抑郁症患者表情实时识别系统研究与设计

Research and design of a real⁃time expression recognition system for depressed patients

面部表情信息是评估抑郁症患者行为特点的重要依据之一。文中对当前以深度学习为基础的表情特征提取方法进行改进,解决运行时间长、推广性差的问题;并以此为表情识别算法依据,提出一种抑郁症患者表情实时识别系统。该系统具备表情识别、实时记录、数据管理等功能。经验证,文中算法在Fer2013测试集上的准确率为80.19%,能准确识别患者表情。所提系统的响应时间较短,负载压力较强,可作为自动化手段分析患者表情变化情况,以进行辅助诊断。

Facial expression information is one of the important bases for assessing the behavioural characteristics of depressed patients.The expression feature extraction method based on deep learning is improved to solve the problems of long running time and poor generalisability.A real⁃time expression recognition system for depressed patients is proposed based on this expression recognition algorithm.The system has the functions of expression recognition,real⁃time recording and data management.It is verified that the accuracy of the algorithm in this paper is 80.19%on the Fer2013 test set,which can accurately identify patients′expressions;the response time of the proposed system is short and the load pressure is strong,and it can be used as an automated means to analyse the changes of patients′expressions for auxiliary diagnosis.