首页> 外文会议>Science and Information Conference >Biometric linkage between identity document card and its holder based on real-time facial recognition
【24h】

Biometric linkage between identity document card and its holder based on real-time facial recognition

机译:基于实时面部识别的身份证件与其持有人之间的生物识别链接

获取原文

摘要

Access control systems based on RFiD cards are very popular in many companies and public institutions. There is an assumption, that if one has a card and passes verification, the access is granted. Such an approach entails few threats: using of cards being in possession of unauthorized people (stolen or just borrowed), risk of cloning card. We strongly believe that prevention is a better way to obtain desired security level. The usability of the system is essential, but it has to be pragmatic, that is why we accept in that case higher value for False Acceptance Rate, simultaneously getting lower False Rejection Rate. We aim to discourage, in significant way, any attempts of stealing or borrowing access cards from third parties or cloning them. Our solution verifies the biometric linkage between the signed facial image of the document holder embedded in personal identity document and the user facial image captured by a camera during usage of a card. In order to avoid any trials of real-time facial substitution, we introduced depth camera and infrared flash. Our goal is to compare the similarity of faces of document holder and user, but at the same time to prevent misuse of this high quality digital data. In order to support that assumption we process captured image in the reader and send it to the card for matching (match-on-card).
机译:基于RFiD卡的访问控制系统在许多公司和公共机构中非常流行。有一个假设,即如果有人拥有卡并通过验证,则将授予访问权限。这种方法几乎没有威胁:使用未经授权的人员拥有的卡(被盗或刚借的卡),克隆卡的风险。我们坚信,预防是获得所需安全级别的更好方法。系统的可用性至关重要,但是必须实用,这就是为什么在这种情况下我们接受更高的错误接受率值,同时获得更低的错误拒绝率的原因。我们旨在以重大方式阻止任何企图从第三方那里窃取或借用访问卡或对其进行克隆的尝试。我们的解决方案验证了嵌入个人身份文件中的证件夹的已签名面部图像与在使用卡期间由相机捕获的用户面部图像之间的生物识别链接。为了避免进行实时面部替代的任何尝试,我们引入了深度摄像头和红外闪光灯。我们的目标是比较文件持有人和用户面部的相似性,但同时要防止滥用这种高质量的数字数据。为了支持该假设,我们在读取器中处理捕获的图像,并将其发送到卡进行匹配(卡上匹配)。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号