Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Dynamic Facial Expression Recognition Method

A technology for facial expression recognition and facial expression, which is applied in the field of image features or characteristics of graphics, and can solve problems such as being easily affected by light, affecting the recognition rate of facial expressions, and high feature dimension and time complexity

Active Publication Date: 2020-08-04
HEBEI UNIV OF TECH +1
View PDF22 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to provide a dynamic human facial expression recognition method, which is a dynamic human facial expression recognition method based on geometric features and semantic features, which overcomes the common problems of poor real-time performance, easy to be affected by light, and characteristic features in the prior art. High dimensionality and time complexity affect the defect that the facial expression recognition rate meets the requirements

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Dynamic Facial Expression Recognition Method
  • Dynamic Facial Expression Recognition Method
  • Dynamic Facial Expression Recognition Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0112] A kind of dynamic facial expression recognition method based on geometric features and semantic features of the present embodiment, the specific steps are as follows:

[0113] The first step is the preprocessing of the dynamic face image sequence:

[0114] First normalize the size of each frame of facial expression images in the input dynamic face image sequence to a size of 640×480 pixels, and then use the following Equation (1) is converted from RGB space to grayscale space, and each frame of facial expression grayscale image I is obtained gray_tn ,

[0115] I gray_tn =0.299I R +0.587I G +0.114I B (1),

[0116] In formula (1), I R , I G , I B They are the three channel components of red, green and blue of each frame of facial expression image in the input dynamic facial image sequence, and keep each frame of facial expression grayscale image I gray_tn , for facial expression frame detection and feature point labeling in the second step below;

[0117] In t...

Embodiment 2

[0221] This embodiment is an experimental verification of the dynamic facial expression recognition method of the present invention.

[0222] A. Select 262 dynamic face image sequences in the CK+ data set. Each dynamic face image sequence contains 2 images, namely neutral frame and peak frame, and a total of 524 facial expression image frames are used for experiments.

[0223]In the CK+ data set, the geometric feature extraction method TGF and the geometric semantic feature extraction method SA-TGF in the present invention, the recognition rate obtained after the ten-fold cross-validation experiment is compared with the literature 1, literature 2, literature 3, and literature in the background technology. The recognition rate comparison in 4 is shown in Table 17:

[0224] Table 17. Comparison of recognition rates on CK+ dataset

[0225]

[0226] B. Select 208 dynamic face image sequences in the MMI data set, each dynamic face image sequence contains 2 images, namely the ne...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a dynamic facial expression recognition method, relates to a method for recognizing image features or characteristics of graphs, and discloses a dynamic facial expression recognition method based on geometric features and semantic features. The method comprises the following steps of: preprocessing a dynamic facial image sequence; carrying out facial expression frame detection and feature point labeling on the facial expression grayscale image; calibrating a facial expression triangular area on the facial expression grayscale image; extracting geometric features of a facial expression triangular region on the facial expression grayscale image; analyzing and extracting semantic features on the face expression grayscale image; training the SVM classifier to obtain a classification result; and completing the recognition of the dynamic facial expression. The method overcomes the defects that the real-time performance is poor, the method is easily influenced by illumination, the feature dimension and the time complexity are high, and then the facial expression recognition rate meets the requirement in the prior art.

Description

technical field [0001] The technical solution of the present invention relates to a method for recognizing image features or characteristics of graphics, specifically a method for recognizing dynamic human facial expressions. Background technique [0002] Facial expression is the most effective way in human emotional communication. With the development of computer technology, facial expression recognition has important applications in fields involving machine vision systems and pattern recognition, such as psychological research, video conferencing, intelligent Human-computer interaction, affective computing and the medical industry. With the comprehensive development of human-computer interaction technology, research on how to make computers automatically perceive human emotions is the focus of artificial intelligence. [0003] Early facial expression recognition methods focused on studying facial expression features in static images. However, facial expression recognitio...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00
Inventor 于明苗少栋王岩郭迎春刘依朱叶阎刚于洋师硕郝小可
Owner HEBEI UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com