Facial expression recognition based on bidirectional gated recurrent units within deep residual network  

作　　者：Wenjuan Shen Xiaoling Li 

机构地区：[1]Gongqing College of Nanchang University,Gongqingcheng,China

出　　处：《International Journal of Intelligent Computing and Cybernetics》2020年第4期527-543,共17页智能计算与控制论国际期刊（英文）

基　　金：supported by a fund:science and technology research project of education department of Jiangxi province in 2019.(No GJJ191573).

摘　　要：Purpose-recent years,facial expression recognition has been widely used in human machine interaction,clinical medicine and safe driving.However,there is a limitation that conventional recurrent neural networks can only learn the time-series characteristics of expressions based on one-way propagation information.Design/methodology/approach-To solve such limitation,this paper proposes a novel model based on bidirectional gated recurrent unit networks(Bi-GRUs)with two-way propagations,and the theory of identity mapping residuals is adopted to effectively prevent the problem of gradient disappearance caused by the depth of the introduced network.Since the Inception-V3 network model for spatial feature extraction has too many parameters,it is prone to overfitting during training.This paper proposes a novel facial expression recognition model to add two reduction modules to reduce parameters,so as to obtain an Inception-W network with better generalization.Findings-Finally,the proposed model is pretrained to determine the best settings and selections.Then,the pretrained model is experimented on two facial expression data sets of CKþand Oulu-CASIA,and the recognition performance and efficiency are compared with the existing methods.The highest recognition rate is 99.6%,which shows that the method has good recognition accuracy in a certain range.Originality/value-By using the proposed model for the applications of facial expression,the high recognition accuracy and robust recognition results with lower time consumption will help to build more sophisticated applications in real world.

关 键 词：Facial expression recognition Inception-W model Bi-GRUs structure Spatial and temporal features Deep residual networks 

分 类 号：TP1[自动化与计算机技术—控制理论与控制工程]