Multi-angle facial expression recognition method under natural state

A natural state and expression recognition technology, applied in the field of face recognition, can solve the problems of lack of robustness and practicability, extraction, and inability to adjust features, and achieve the effect of fast computing speed, high recognition rate, and real-time computing.

Inactive Publication Date: 2018-11-16
SHANGHAI UNIVERSITY OF ELECTRIC POWER
View PDF2 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this type of algorithm needs to manually mark feature points, cannot adjust fe

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
  • Multi-angle facial expression recognition method under natural state
  • Multi-angle facial expression recognition method under natural state
  • Multi-angle facial expression recognition method under natural state

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] 1. Facial expression data preprocessing

[0021] Input multi-angle facial expression image data, and preprocess the image data, such as figure 1 , 2 , 3 is a schematic diagram of the preprocessing of the established facial expression data.

[0022] (1) Face detection and cropping: use the Multi-Task Convolutional Neural Network (MTCNN) method to detect the face and roughly locate the key points, and then crop the face area, such as figure 1 Schematic diagram of multi-view face detection and cropping shown. MTCNN is a neural network with a deep level (24 layers) multi-task framework. First, adjust the size of the image to different scales to build an image pyramid, and use it as the input of the three-level cascade framework. The cascade structure of the three-level deep convolutional network as follows:

[0023] P-Net (Proposal Network): Generate a candidate face frame and its bounding box regression vector, then use the bounding box regression vector to calibrate t...

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 relates to a multi-angle facial expression recognition method under a natural state. The method comprises the following steps: inputting multi-angle facial expression image data, and carrying out pretreatment on the image data; establishing an MVFE-LightNet network structure; after pretreatment, inputting the image data to an input layer, and after passing through two two-dimensionalconvolution layers, extracting image low-level edge features; then, carrying out image deep feature extraction and processing through 6 residual-depth-separable convolution layers, one two-dimensional convolution layer and one GlobalAveragePooling2D layer in sequence; and finally, sending the extracted image deep features to a Softmax classifier for training and recognition, and realizing classification output. The method is suitable for multi-angle facial expression recognition; the method is fast in operation speed and can realize real-time operation; the method is high in recognition ratein different angles of facial expression recognition; and the method effectively solves the overfitting problem due to increase of network layer number.

Description

technical field [0001] The invention relates to a face recognition technology, in particular to a multi-angle face expression recognition method in a natural state based on an MVFE-LightNet network. Background technique [0002] Facial expression recognition is one of the research hotspots in the field of computer vision and pattern recognition. It is the development trend of human-computer interaction and affective computing technology research. It has been widely used in intelligent security, robot manufacturing, medical treatment, communication and automobile fields. People's emotions are largely presented in facial expressions, and people's inner thoughts can be judged through changes in expressions. With the development of artificial intelligence, human beings' demand for intelligent and comfortable life is increasing. The analysis of human emotional information obtained through facial expressions is not only of scientific research value, but also of great significance ...

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): G06K9/00G06K9/62G06N3/04
CPCG06V40/175G06V40/168G06N3/045G06F18/2431
Inventor 邵洁钱勇生
Owner SHANGHAI UNIVERSITY OF ELECTRIC POWER
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