Facial recognition methods, devices, electronic devices and media for children across ages

A face recognition and children's technology, applied in the field of image processing, can solve the problem of low recognition accuracy, achieve the effect of improving accuracy and reducing feature differences

Active Publication Date: 2022-07-08
广东红橙云大数据有限公司
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Existing facial recognition systems are less accurate as faces change with age

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 recognition methods, devices, electronic devices and media for children across ages
  • Facial recognition methods, devices, electronic devices and media for children across ages
  • Facial recognition methods, devices, electronic devices and media for children across ages

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0090] Since the facial features of children will change with age, after obtaining the facial image to be recognized that includes the face of the current child, you can use the stored facial features of each child in different age groups to analyze the current children's facial features in other age groups. The facial features are predicted to determine the current child identity information, and the specific method is described in detail below.

[0091] figure 1 This is a schematic flowchart of a method for facial recognition of children across ages provided by the embodiments of the present application. like figure 1 As shown, the method can include:

[0092] Step S110: Acquire the target age of the current child in the to-be-recognized facial image and the to-be-recognized facial features of the to-be-recognized facial image.

[0093] The head posture of the current child in the face image to be recognized satisfies the preset posture.

[0094] In the specific implemen...

Embodiment 2

[0116] For the feature prediction network applied in the children's cross-age face recognition method of Embodiment 1, the training process of the feature prediction network may include the following steps:

[0117] Step A: Collect training data.

[0118] The training data includes a first facial image identified by each child in a first age group and a second facial image identified by a corresponding child in a second age group. The first age group and the second age group are different, and the head state of the child in the first facial image and the second facial image satisfies the preset posture.

[0119] Step B, acquiring the facial features of the first facial image and the facial features of the second facial image.

[0120] Step C, based on the facial features of the first facial image in the first age segment and the corresponding number, calculate the average facial feature corresponding to the first age segment, and based on the facial feature and the correspond...

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 present application provides a method, device, electronic device and medium for children's cross-age facial recognition. After acquiring the target age of the current child in the facial image to be recognized and the facial features to be recognized in the facial image to be recognized, the method searches for the facial features of multiple child identifiers corresponding to the predicted age groups in the stored facial feature database, and the corresponding predicted age groups The average facial features corresponding to the target age group and the average facial features corresponding to the target age group; input the facial features corresponding to any child identification, the average facial features corresponding to the predicted age group and the average facial features corresponding to the target age group into the trained In the feature prediction network, the predicted facial features of the child identification in the target age range are obtained, and if the predicted facial features corresponding to the target child identification match the facial features to be recognized, the target child identification is determined as the child identification of the current child. The method improves the accuracy of facial recognition for children.

Description

technical field [0001] The present application relates to the technical field of image processing, and in particular, to a method, device, electronic device and medium for children's facial recognition across ages. Background technique [0002] At present, the accuracy of facial recognition systems on the market has achieved great success on adult photos, but the performance on children's photos is far from adult photos. Taking FRVT's face authentication rankings in January 2020 organized by the National Institute of Standards and Technology (NIST) as an example, the champion model in the unconstrained environment photo competition has a false match rate (FalseMatch Rate, FMR) <=0.00001 (ten 1 in 10,000), the False None-Match Rate (FNMR) was 3.01%, while in the non-binding environment of children’s photos, the false match rate (False Match Rate, FMR) <=0.01 (percentage). 1), its rejection match rate (False None-Match Rate, FNMR) was 34.22%. It can be seen that there ...

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): G06V40/16G06V10/44G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/04G06N3/08G06F18/241
Inventor 林伟辉
Owner 广东红橙云大数据有限公司
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