Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Training method of deep convolutional neural network for face recognition

A deep convolution and face recognition technology, applied in the training field of deep convolutional neural network, can solve problems such as difficult convergence, single action type, and difficult aggregation of feature points, so as to speed up convergence, improve degrees of freedom, and class-to-class more diffuse effects

Active Publication Date: 2020-11-24
SOUTH CHINA UNIV OF TECH +1
View PDF6 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the constraints generated by these loss functions have the problem of a single type of action, which cannot effectively and flexibly drive feature points to achieve a better distribution
In addition, there are still some feature points distributed near the spatial origin during training, and these feature points are difficult to gather to their respective class centers, which makes it difficult for the entire training process to converge

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
  • Training method of deep convolutional neural network for face recognition
  • Training method of deep convolutional neural network for face recognition
  • Training method of deep convolutional neural network for face recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The present invention will be further described below in conjunction with specific examples.

[0047] Such as figure 1 As shown, the training method of the deep convolutional neural network for face recognition provided by this example includes the following steps:

[0048] 1) Prepare the face image data set, including training set and verification set. Before training or verification, several preprocessing steps are required for the face image data in the dataset, including face detection and alignment, image pixel value normalization, image size normalization, image enhancement, and dataset cleaning, etc. In this example, the image pixel value is normalized to a 32-bit floating-point number between 0 and 1, and the image size is normalized to 112×112×3 (respectively representing the length, width and color channel number of the input image). The used Image enhancement methods include gray scale linear transformation, histogram equalization transformation, pixel colo...

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 training method of a deep convolutional neural network for face recognition. The method comprises the following steps: 1) preparing a face image data set, dividing the face image data set into a training set and a verification set, and selecting the type, the structure, the hyper-parameter and the magnitude of a deep convolutional neural network model according to the scale and the complexity of the training set and the performance index of face recognition which should be achieved; 2) extracting features of face pictures input by the training set by using the model,and taking the features as input in the step 3); 3) constructing a loss layer, and iteratively calculating a loss value for the training; 4) comparing the loss value calculated in the step 3) with a preset threshold value, judging whether training is stopped or the gradient is calculated, and updating model parameters; and 5) verifying the model performance, and determining whether to stop training. According to the method, the human face features can be constrained by using a multivariate acting force from two aspects of an Euclidean space and an angle space during training, so that the deepconvolutional neural network model can learn the human face features with higher discrimination and robustness.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a training method for a deep convolutional neural network for face recognition. Background technique [0002] Face recognition is a biometric technology for identification based on human facial image information. Compared with identification technologies such as fingerprints and pupils, it has significant advantages such as non-contact, low threshold for collecting information, and high recognition rate. When conducting face verification, since the collection process is non-contact, it can not only reduce the resistance of the person being collected, but also ensure the hygiene and safety of the collection process, especially for epidemic prevention inspections during the epidemic, which can effectively reduce the possibility of virus infection. In addition, face recognition has a wide range of application scenarios, such as building access control systems, monitoring s...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/172G06V40/168G06N3/045G06F18/2411Y02T10/40
Inventor 田联房孙峥峥杜启亮
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
Eureka Blog
Learn More
PatSnap group products