Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

High-performance human face recognition method and system

A face recognition, high-performance technology, applied in the field of high-performance face recognition methods and systems, can solve the problems of complex calculation, difficult to promote three-dimensional face recognition, low recognition rate, etc. The effect of face recognition performance

Active Publication Date: 2015-01-07
UNIV OF SCI & TECH OF CHINA
View PDF2 Cites 61 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the 3D face recognition technology is not mature enough, the 3D data is too large, the calculation is complicated, the recognition rate is low, and the 3D data acquisition equipment is expensive, and the 3D data acquisition conditions are limited. Therefore, it is difficult to promote 3D face recognition in practical applications.

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
  • High-performance human face recognition method and system
  • High-performance human face recognition method and system
  • High-performance human face recognition method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] figure 1 It is a flow chart of a high-performance face recognition method provided by Embodiment 1 of the present invention. Such as figure 1 As shown, the method mainly includes the following steps:

[0052] Step 11. Perform face detection based on multiple perspectives, and perform image normalization after determining the face region according to the face feature point positioning algorithm, and obtain normalized two-dimensional face images from multiple perspectives.

[0053] Specifically, this step includes: Face detection based on multiple perspectives: Grayscale the input color image and perform histogram equalization, and use the front, left and right face detectors to detect faces Detect and remove face detection results whose area is smaller than a predetermined value, and obtain a multi-view face image;

[0054] Face positioning and normalization: Based on the hybrid tree structure feature point model of HOG (Histogram of Oriented Gradient) feature, the fe...

Embodiment 2

[0064] In order to facilitate understanding of the present invention, below in conjunction with figure 2 The present invention is further introduced.

[0065] Such as figure 2 As shown, the present invention mainly includes three parts of the processing flow: 1) the processing flow for the input image; 2) the processing flow for the sample library; 3) the process of face recognition combining 1) and 2).

[0066] 1. The process of processing and recognizing the input image on the right mainly includes: face detection, face positioning and normalization, and feature extraction.

[0067] 1) Face detection is a multi-view face detection mainly includes: graying the image and performing histogram equalization to reduce the influence of too dark or too bright light, and then using Adaboost (a kind of Iterative Algorithm) face detection algorithm for face detection.

[0068] Considering that the Haar features of multi-pose faces are obviously different, so in the embodiment of t...

Embodiment 3

[0102] Figure 4 It is a schematic diagram of a high-performance face recognition system provided by Embodiment 3 of the present invention. Such as Figure 4 As shown, the system mainly includes:

[0103] The human face image acquisition module 41 is used for performing human face detection based on multiple perspectives, and performing image normalization after determining the human face region according to the facial feature point positioning algorithm, and obtaining normalized two-dimensional human faces of multiple perspectives image;

[0104] The final feature acquisition module 42 is used to extract the directional gradient histogram HOG feature and the local binary pattern LBP feature of the normalized two-dimensional face image of each viewing angle, and connect them to obtain the final feature;

[0105] The sample feature extraction and training module 43 is used to extract the HOG feature and LBP feature of the face image in the sample library, and adopts tree str...

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 high-performance human face recognition method and system. The high-performance human face recognition method includes the steps that human face detection is carried out based on multiple visual angles, and after a human face area is determined according to a human face feature point positioning algorithm, image normalization is carried to obtain a human face image at multiple visual angles after normalization; the extracted HOG feature of the normalized image is connected with a LBP feature, and the final feature is obtained; the HOG feature and the LBP feature of the human image in a sample base are extracted, trained and matched with the final feature, so that high-performance human face recognition is realized; the human face image in the sample base is a multi-visual-angle projection image of a three-dimensional human face model with illumination. By means of the high-performance human face recognition method and system, the human recognition performance is not affected by changes of the illuamition condition and the human face posture factor, the human recognition speed is increased, and calculation complexity is reduced.

Description

technical field [0001] The invention relates to the technical field of image processing and pattern recognition, in particular to a high-performance face recognition method and system. Background technique [0002] Face recognition technology is to identify the identity by analyzing the unique shape and position of facial organs. It is an important biometric technology and is widely used in bank monitoring, access control systems, entry-exit inspections, criminal investigations and suspect tracking, and venue camera surveillance. , information security, home entertainment and many other fields. [0003] At present, face recognition systems are divided into two-dimensional face recognition systems and three-dimensional face recognition systems according to the data processed. [0004] Among them, the method adopted by the two-dimensional face recognition system is relatively mature, and the eigenfaces (Eigenfaces) method proposed by Turk and Pentland earlier in 1991 has a be...

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): G06K9/00G06K9/66
CPCG06V40/16G06V40/168G06V10/758G06F18/2411
Inventor 董兰芳任乐乐
Owner UNIV OF SCI & TECH OF CHINA
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com