基于金字塔分割注意力和联合损失的表情识别模型  

An Expression Recognition Model Based on Pyramid Split Attention and Joint Loss

作　　者：谷瑞 顾家乐[2] 宋翠玲 GU Rui;GU Jiale;SONG Cuiling(School of Digital Economy and Management,Nanjing University,Nanjing 210003,China;Suzhou Industrial Park Institute of Services Outsourcing,Suzhou 215123,China)

机构地区：[1]南京大学数字经济与管理学院,南京210003 [2]苏州工业园区服务外包职业学院,苏州215123

出　　处：《数据采集与处理》2024年第6期1493-1504,共12页Journal of Data Acquisition and Processing

基　　金：2023年江苏省高职院校教师专业带头人高端研修项目(2023TDFX010)。

摘　　要：如何提取多尺度特征和建模远程通道间的语义依赖仍是表情识别网络面临的挑战。本文提出一种基于金字塔分割注意力的残差网络(Residual network based on pyramid split attention, PSA-ResNet)模型,该模型将ResNet50残差模块中的3×3卷积替换成金字塔分割注意力,以有效提取多尺度特征,增强跨通道语义信息的相关性。同时,为缩小同类表情之间的差异,扩大不同类表情之间的距离,在训练过程中引入了Softmax loss和Center loss联合损失函数优化模型参数。本文所提出的方法在Fer2013和CK+两个公开的数据集上进行仿真实验,分别取得了74.26%和98.35%的准确率,进一步证实了该方法相比前沿算法具有更好的表情识别效果。How to extract multi-scale features and model semantic dependencies between remote channels remains a challenge for expression recognition networks.This paper proposes a residual network based on pyramid split attention(PSA-ResNet),which replaces the 3×3 convolution in the ResNet50 residual module with PSA to effectively extract multi-scale features and enhance the correlation of cross channel information.In order to reduce the differences between similar expressions and expand the distance between different types of expressions,a joint loss function optimization parameter of Softmax loss and Center loss is introduced during the training process.The proposed model is simulated on two publicly available datasets,Fer2013 and CK+,and achieves accuracies of 74.26%and 98.35%,respectively,further confirming that this method has better recognition results compared to cutting-edge algorithms.

关 键 词：表情识别 金字塔分割注意力 多尺度特征 残差网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]