<?xml version="1.0"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD Journal Publishing DTD v2.1 20050630//EN" "http://uploads.ingentaconnect.com/docs/dtd/ingenta-journalpublishing.dtd">
<article article-type="research-article">
  <front>
    <journal-meta>
      <journal-id journal-id-type="aggregator">72010350</journal-id>
      <journal-title>Color and Imaging Conference</journal-title>
      <abbrev-journal-title>color imaging conf</abbrev-journal-title>
      <issn pub-type="ppub">2166-9635</issn><issn pub-type="epub"/>
      <publisher>
        <publisher-name>Society of Imaging Science and Technology</publisher-name>
        <publisher-loc>7003 Kilworth Lane, Springfield, VA 22151, USA</publisher-loc>
      </publisher>
    </journal-meta>
    <article-meta><article-id pub-id-type="doi">10.2352/CIC.2001.9.1.art00031</article-id>
      <article-id pub-id-type="sici">2166-9635(20010101)2001:1L.168;1-</article-id>
      <article-id pub-id-type="publisher-id">cic_v2001n1/splitsection31.xml</article-id>
      <article-id pub-id-type="other">/ist/cic/2001/00002001/00000001/art00031</article-id>
      <article-categories>
        <subj-group>
          <subject>Articles</subject>
        </subj-group>
      </article-categories>
      <title-group>
        <article-title>Which color similarity measure is most effective for background-frame differencing ?</article-title>
      </title-group>
      <contrib-group>
        <contrib>
          <name>
            <surname>Hild</surname>
            <given-names>Michael</given-names>
          </name>
        </contrib>
        <contrib>
          <name>
            <surname>Emura</surname>
            <given-names>Toshinobu</given-names>
          </name>
        </contrib>
      </contrib-group>
      <pub-date>
        <day>01</day>
        <month>01</month>
        <year>2001</year>
      </pub-date>
      <volume>2001</volume>
      <issue>1</issue>
      <fpage>168</fpage>
      <lpage>173</lpage>
      <permissions>
        <copyright-year>2001</copyright-year>
      </permissions>
      <abstract>
        <p>We examine twelve known color similarity measures with regard to their effectiveness for the background-frame differencing task. The RGB sensor space and the CIE L*a*b* color space is used to represent color. Based on experiments conducted on twelve different scenes (indoor
 as well as outdoor scenes under various lighting conditions and photographed with two different types of CCD cameras) we show that the <italic>Absolute-value exponent method</italic> is superior.</p>
      </abstract>
    </article-meta>
  </front>
</article>
