Facial expression recognition method based on complexity perception classification algorithm

A facial expression and classification algorithm technology, which is applied in the field of image recognition, can solve the problems of different facial feature distribution and feature complexity, and achieve the effect of alleviating the inconsistency of sample feature distribution, improving accuracy, and alleviating misclassification of easily confused expression categories

Inactive Publication Date: 2018-11-09
SOUTH CHINA UNIV OF TECH
View PDF4 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, in an uncontrolled environment, human faces are easily affected by factors such as race, age, gender, hair, and surrounding ...

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 based on complexity perception classification algorithm
  • Facial expression recognition method based on complexity perception classification algorithm
  • Facial expression recognition method based on complexity perception classification algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0028] This embodiment provides a facial expression recognition method based on a complexity-aware classification algorithm, the flow chart of the method is as follows figure 1 shown, including the following steps:

[0029] S1, the facial expression images from the facial expression data set Fer2013 are cut, whitened, normalized and preprocessed as a training data set;

[0030] S2. Design a deep convolutional neural network based on the improved residual block to train the training data set and extract facial features;

[0031] S3. According to the complexity-aware classification algorithm, by evaluating the complexity of the facial features extracted from the training data set, the training data set is divided into an easy training sample set and a difficult training sample set, and the two types of sub-sample sets are respectively Train an easy sample classifier and a hard sample classifier;

[0032] S4. Mark the {+} label on the easy training sample set and the {-} label ...

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 facial expression recognition method based on a complexity perception classification algorithm. Firstly, a deep convolution neural network based on improved residual blocks is designed to train pre-processed training data sets, and facial features of a person are extracted; according to the complexity perception classification algorithm, the complexity of the facial features of the person is evaluated, the training data sets are divided to an easy sample set and a difficult sample set, and the two classes of sub sample sets are trained respectively to obtain an easy sample classifier and a difficult sample classifier; in view of the two classes of sub sample sets, a binary-class sample complexity identification classifier is trained; and after the test data sets are subjected to preprocessing and facial feature extraction, the facial features of the person extracted by the test data sets are subjected to complexity identification through the sample complexityidentification classifier, the test data sets are inputted to the easy sample classifier and the difficult sample classifier to complete recognition on the class of a facial expression according to the complexity of the facial features of the person.

Description

technical field [0001] The invention relates to the field of image recognition in computer applications, in particular to a facial expression recognition method based on a complexity perception classification algorithm. Background technique [0002] Facial micro-expression recognition has extremely broad research prospects in the fields of human-computer interaction and affective computing, including polygraph detection, intelligent security, entertainment, Internet education, and smart medical care. Facial expressions are often used by humans as a main way to express emotions, so the main task of expression recognition is how to automatically, reliably, and efficiently recognize the information conveyed by human facial expressions. In the research of expression recognition, seven basic expression categories are defined: surprise, fear, disgust, anger, happiness, sadness, and peace. These seven categories are usually used as the basic labels of expression recognition. Most ...

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
CPCG06V40/172G06V40/168G06V40/174
Inventor 文贵华常天元
Owner SOUTH CHINA UNIV OF 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