[1]姜丽莉,黄承宁.融合注意力机制改进残差网络的表情识别方法[J].计算机技术与发展,2022,32(05):42-46.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 007]
　JIANG Li-li,HUANG Cheng-ning.An Expression Recognition Method Based on Fusion of Attention Mechanism and Improved Residual Network[J].,2022,32(05):42-46.[doi:10. 3969 / j. issn. 1673-629X. 2022. 05. 007]
点击复制

融合注意力机制改进残差网络的表情识别方法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年05期
页码:
42-46
栏目:
图形与图像
出版日期:
2022-05-10

文章信息/Info

Title:
An Expression Recognition Method Based on Fusion of Attention Mechanism and Improved Residual Network
文章编号:
1673-629X(2022)05-0042-05
作者:
姜丽莉; 黄承宁
南京工业大学浦江学院 计算机与通信工程学院,江苏 南京 211200
Author(s):
JIANG Li-li; HUANG Cheng-ning
School of Computer and Communication Engineering,Nanjing Technology University Pujiang Institute,Nanjing 211200,China
关键词:
表情识别; 神经网络; 深度学习; 通道注意力机制; 残差网络
Keywords:
expression recognition; neural network; deep learning; channel attention mechanism; residual network
分类号:
TP391. 4
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 05. 007
摘要:
为提高大数据挖掘过程中表情识别的计算速度和准确率,在 ResNet-50 模型的基础上,融合通道注意力机制与改进残差网络,提出一种表情识别方法的改进模型( SE-ResNet-50 +Swish) 。 改进模型在 ResNet-50 的基础上,引入多个带有通道注意力模块 SE 的特征层对表情样本进行特征提取,利用注意力机制增强关键的特征通道,增强网络的特征表达与鲁棒性,且能够有效减少计算量,并利用激活函数 Swish 替代 ReLU 激活函数,以达到进一步提升表情识别准确率的目的。在 CAS-PEAL-R1 数据库上进行验证的结果表明,SE-ResNet-50 在 ResNet-50 的基础上引入 SE 模块之后,虽然增加了网络层数,但计算速度以及表情识别的准确率有明显提高;改进模型利用 Swish 替代 ReLU 后,相比于 SE-ResNet-50 的参数数量与计算量等无显著增多,但表情识别准确率有提升;以上结果表明,改进模型能够有效减少计算量,并增强网络的特征表达与鲁棒性,从而达到提升表情识别计算速度与识别准确率的目的。
Abstract:
In order to improve the computational speed and accuracy of expression recognition in big data mining,an improved expression recognition? model ( SE-ResNet-50+Swish) is proposed based on ResNet-50 model,which combines channel attention mechanism and improved residual network. On the basis of ResNet - 50,the improved model introduces multiple feature layers with channel attention module SE to extract the features of the expression samples. The attention mechanism is used to enhance the key feature channels,enhance the feature expression and robustness of the network,and effectively reduce the amount of calculation. The activation function Swish is used to replace the ReLU activation function in order to further improve the accuracy of expression recognition. The results on CAS-PEAL-R1 database show that after SE - ResNet - 50 introduces SE module on the basis of ResNet - 50,although the number of network layers is increased,the calculation speed and the accuracy of expression recognition are significantly improved. Compared with SE-ResNet-50,the number of parameters and the amount of calculation of the improved model using Swish instead of ReLU are not significantly increased,but the accuracy of expression recognition is improved. The above results show that the improved model can effectively reduce the amount of computation, and enhance the feature expression and robustness of the network, so as to achieve the purpose of improving the computational speed and recognition accuracy of expression recognition.

相似文献/References:

[1]路川 胡欣杰.区域航空市场航线客流量预测研究[J].计算机技术与发展,2010,(04):84.
　LU Chuan,HU Xin-jie.Analysis of Regional Airline Passenger Forecast Title[J].,2010,(05):84.
[2]高峥 陈蜀宇 李国勇.混合入侵检测系统的研究[J].计算机技术与发展,2010,(06):148.
　GAO Zheng,CHEN Shu-yu,LI Guo-yong.Research of a Hybrid Intrusion Detection System[J].,2010,(05):148.
[3]包力伟 周俊.铸锻企业生产质量控制系统的开发[J].计算机技术与发展,2008,(04):174.
　BAO Li-wei,ZHOU Jun.Development of a Manufacture Quality Control System in Casting Company[J].,2008,(05):174.
[4]李志俊 程家兴 金奎 饶玉佳.基于样本期望训练数的BP神经网络改进研究[J].计算机技术与发展,2009,(05):103.
　LI Zhi-jun,CHENG Jia-xing,JIN Kui,et al.BP Algorithm Improvement Based on Sample Expected Training Number[J].,2009,(05):103.
[5]李龙澍 葛瑞峰 王慧萍.基于神经网络的批强化学习在Robocup中的应用[J].计算机技术与发展,2009,(07):98.
　LI Long-shu,GE Rui-feng,WANG Hui-ping.Application of Batch Reinforcement Learning Based on NN to Robocup[J].,2009,(05):98.
[6]贾志先.神经网络在空白试卷识别中的应用[J].计算机技术与发展,2009,(08):208.
　JIA Zhi-xian.Application of Neural Network in Recognization Blank Examination Paper[J].,2009,(05):208.
[7]肖宜龙 路游 亓永刚.基于神经网络的NURBS曲面重建[J].计算机技术与发展,2009,(09):65.
　XIAO Yi-long,LU You,QI Yong-gang.NURBS Surface Reconstruction Based on Neural Network[J].,2009,(05):65.
[8]蔡秋茹 罗烨 柳益君 叶飞跃.企业资信的BP神经网络评估模型研究[J].计算机技术与发展,2009,(10):117.
　CAI Qiu-ru,LUO Ye,LIU Yi-jun,et al.Research on BP Neural Network Model for Corporation Credit Rating[J].,2009,(05):117.
[9]王晓敏 刘希玉 戴芬.BP神经网络预测算法的改进及应用[J].计算机技术与发展,2009,(11):64.
　WANG Xiao-min,LIU Xi-yu,DAI Fen.Improvement and Application of BP Neural Network Forecasting Algorithm[J].,2009,(05):64.
[10]崔海青 刘希玉.基于粒子群算法的RBF网络参数优化算法[J].计算机技术与发展,2009,(12):117.
　CUI Hai-qing,LIU Xi-yu.Parameter Optimization Algorithm of RBF Neural Network Based on PSO Algorithm[J].,2009,(05):117.

更新日期/Last Update: 2022-05-10