权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

電子透かしアルゴリズムの安全性評価手法とバイオメトリクスへの応用に関する研究

数字水印算法安全评估方法及其在生物识别中的应用研究

基本信息

批准号：
12450157
负责人：
KOMATSU Naohisa
金额：
$ 4.48万
依托单位：
Waseda University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
2000
资助国家：
日本
起止时间：
2000 至 2002
项目状态：
已结题

项目摘要

In this study, we propose a new watermarking method for speech by manipulating the articulation in the process of speech production. The proposed method is performed by modeling a quasi vocal tract model equivalent to the speech production process. The watermarked vocal tract model is expressed by codebocks made by LSP(Line Spectrum Pair) parameters. The procedure of watermark for speech is as follows ; 1) LSPs are extracted from the speech 2) Some of the extracted LSPs are replaced by the codebook vectors. 3) Speech is synthesized using replaced LSPs. In the process above, watermarks are embedded indirectly into the speech Evaluation tests on speech quality and accuracy of the proposed method a scheme is applied to the ITU-T G. 729 (CS-ACELP).Digital watermark is expected to have robustness against various alterations. However, many watermark do not necessarily have robustness against all kinds of alterations. To avid this problem, we propose a hybrid type of digital watermark. The hybrid type of digital watermark is rather robust than the conventional ones by using some digital watermarks that have robustness against different kinds of alterations.Furthermore, we apply our proposed watermark method to a fingerprint identification system In a fingerprint identification system a leak of a fingerprint data is a serious problem Is is difficult for a user to blame administrators for the leak, because the user cannot prove where the images leaked out from We propose a fingerprint identification system that enables a user to prove it by using digital watermarks.

在本研究中，我们提出了一种新的语音水印方法，通过对语音产生过程中的发音进行控制。提出的方法是通过建立一个相当于语音产生过程的准声道模型来实现的。带水印的声道模型由LSP（Line Spectrum Pair）参数组成的码块表示。语音水印的处理步骤如下：1)从语音中提取lsp； 2)用码本向量替换部分提取的lsp。3)使用替换的lsp合成语音。在上述过程中，将水印间接嵌入到语音中，对所提方法的语音质量和准确性进行了评价测试，并将该方案应用于ITU-T G. 729 （CS-ACELP）。数字水印应具有抗各种变化的鲁棒性。然而，许多水印并不一定具有抗各种变化的鲁棒性。为了解决这个问题，我们提出了一种混合型数字水印。混合类型的数字水印通过使用一些对不同类型的变化具有鲁棒性的数字水印，比传统的数字水印具有更强的鲁棒性。此外，我们将所提出的水印方法应用于指纹识别系统中，在指纹识别系统中，指纹数据的泄露是一个严重的问题。由于用户无法证明图像是从哪里泄露出来的，因此用户很难将泄漏的责任归咎于管理员。我们提出了一种用户可以通过使用数字水印来证明的指纹识别系统。