Facial expression recognition method, system, device and readable storage medium

A facial expression recognition and facial expression technology, applied in the field of computer vision, can solve the problems of learning facial expression features and not being able to guarantee the model, and achieve the effect of enhancing robustness and improving recognition accuracy

Active Publication Date: 2021-09-03
HANGZHOU XINHE SHENGSHI TECH
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] At present, there are other methods for recognizing facial expressions based on deep learning, but there are still some problems. Most of the methods cannot guarantee that the model can learn facial expression features that are not related to occlusion.

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
  • Facial expression recognition method, system, device and readable storage medium
  • Facial expression recognition method, system, device and readable storage medium
  • Facial expression recognition method, system, device and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0071] A facial expression recognition method, such as figure 1 shown, including the following steps:

[0072] S100. Obtain a facial expression image, and form an unoccluded facial expression image data set with the obtained facial expression image;

[0073] S200. Perform random occlusion processing on each facial expression image sample in the unoccluded facial expression image dataset to form an occluded facial expression image dataset;

[0074] S300. Mix the unoccluded facial expression image data set and the occluded facial expression image data set in equal proportions to obtain an image data set, and divide the image data set into a training data set and a verification data set in equal proportions;

[0075] S400. Build a masked facial expression recognition model, and train the masked facial expression recognition model based on the training data set. After the training is completed, use the verification data set as input to verify the training results to obtain a mask...

Embodiment 2

[0111] A facial expression recognition system, such as figure 2 As shown, it includes an image acquisition module 100, an occlusion processing module 200, a data set acquisition module 300, a model construction and training module 400 and a result acquisition module 500;

[0112] The image acquisition module 100 is used to acquire facial expression images, and the acquired facial expression images are formed into an unblocked facial expression image data set;

[0113] The occlusion processing module 200 is used to perform occlusion processing on each facial expression image sample in the unoccluded facial expression image data set to form a occluded facial expression image data set;

[0114] The data set acquisition module 300 is used to mix the unoccluded facial expression image data set and the occluded facial expression image data set in equal proportions to obtain an image data set, and divide the image data set into a training data set and a verification data set in equa...

Embodiment 3

[0144] A computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following method steps are implemented:

[0145] Obtain the facial expression image, and form the unoccluded facial expression image data set with the acquired facial expression image;

[0146] Carry out occlusion processing to each facial expression image sample in the unoccluded facial expression image data set to form a occluded facial expression image data set;

[0147] The unoccluded facial expression image data set and the occluded facial expression image data set are mixed in equal proportions to obtain an image data set, and the image data set is divided into a training data set and a verification data set in equal proportions;

[0148] Build an occluded facial expression recognition model, and train the occluded facial expression recognition model based on the training data set. After the training is c...

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 human facial expression recognition method, which comprises the following steps: obtaining a human facial expression image, forming an unoccluded human facial expression image data set from the acquired human facial expression image; The facial expression image samples are occluded to form a occluded facial expression image dataset; the unoccluded facial expression image dataset and the occluded facial expression image dataset are mixed in equal proportions to obtain an image dataset, and the image dataset is divided into equal proportions Training data set and verification data set; Construct the occluded facial expression recognition model, and train the occluded facial expression recognition model based on the training data set. After the training is completed, the verification data set is used as input to verify the training results, and the occluded facial expression recognition model is obtained. Facial expression recognition model; based on the occluded facial expression recognition model, facial expression recognition is performed on facial expression images with or without occluders to be detected, and facial expression images and expression recognition results are obtained.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a facial expression recognition method, system, device and readable storage medium. Background technique [0002] At present, facial expression recognition technology has broad application prospects and value in the fields of human-computer interaction, smart education, and auxiliary medical care, and has received extensive attention from experts in computer science, psychology, and pedagogy. The current facial expression recognition system can better recognize various facial expressions in natural scenes. However, facial expression images collected in actual scenes often have various occlusions, such as glasses, masks, scarves, hair, and Self-occlusion induced by human limb swing under spontaneous conditions. The size and shape of the occluder are rich and changeable, and may cover any position of the face. Facial expression recognition under occlusion cond...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/174G06V40/172G06F18/214
Inventor 孙国辉
Owner HANGZHOU XINHE SHENGSHI TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap