a reference does not exist. We propose an original univariant (i.e., without a reference) method based on the use of artificial neural networks. The principle behind it is to first teach a neural network to assess image quality using images taken from a pool of known examples, then use it to assess the quality of unknown images. The defects considered are compression artifacts, ringing, local singularities, etc. To simplify, only images with defects that are not mixed with one another were first used. Two illustrative examples are presented: assessment of the quality of JPEG compressed images and detection of local defects. The quality of the images is assessed without a reference and with error less than 6 -7 compared to the bivariant method that was learned. Our method can even be used to model some very simple visual comportment. ©
doi_str_mv 10.1117/1.1482096
format article
fullrecord <record><control><sourceid>hal_cross</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_00149735v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>oai_HAL_hal_00149735v1</sourcerecordid><originalsourceid>FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3</originalsourceid><addsrcrecordid>eNpFUE1PwzAMjRBIjMGBf9ArhxY7H01ynKbChoq4MIlblHUJK-rW0ZRJ-_ekbBoH28_We7b1CLlHyBBRPmKGXFHQ-QUZocghpVR_XEYMKFOtQV-TmxC-ABAVxxHJFtt6b7vabvvEhuBC2LgIW5_0a5d8_9im7g9DW2_spwu35MrbJri7Ux2TxVPxPp2l5dvzfDop04oB71PJpBJKos01A-VzJZVlEXNlPee0WnHlNM2FiH9IVjmtKp7bJVDgS0_Zio3Jw3Hv2jZm18Xj3cG0tjazSWmGWRRyLZnY4z-36toQOufPAgQzmGLQnEyJXHrkhl3tzryXYv5axF8GV4YMbAjB_zCyX9JZXlw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Univariant assessment of the quality of images</title><source>SPIE Digital Library</source><creator>Jung, Mathieu ; Le´ger, Dominique ; Gazalet, Marc</creator><creatorcontrib>Jung, Mathieu ; Le´ger, Dominique ; Gazalet, Marc</creatorcontrib><description>To evaluate the quality of images, most methods compare a degraded image to a perfect reference. Nevertheless in many cases, a reference does not exist. We propose an original univariant (i.e., without a reference) method based on the use of artificial neural networks. The principle behind it is to first teach a neural network to assess image quality using images taken from a pool of known examples, then use it to assess the quality of unknown images. The defects considered are compression artifacts, ringing, local singularities, etc. To simplify, only images with defects that are not mixed with one another were first used. Two illustrative examples are presented: assessment of the quality of JPEG compressed images and detection of local defects. The quality of the images is assessed without a reference and with error less than 6 -7 compared to the bivariant method that was learned. Our method can even be used to model some very simple visual comportment. ©</description><identifier>ISSN: 1017-9909</identifier><identifier>EISSN: 1560-229X</identifier><identifier>DOI: 10.1117/1.1482096</identifier><identifier>CODEN: JEIME5</identifier><language>eng</language><publisher>SPIE and IS&amp;T</publisher><ispartof>Journal of electronic imaging, 2002, Vol.11 (3), p.354-364</ispartof><rights>2002 SPIE and IS&amp;T</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3</citedby><cites>FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://www.spiedigitallibrary.org/journalArticle/Download?urlId=10.1117/1.1482096$$EPDF$$P50$$Gspie$$H</linktopdf><linktohtml>$$Uhttp://dx.doi.org/10.1117/1.1482096$$EHTML$$P50$$Gspie$$H</linktohtml><link.rule.ids>230,314,780,784,885,4024,18965,27923,27924,27925,55386,55387</link.rule.ids><backlink>$$Uhttps://hal.science/hal-00149735$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Jung, Mathieu</creatorcontrib><creatorcontrib>Le´ger, Dominique</creatorcontrib><creatorcontrib>Gazalet, Marc</creatorcontrib><title>Univariant assessment of the quality of images</title><title>Journal of electronic imaging</title><description>To evaluate the quality of images, most methods compare a degraded image to a perfect reference. Nevertheless in many cases, a reference does not exist. We propose an original univariant (i.e., without a reference) method based on the use of artificial neural networks. The principle behind it is to first teach a neural network to assess image quality using images taken from a pool of known examples, then use it to assess the quality of unknown images. The defects considered are compression artifacts, ringing, local singularities, etc. To simplify, only images with defects that are not mixed with one another were first used. Two illustrative examples are presented: assessment of the quality of JPEG compressed images and detection of local defects. The quality of the images is assessed without a reference and with error less than 6 -7 compared to the bivariant method that was learned. Our method can even be used to model some very simple visual comportment. ©</description><issn>1017-9909</issn><issn>1560-229X</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2002</creationdate><recordtype>article</recordtype><recordid>eNpFUE1PwzAMjRBIjMGBf9ArhxY7H01ynKbChoq4MIlblHUJK-rW0ZRJ-_ekbBoH28_We7b1CLlHyBBRPmKGXFHQ-QUZocghpVR_XEYMKFOtQV-TmxC-ABAVxxHJFtt6b7vabvvEhuBC2LgIW5_0a5d8_9im7g9DW2_spwu35MrbJri7Ux2TxVPxPp2l5dvzfDop04oB71PJpBJKos01A-VzJZVlEXNlPee0WnHlNM2FiH9IVjmtKp7bJVDgS0_Zio3Jw3Hv2jZm18Xj3cG0tjazSWmGWRRyLZnY4z-36toQOufPAgQzmGLQnEyJXHrkhl3tzryXYv5axF8GV4YMbAjB_zCyX9JZXlw</recordid><startdate>2002</startdate><enddate>2002</enddate><creator>Jung, Mathieu</creator><creator>Le´ger, Dominique</creator><creator>Gazalet, Marc</creator><general>SPIE and IS&amp;T</general><scope>AAYXX</scope><scope>CITATION</scope><scope>1XC</scope></search><sort><creationdate>2002</creationdate><title>Univariant assessment of the quality of images</title><author>Jung, Mathieu ; Le´ger, Dominique ; Gazalet, Marc</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2002</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Jung, Mathieu</creatorcontrib><creatorcontrib>Le´ger, Dominique</creatorcontrib><creatorcontrib>Gazalet, Marc</creatorcontrib><collection>CrossRef</collection><collection>Hyper Article en Ligne (HAL)</collection><jtitle>Journal of electronic imaging</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Jung, Mathieu</au><au>Le´ger, Dominique</au><au>Gazalet, Marc</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Univariant assessment of the quality of images</atitle><jtitle>Journal of electronic imaging</jtitle><date>2002</date><risdate>2002</risdate><volume>11</volume><issue>3</issue><spage>354</spage><epage>364</epage><pages>354-364</pages><issn>1017-9909</issn><eissn>1560-229X</eissn><coden>JEIME5</coden><abstract>To evaluate the quality of images, most methods compare a degraded image to a perfect reference. Nevertheless in many cases, a reference does not exist. We propose an original univariant (i.e., without a reference) method based on the use of artificial neural networks. The principle behind it is to first teach a neural network to assess image quality using images taken from a pool of known examples, then use it to assess the quality of unknown images. The defects considered are compression artifacts, ringing, local singularities, etc. To simplify, only images with defects that are not mixed with one another were first used. Two illustrative examples are presented: assessment of the quality of JPEG compressed images and detection of local defects. The quality of the images is assessed without a reference and with error less than 6 -7 compared to the bivariant method that was learned. Our method can even be used to model some very simple visual comportment. ©</abstract><pub>SPIE and IS&amp;T</pub><doi>10.1117/1.1482096</doi><tpages>11</tpages></addata></record>
fulltext fulltext
identifier ISSN: 1017-9909
ispartof Journal of electronic imaging, 2002, Vol.11 (3), p.354-364
issn 1017-9909
1560-229X
language eng
recordid cdi_hal_primary_oai_HAL_hal_00149735v1
source SPIE Digital Library
title Univariant assessment of the quality of images
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-07T17%3A45%3A31IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-hal_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Univariant%20assessment%20of%20the%20quality%20of%20images&rft.jtitle=Journal%20of%20electronic%20imaging&rft.au=Jung,%20Mathieu&rft.date=2002&rft.volume=11&rft.issue=3&rft.spage=354&rft.epage=364&rft.pages=354-364&rft.issn=1017-9909&rft.eissn=1560-229X&rft.coden=JEIME5&rft_id=info:doi/10.1117/1.1482096&rft_dat=%3Chal_cross%3Eoai_HAL_hal_00149735v1%3C/hal_cross%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c304t-73785871a69308f6878a369348af442cd48e9265500173ce98c46ab0204bf23d3%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true