基于视频图像的人脸面部表情快速识别研究

Research of Rapid Facial Expression Recognition based on Video Images

传统的信息安全管理方式虽然可以在一定程度上维护个人的信息安全,但是存在数据泄密、密码盗用等弊端。为提高信息安全性,可以采用人脸面部表情识别进行信息安全的维护。为实现对人脸面部表情的快速、准确识别,建立人脸面部表情识别系统。该系统主要包括图像检测模块、人脸自动检测模块、特征提取模块和表情识别模块。通过对系统的各模块进行设计,并对获取的图像进行预处理、人脸初识别和人脸验证,确定了人脸位置并可以追踪。充分考虑了人脸面部表情特征,确定人脸提取特征以及提取算法,并对特征进行分类、识别。采用BP神经网络作为分类器进行表情的自动识别,并对相关算法进行了设计。为验证该人脸面部表情识别系统的性能,对其进行人脸识别试验和人脸表情快速识别试验。试验结果表明:该系统可以实现对人脸表情的快速、准确识别,符合人们对人脸面部表情识别系统的要求。

Although the traditional information security management could maintain personal information security to a certain extent,there are drawbacks such as data leakage and password theft.To improve information security,facial expression recognition could be used to maintain information security.The system was mainly constituted of image detection module automatic face detection module,feature extraction module and expression recognition module.Through the design of each module of the system,and the acquired image preprocessing,face initial recognition and face verification,determine the face position and could be traced.Fully consider the face facial expression features,determine the face extraction features and extraction algorithm,and the feature classification,recognition.BP neural network was used as classifier for automatic expression recognition,and the correlation algorithm was designed.To verify the performance of the facial expression recognition system,face recognition test and facial expression fast recognition test were carried out.The test results show that the system could realize the rapid and accurate recognition of facial expression,which meets the requirements of people on facial expression recognition system.