基于脑功能网络连接的隐藏信息检测研究  

Study of Concealed Information Test Based on Functional Brain Network

作　　者：常文文[1] 王宏[1] 化成城 王翘秀[1] 原玥[1] 刘冲[1] CHANG Wen-wen;WANG Hong;HUA Cheng-cheng;WANG Qiao-xiu;YUAN Yue;LIU Chong(School of Mechanical Engineering and Automation,Northeast University Shenyang 110819)

机构地区：[1]东北大学机械工程与自动化学院,沈阳110819

出　　处：《电子科技大学学报》2018年第5期775-780,共6页Journal of University of Electronic Science and Technology of China

基　　金：国家自然科学基金(51405073;51505069)

摘　　要：大脑在视觉或听觉刺激作用下对是否隐藏的信息有不同的认知反应,而大脑的这种反应涉及到不同脑区的协同和信息流动。该文基于传统视觉刺激隐藏信息测试方法,设计了视听同步刺激的对比试验;并针对当前测试方法主要集中于脑中央区电极点的这一缺点,通过记录全脑区导联的信号来分析不同脑区神经活动的变化。首先用视觉刺激和视听同步刺激相关脑电位构建了脑功能网络,并计算脑网络聚集系数和特征路径长度作为基本特征量,同时构建了一种量子门节点神经网络分类器,将其应用于脑电特征的分类。实验结果表明,结合脑网络特征和量子神经网络分类器的方法,能够较为准确地识别隐藏信息,同时视听同步刺激效果好于视觉刺激。Brain has different cognition responses to the concealed information under visual and auditory stimuli,and this process involves the coordination and information flow between different regions.In this paper,based on the traditional visual stimuli for concealed information,we designed the video-audio synchronization test for comparison.For the defect in current research that mainly focus on the electrodes in central of the brain,we recorded the signals from the whole brain to reflect the neural activity of the brain.Firstly,we constructed the brain functional network using the visual and video-audio stimuli related potentials,then calculated the clustering coefficient and path length as the features of the signals,lastly,we build a quantum gated neural network as the classification for the features.The experimental results show that combining the characteristics of brain network with quantum neural network classifier,the concealed information can be indentified accurately,and the video-audio stimuli is better than visual stimuli.

关 键 词：脑网络 隐藏信息测试 脑电 量子神经网络 视听刺激 

分 类 号：R318.04[医药卫生—生物医学工程] R853[医药卫生—基础医学]