首页 | 本学科首页   官方微博 | 高级检索  
     

基于DCT-BP神经网络的人脸表情识别
引用本文:董戎萍,唐伯良. 基于DCT-BP神经网络的人脸表情识别[J]. 微计算机信息, 2005, 0(18)
作者姓名:董戎萍  唐伯良
作者单位:上海大学通信与信息工程学院 200072(董戎萍),上海大学通信与信息工程学院 200072(唐伯良)
基金项目:上海市科技发展基金项目(7A17746)
摘    要:提出一种基于DCT-BP神经网络的人脸表情识别算法,先对图像进行灰度均衡与图像平滑的预处理,然后利用离散余弦变换提取图像的表情特征参数,变换后的数据量大大减小,而且不会丢失图像所携带的关键信息,最后利用前向反馈神经网络算法进行识别。

关 键 词:表情识别  离散余弦变换  误差向传播算法  前向神经网络

Facial Expression Recognition Using DCT- Back Propagation Neural Network
Dong,Rongping Tang,Boliang. Facial Expression Recognition Using DCT- Back Propagation Neural Network[J]. Control & Automation, 2005, 0(18)
Authors:Dong  Rongping Tang  Boliang
Affiliation:(School of Communication and Information Engi- neering,Shanghai University,Shanghai 200072) Dong,Rongping Tang,Boliang
Abstract:An effective method for facial expression recognition is proposed, which uses the discrete cosine transform (DCT) and error back propagation (BP) neutral network. First, processing including normalization and filtering is carried out on facial im- ages. Then two- dimensional discrete cosine transform (2 - D DCT) is introduced. After that, the size of expression feature da- ta is reduced without sacrificing key attributes that play impor- tant role in the recognition process. Finally, classification is made based on the back propagation (BP) algorithm.
Keywords:facial expression recognition   discrete cosine trans- form   error back propagation algorithm   feedforward neural network
本文献已被 CNKI 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