基于深度神经网络的动态人脸识别方法及应用

Dynamic Facial Recognition Method Based on Deep Neural Networks and Its Application

人脸识别的核心问题是人脸特征的提取与识别,深度神经网络是人工智能与模式识别领域新兴的研究热点,利用其具备视觉特征自学习的特点,结合基于类间约束与特征间约束的人脸识别方法,提出了一种能适用于室外开放场合,用户无感的动态人脸识别方法。研制了用于平安校园的人脸识别系统,实现了校内人员的人脸识别功能,并在交互系统中实时展现,经过实际部署和长时间运行,对校园内人员实现了有效管控,满足了实际应用需求。

Face feature extraction and recognition is the com issue of face recognition.Deep Neural Network(DNN) is an emerging hot area of research in the field of artificial intelligence and pattern recognition.Combined with the face recognition method based on Constraints between classes and features,this paper proposes a dynamic face recognition method which is suitable for outdoor open occasions and user imperceptions by using its characteristics of the visual feature of self-learning.Then it develops a face recognition system based on the proposed method for peaceful campus to realize the function of face recognition for school staff.And it shows the function in the interactive system at all time.After actual deployment and long time running,it realizes efficient control of the school staff,and meets the demand of actual application.