Multi-algorithm fused face recognition method

A face recognition and multi-algorithm technology, applied in the field of multi-algorithm fusion face recognition, can solve problems such as Gaussian distribution and normalization deviation that are difficult to achieve, and achieve the effects of easy update, fast fusion speed, and easy deployment

Pending Publication Date: 2019-12-03
THE FIRST RES INST OF MIN OF PUBLIC SECURITY +1
View PDF4 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In practical applications, the collection of data sets is limited by subjective and objective conditions, and it is difficult to achieve a Gaussian distribution, so there will be a large deviation during normalization

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-algorithm fused face recognition method
  • Multi-algorithm fused face recognition method
  • Multi-algorithm fused face recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0076] The flow process of the present invention is specifically stated below, as image 3 Shown:

[0077] 1. Build a test image library

[0078] There are 10,000 people in the registration set, one image for each person; 10,000 people in the probe set, one image for each person; everyone in the registration set has an image in the probe set;

[0079] 2. Integration process:

[0080] Step 201: The first face recognition algorithm extracts feature templates from the registration set to generate a registration set template set G 1 , then G 1 =1 1 ,G 2 1 ,...,G i 1 ,...G 10000 1 >, where 1≤i<10000;

[0081] Step 202: The first face recognition algorithm performs feature template extraction on the probe set to generate a probe set template set T 1 , then T 1 =1 1 , T 2 1 ,...,T i 1 ,... T 10000 1 >

[0082] Step 203: The probe set template T of the first face recognition algorithm 1 All templates with registration set G 1 The templates are cross-compared to ...

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 multi-algorithm fusion face recognition method, and the method comprises the following steps: constructing a unified benchmark test image library, wherein the benchmark testimage library construction method comprises a registration set and a probe set; based on the benchmark test image library, enabling each face recognition algorithm to be subjected to algorithm test; calculating the FAR and the FRR of each algorithm on the benchmark test image library; calculating a trust degree function of each algorithm; normalizing the comparison score of each algorithm; carrying out score fusion by adopting a credibility-based weighting mode; and outputting a fusion comparison result according to the fusion comparison score. On the basis of establishing a unified test benchmark library, performance test is carried out on a plurality of different face recognition algorithms, comparison similarity scores are converted into corresponding credibility, and the converted scores meet credibility orderliness, so that the converted scores have comparability.

Description

technical field [0001] The invention relates to the technical field of computer software development and programming, in particular to a multi-algorithm fusion face recognition method. Background technique [0002] With the advancement of the "Internet +" action plan and the national artificial intelligence strategy, face recognition technology, as a technical defense line of artificial intelligence, is developing rapidly and has urgent social needs. Widely used in security, criminal investigation and other fields. [0003] Although the accuracy of face recognition technology is high, it is still affected by conditions such as poor lighting, facial occlusion, large postures, and changing facial expressions, resulting in certain refusal and misrecognition, thus affecting the use effect. In order to increase the accuracy of the system, users usually choose multiple algorithms from different face recognition algorithm manufacturers to complement each other, and improve the sys...

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): G06K9/00G06K9/62
CPCG06V40/16G06F18/251
Inventor 田强李志远周卫东吴国英张治安邱旭华
Owner THE FIRST RES INST OF MIN OF PUBLIC SECURITY
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