%0 Journal Article
%T DCT-domain Watermark Detection Using Gaussian Mixture Model<br>基于高斯混合模型的DCT域水印检测方法
%A LIN Xiao-Dan
%A <br>林晓丹
%J 自动化学报
%D 2012
%I 
%X An audio watermark detector is developed based on the statistical characteristics of audio in DCT domain. Watermarks are embedded in DCT domain on a spread spectrum basis. A Gaussian mixture model (GMM) is introduced to model both the watermarked and unwatermarked audio signals in DCT domain. At the receiver a maximum likelihood detector is employed. Simulation results indicate that the proposed method allows for common signal processing and provides a better detecting accuracy compared with the linear correlation detector.
%K Watermark detection
%K DCT domain
%K Gaussian mixture model (GMM)
%K maximum likelihood<br>水印检测
%K DCT域
%K 高斯混合模型
%K 最大似然
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=E76622685B64B2AA896A7F777B64EB3A&aid=21CB0722CA48644F8177CA24A4DDB868&yid=99E9153A83D4CB11&vid=16D8618C6164A3ED&iid=9CF7A0430CBB2DFD&sid=B93F010195432A97&eid=477F4C97C1F396EF&journal_id=0254-4156&journal_name=自动化学报&referenced_num=0&reference_num=15