%0 Journal Article
%T Color Pixel Categorization Based on ART2 Network<br>基于ART2网络的彩色像素分析及其应用
%A CHEN Zhong
%A CAI Zi-xing
%A YE Qing
%A <br>陈众
%A 蔡自兴
%A 叶青
%J 中国图象图形学报
%D 2008
%I 
%X Pixel analysis is the primary step for region,shape and texture analysis and even for semantic analysis. The correctness of the color and luminance analysis of certain pixel in an image or a series of video streams is a guarantee to the acceptable result of other image processing. Based on a reasonable mapping operation to vectors in RGB color space,this paper applies the ART2 to the layered detecting approach to categorize color pixels. The processing steps and final results not only demonstrate the functions of "neoteny learning" and adjusting of vigilance value,but also illustrate that the method is coherent with the human psychological and physiological process of observing an image and also has strong adaptability for shadow noise suppression.
%K pixel analysis
%K shadow
%K ART2
%K image processing<br>像素分析
%K 阴影
%K ART2
%K 图像处理
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=D06194629680C940ACE75262F54B9D85&aid=A1EC5B012FA03D48EB8A241C72A52C11&yid=67289AFF6305E306&vid=FC0714F8D2EB605D&iid=E158A972A605785F&sid=20ED669EB429E15C&eid=9A596D09E9486F3E&journal_id=1006-8961&journal_name=中国图象图形学报&referenced_num=0&reference_num=15