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

Emotion classification method and system based on facial blood flow distribution

A technology of emotion classification and blood flow, applied in blood flow measurement, sensors, computer components, etc., can solve problems such as emotion classification errors and achieve the effect of improving accuracy

Active Publication Date: 2020-06-09
TIANJIN POLYTECHNIC UNIV
View PDF12 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this classification method will vary due to individual differences. Some people are good at controlling their facial expressions or controlling their voice intonation to conceal their emotions at the time, so the above method may cause emotional classification errors

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
  • Emotion classification method and system based on facial blood flow distribution
  • Emotion classification method and system based on facial blood flow distribution
  • Emotion classification method and system based on facial blood flow distribution

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] Such as figure 1 As shown, the present invention provides a kind of emotion classification method based on facial blood flow distribution, and described emotion classification method comprises the steps:

[0063] Obtain facial video data to be classified; use a frame of facial video data as a facial image data;

[0064] Select the area on both sides of the alar of the face image data and the area of ​​the cheek as the ROI area;

[0065] Determine the optimal pulse wave for the ROI area.

[0066] The optimal pulse wave for determining the ROI region specifically includes: respectively adopting the rPPG algorithm to calculate the R channel, G channel, and B channel average pixel gray value of the ROI area as the R channel pulse wave, the G channel pulse wave, and the B channel pulse wave. detrend, band-pass filter and independent component analysis on the R channel pulse wave, G channel pulse wave and B channel pulse wave, respectively, to obtain the R channel independe...

Embodiment 2

[0076] In this embodiment, emotions are classified into two categories, and emotions are divided into positive emotions and negative emotions through facial blood flow distribution. The specific process is as follows: figure 1 shown.

[0077] 1. Collect facial color video and use video to stimulate emotional changes;

[0078] Step 1: Collect face video through the camera, place the camera in the center of the fill light, and evenly fill the face of the subject with light. When shooting, the face of the person under test is facing the camera, and the camera image optimization function is turned off to maintain the originality of the image. The distance between the face of the tested person and the camera is about 30 cm, and the tested person should try to avoid behaviors such as shaking the head during the test

[0079] Step 2: Before the emotional stimulation experiment starts, it is first necessary to reduce the influence from one's own emotional state. In order to achieve...

Embodiment 3

[0097] The present invention also provides an emotion classification system based on facial blood flow distribution, the emotion classification system comprising:

[0098] The facial video data acquisition module is used to acquire the facial video data to be classified; a frame of the facial video data is used as a facial image data.

[0099] The ROI area selection module is used to select the areas on both sides of the nose and cheeks of the facial image data as the ROI area.

[0100] The optimal pulse wave determination module in the ROI area is used to determine the optimal pulse wave in the ROI area.

[0101]The best pulse wave determination module in the ROI area specifically includes: an average pixel gray value calculation submodule, which is used to calculate the R channel, G channel, and B channel average pixel gray value of the ROI area using the rPPG algorithm, as R Channel pulse wave, G channel pulse wave, B channel pulse wave; independent component acquisition s...

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 provides an emotion classification method and system based on facial blood flow distribution. The method comprises the steps of obtaining to-be-classified facial video data; selecting areas of two sides of the nose wing and the cheek of the face image data as ROI areas; determining the optimal pulse wave of the ROI; extracting basic features from the time domain and frequency domaininformation of the optimal pulse wave of the ROI; extracting facial blood flow distribution from the facial image data by utilizing the optimal pulse wave of the ROI; sequentially arranging the basicfeatures and the facial blood flow distribution of each piece of facial image data to form a facial blood flow distribution feature sequence as input feature data; and inputting the trained SVM model,and carrying out emotion classification. According to the method, physiological information such as facial blood flow distribution characteristics and the like is applied in the emotion classification process, so that emotion classification errors caused by the fact that a classification object masks the current emotion due to the fact that the classification object controls the facial expressionor controls the voice intonation of the classification object are avoided, and the emotion classification accuracy is improved.

Description

technical field [0001] The invention relates to the field of emotion classification, in particular to an emotion classification method and system based on facial blood flow distribution. Background technique [0002] With the rapid development of artificial intelligence, it is crucial for machines to recognize and "own" emotions, and emotion classification has become a key factor in natural human-computer interaction. Scientists have done a lot of research on this, and a large part of them focus on the features of emotion classification based on people's facial expressions and voice information, and the classification effect is good. However, this classification method will vary due to individual differences. Some people are good at controlling their facial expressions or controlling their voice intonation to conceal their emotions at the time, so the above methods may cause emotional classification errors. Contents of the invention [0003] The purpose of the present inv...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/32G06K9/62G06K9/00A61B5/16A61B5/0205A61B5/026A61B5/00
CPCA61B5/165A61B5/0205A61B5/026A61B5/02405A61B5/7267A61B5/748G06V40/168G06V10/25G06F2218/10G06F2218/12G06F18/2411G06F18/214
Inventor 王慧泉何森梁晓韵陈瑞娟王金海
Owner TIANJIN POLYTECHNIC UNIV
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