Face quality evaluation method and device based on granular ball clustering

A quality assessment and clustering technology, applied in the field of face quality assessment of granule clustering, can solve problems in face recognition and affect the accuracy of face recognition, so as to avoid database redundancy and streamline face database Effect

Pending Publication Date: 2022-03-29
CHONGQING UNIV OF POSTS & TELECOMM
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a face quality assessment method and device based on granule clustering to solve the problem of face recognition caused by too many repeated face pictures in the database, thereby affecting the accuracy of face recognition

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
  • Face quality evaluation method and device based on granular ball clustering
  • Face quality evaluation method and device based on granular ball clustering
  • Face quality evaluation method and device based on granular ball clustering

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] Such as figure 1 As shown, the first embodiment provides a face quality assessment method of granule clustering, including,

[0040] Step S10, collecting a plurality of face pictures to be identified, performing feature extraction on the plurality of face pictures to be identified respectively, and obtaining a plurality of feature vectors;

[0041] Step S20, performing clustering training on multiple feature vectors until multiple clustering areas with feature vectors of the same category are obtained, and saving the results of the multi-clustering areas in multiple files;

[0042] Step S30, input the results of multiple files into the quality assessment model for quality assessment, and obtain the best face picture.

[0043]The face picture quality assessment method in this embodiment can be applied in the face database, and is used for evaluating the picture with the best quality of the same person in a database, and can be used as some follow-up face access control ...

Embodiment 2

[0059] Such as Figure 4 As shown, the present embodiment 2 provides a face quality assessment device based on granule clustering on the basis of the first embodiment, to perform the steps of the face quality assessment method mentioned in the first embodiment, and the device includes a feature extraction unit 220. Feature clustering unit 330 and quality evaluation unit 440;

[0060] The feature extraction unit 220 is used to collect a plurality of human face pictures to be identified, and perform feature extraction on a plurality of human face pictures to be identified respectively, to obtain a plurality of feature vectors;

[0061] A feature clustering unit 330, configured to perform clustering training on multiple feature vectors until multiple clustering areas with feature vectors of the same category are obtained, and save the results of the multi-clustering areas in multiple files;

[0062] The quality evaluation unit 440 is configured to input the results of multiple f...

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 face quality evaluation method and device based on particle and ball clustering, relates to the field of face image quality detection, and solves the problem that too many repeated face images in a database cause problems in face recognition and affect the accuracy of face recognition. Performing feature extraction on the plurality of to-be-recognized face pictures to obtain a plurality of feature vectors; performing clustering training on the plurality of feature vectors until a plurality of clustering regions with the same category of feature vectors are obtained, and storing results of the clustering regions in a plurality of files; and inputting results in the plurality of files into a quality evaluation model for quality evaluation to obtain an optimal face picture. According to the method, repeated and poor-quality face pictures in the face database are deleted, and database redundancy is avoided.

Description

technical field [0001] The invention relates to the field of human face image quality detection, more specifically, it relates to a human face quality evaluation method and device based on granule clustering. Background technique [0002] In recent years, face recognition technology has developed rapidly, and face recognition is used in many places, such as face recognition in class, work, payment and so on. The function of face recognition is mainly to identify a photo of a person's face and obtain his (her) identity information. For a machine, it cannot remember every detail of a person's face. Even after remembering, a person's face cannot always be the same. For example, expressions and other lighting will affect face recognition, and Not all pictures are very clear. For example, in a certain video, the person is profiled many times, so it is difficult to identify. The most important thing is that in a video, a person is in different pictures. It will appear many times...

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): G06V40/16G06N3/04G06V10/774G06V10/762G06V10/764
CPCG06N3/045G06F18/23G06F18/214
Inventor 夏书银夏岑竣张勇刘瑶
Owner CHONGQING UNIV OF POSTS & TELECOMM
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