基于深度学习的人脸表情识别技术分析
Analysis of Facial Expression Recognition Technology Based on Deep Learning
赵彬宇1
作者信息
摘要
阐述一种新的基于深度学习的表情识别方法,可以对已有表情识别方法中的特征提取进行优化,采用神经网络(NN)和支持向量机(SVM)的集成分类器对正常、快乐、悲伤、惊讶、恐惧和愤怒等面部表情进行分类,并使用JAFFE、CK+、Pie数据集和一些真实世界的图像评估所提出优化算法的性能.
Abstract
This paper describes a new deep learning based expression recognition method that can optimize feature extraction in existing expression recognition methods.An integrated classifier of neural networks(NN)and support vector machines(SVM)is used to classify facial expressions such as normal,happy,sad,surprised,fearful,and angry.The performance of the proposed optimization algorithm is evaluated using JAFFE,CK+,Pie datasets,and some real-world images.
关键词
人脸表情识别/深度学习/集成学习/卷积神经网络Key words
facial expression recognition/deep learning/ensemble learning/convolutional neural networks引用本文复制引用
出版年
2024
NETL
NSTL
万方数据