首页 | 本学科首页   官方微博 | 高级检索  
     

基于CNN集成学习的人脸表情识别系统的设计
引用本文:陈佳,褚丽莉,周影. 基于CNN集成学习的人脸表情识别系统的设计[J]. 电脑与信息技术, 2021, 29(1): 10-12. DOI: 10.3969/j.issn.1005-1228.2021.01.004
作者姓名:陈佳  褚丽莉  周影
作者单位:辽宁工业大学,辽宁锦州 121000;辽宁工业大学,辽宁锦州 121000;辽宁工业大学,辽宁锦州 121000
摘    要:随着计算机计算资源的提升以及深度学习理论的不断丰富,自动的人脸表情识别技术已经得到了进一步的发展。但由于表情存在复杂性以及微妙性,实现实时的人脸表情识别仍是一大难题。文章设计了一种基于CNN集成学习的人脸表情识别系统,该系统在FER2013数据集上表情的识别准确率达到70.84%,能够实现实时的、高精度的表情识别。

关 键 词:深度学习  CNN  集成学习  FER2013  表情识别

Design of Facial Expression Recognition System Based on CNN Integrated Learning
CHEN Jia,CHU Li-li,ZHOU Ying. Design of Facial Expression Recognition System Based on CNN Integrated Learning[J]. Computer and Information Technology, 2021, 29(1): 10-12. DOI: 10.3969/j.issn.1005-1228.2021.01.004
Authors:CHEN Jia  CHU Li-li  ZHOU Ying
Affiliation:(Liaoning University of Technology,Jinzhou,121000,China)
Abstract:With the improvement of computer computing resources and the continuous enrichment of deep learning theory,automatic facial expression recognition technology has been further developed.But because of the complexity and subtlety of facial expressions,it is still a big problem to realize real-time facial expression recognition.Therefore,we design a facial expression recognition system based on CNN integrated learning in this paper,which can realize real-time and high-performance facial expression recognition with an accuracy rate of 70.84%on FER2013 data set.
Keywords:Deep learning  CNN  Integrated learning  FER2013  Facial expression recognition
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号