摘要
微表情的变化是非常微小的,这使得微表情的研究非常困难。微表情是不能伪造和压制的,因此也成为判断人们主观情感的重要依据。本文提出了以卷积神经网络及改进长短时记忆网络特征融合为依托的微表情识别方法,先介绍了相关的背景知识,再介绍了实验的预处理过程、特征提取以及相应的特征融合的过程,将所得的结果用于实验模型的预测分类。实验结果表明,新模型具有更好的识别率。
Changes in micro-expressions are very small,which makes it hard to study micro-expression.Besides,micro-expression cannot be forged and suppressed,which makes it an important basis for judging people’s emotions.This paper proposes a micro-expression recognition method based on convolutional neural networks and improved Long-Short-Term Memory(LSTM)network feature fusion.First,it introduces relevant background knowledge and then introduces experimental preprocessing process,feature extraction and corresponding feature fusion process.Results are used in prediction classification of experiment model.Experimental results show that the new model has a better recognition rate.
作者
钱泽锋
钱梦莹
QIAN Zefeng;QIAN Mengying(Department of Computer Science and Technology,Zhejiang Sci-tech University,Hangzhou 310018,China;Shaoxing Environmental Protection Bureau,Shaoxing 312000,China)
出处
《软件工程》
2021年第4期26-29,14,共5页
Software Engineering
关键词
微表情识别
特征融合
微表情分类
卷积神经网络
LSTM
micro-expression recognition
feature fusion
micro-expression classification
convolutional neural network
LSTM
