摘要
通过使用面具、变更指纹等技术,个人可以轻松冒充他人,这给物理身份识别带来巨大挑战。针对身份识别过程中属性伪造的问题,提出了一种基于逐阶共识计算的虚假物理身份属性检测方法。该方法首先进行身份识别;然后利用错误识别结果近邻关系的差异,逐阶分析得到共识身份;最后分析共识身份在识别结果中的排序,以检测虚假属性。在人脸、指纹和声纹数据库上的实验结果表明,与多数表决方法相比,本文方法能有效检测出虚假物理身份属性,并具有更高的准确率。
Individuals could easily impersonate others by wearing masks or using altered fingerprints or other technologies,which brings enormous challenges to physical identity recognition.Aiming at the existence of fake attributes during identification,a method based on order-of-consensus-calculation is presented for fake physical identity attributes detection.It first executes identity recognition,then uses the differences of proximity relationship of misidentification result to obtain consensus identity with analysis in order,finally detects fake attributes by analyzing the rank of consensus identity in the recognition result.Experimental results on the face,fingerprint and voiceprint databases demonstrate that compared with the majority decision method,the proposed method can detect fake physical identity attributes effectively and has a higher accuracy.
作者
胡瑞敏
张亚浩
李登实
王晓晨
王超
HU Ruimin;ZHANG Yahao;LI Dengshi;WANG Xiaochen;WANG Chao(National Engineering Research Center for Multimedia Software,School of Computer Science,Wuhan University,Wuhan 430072,Hubei,China;Hubei Key Laboratory of Multimedia and Network Communication Engineering,Wuhan University,Wuhan 430072,Hubei,China;School of Mathematics and Computer Science,Jianghan University,Wuhan 430056,Hubei,China)
出处
《武汉大学学报(理学版)》
CAS
CSCD
北大核心
2020年第2期103-110,共8页
Journal of Wuhan University:Natural Science Edition
基金
国家重点研发计划(2017YFB1002803)
国家自然科学基金(U1736206,61701194)。
关键词
可信计算
虚假检测
共识计算
近邻关系
身份识别
trusted computing
forgery detection
consensus-calculation
proximity relationship
identity recognition