首页> 外国专利> FACIAL EXPRESSION RECOGNITION SYSTEM AND METHOD USING MACHINE LEARNING

FACIAL EXPRESSION RECOGNITION SYSTEM AND METHOD USING MACHINE LEARNING

机译:机器学习的表情表达识别系统及方法

摘要

The present invention relates to a face expression recognition system using machine learning and a method for the same. Specifically, the face expression recognition system includes: a detecting module detecting a face area of a user in a video including a face of the user which is input; an extraction module extracting a feature vector on the face of the user from the face area; a classification module classifying the face of the user by using the feature vector; and a recognition module recognizing a face expression of the user from the face of the user. Also, specifically, the face expression recognition system includes: (1) a step in which the detecting module detects the face area of the user in the video including the face of the user which is input; (2) a step in which the extraction module extracts the feature vector on the face of the user from the face area; (3) a step in which the classification module classifies the face of the user by using the feature vector; and (4) a step in which the recognition module recognizes the face expression of the user from the face of the user. According to the present invention, the face expression recognition system using machine learning can improve speed of recognizing the face expression of the user in comparison with an existing face recognition technique.
机译:本发明涉及一种使用机器学习的面部表情识别系统及其方法。具体地,所述面部表情识别系统包括:检测模块,用于在包括输入的用户面部的视频中检测用户的面部区域;提取模块,从面部区域提取用户面部的特征矢量;分类模块,通过特征向量对用户的面部进行分类;识别模块从用户的面部识别用户的面部表情。并且,具体地,所述面部表情识别系统包括:(1)检测模块在包括输入的用户的面部的视频中检测用户的面部区域的步骤; (2)提取模块从面部区域提取用户面部的特征矢量; (3)分类模块通过特征向量对用户的面部进行分类的步骤; (4)识别模块从用户的面部识别用户的面部表情。根据本发明,与现有的面部识别技术相比,使用机器学习的面部表情识别系统可以提高识别用户的面部表情的速度。

著录项

  • 公开/公告号KR20190038203A

    专利类型

  • 公开/公告日2019-04-08

    原文格式PDF

  • 申请/专利权人 LEE IN GYU;

    申请/专利号KR20170128341

  • 发明设计人 BYOUNGCHUL KO;JAEYEAL NAM;JUNG MI RA;

    申请日2017-09-29

  • 分类号G06K9;G06K9/48;

  • 国家 KR

  • 入库时间 2022-08-21 11:51:12

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号