摘要
针对人脸面部表情特征复杂且目标较小,导致大量人工智能算法难以稳定识别的问题,提出一种多流残差网络结合改进支持向量机(support vector machine,SVM)模型的面部表情识别方法。该方法由信息增强、表情特征提取和表情分类三阶段组成。首先,提出自适应多流信息增强模块,突出图像关键信息,提升特征关联程度;其次,提出残差交互融合模块,提取特征图像的空间信息并突出面部表情特征,输出3条不同尺度的特征图像保证后续稳定识别;最后,对3条特征图像使用改进后的SVM进行表情分类,输出识别结果。实验结果表明:所提方法在CK+、FER2013及JAFFE数据集上的准确率分别达到98.57%、77.28%、96.24%,均优于对比的经典及新颖算法,为AI表情识别领域提供了新思路。
The complex facial expressions and small targets make it difficult for many existing artificial intelligence algorithms to achieve consistent identification.Thus,a multi-stream residual network integrated with an improved Support Vector Machine(SVM)for facial expression recognition is proposed.The method consists of three steps,namely information enhancement,expression feature extraction and expression classification.First,an adaptive multi-stream information enhancement module is proposed to highlight key image information and enhance feature correlation.Next,a residual interaction module is proposed to extract the spatial information of feature images,highlight facial expression features,and output three feature images with different scales.Last,improved SVM is employed to classify the expression of three-branch feature images to further enhance the robustness of the algorithm.The experimental results show that the accuracy of the proposed method on CK+,FER2013 and JAFFE data sets reaches 98.57%,77.28%and 96.24%respectively,all of which are superior to other classic and novel algorithms.Thus,it provides new insights in the recognition of facial expressions by AI.
作者
郝秉华
吴华
HAO Binghua;WU Hua(School of Computer Information Management,Inner Mongolia University of Finance and Economics,Hohhot 100010,China)
出处
《重庆理工大学学报(自然科学)》
北大核心
2023年第11期157-165,共9页
Journal of Chongqing University of Technology:Natural Science
基金
国家自然科学基金项目地区项目(71962025)
内蒙古自治区自然科学基金项目(2023LHMS07002)。
关键词
深度学习
表情识别
支持向量机
交互残差
信息增强
deep learning
expression recognition
support vector machine
interactive residual
information enhancement
作者简介
郝秉华,女,硕士,讲师,主要从事图像识别研究,E-mail:32730299@qq.com。
