Gender identification method based on face shape and generalized self-organizing mapping

A self-organizing mapping and gender recognition technology, applied in the field of face recognition, can solve problems such as the adverse effects of changes in facial posture, expression, and lighting conditions on recognition accuracy, and achieve the effect of improving recognition rate and accuracy.

Inactive Publication Date: 2012-09-19
CHANGSHU RES INSTITUE OF NANJING UNIV OF SCI & TECH +1
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to restore the 3D face shape from a single 2D face image, so as to solve the adverse effects on recognition accuracy caused by chan

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
  • Gender identification method based on face shape and generalized self-organizing mapping
  • Gender identification method based on face shape and generalized self-organizing mapping
  • Gender identification method based on face shape and generalized self-organizing mapping

Examples

Experimental program
Comparison scheme
Effect test

example

[0081] Example: Based on the MAX PLANCK 3D library, using PGA, take the first 60 eigenvectors with the largest eigenvalues ​​to create a 3D face shape restoration model; the output layer neuron configuration of G-SOM is 30 20 regular hexagonal grid; follow 8 8 to block the NeedleMap in MAX PLANCK 3D and input it into G-SOM for training until convergence.

[0082] To verify the effect, we are using a subset of FERET, which contains 200 images of 200 people (103 males, 97 females, one image per person). Each image in the data set is preprocessed: geometric operations such as shearing, rotation, scaling, and alignment are used to obtain a final resolution of 142 124 images, and perform histogram equalization on the images. Then, use such as figure 1 The recognition process shown: For a given 2D face image, after recovering the 3D shape, use G-SOM to recognize the result. Experimental results show that the method used in this patent has achieved a recognition accuracy of 90...

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 gender identification method based on a face shape and generalized self-organizing mapping. The method comprises the following steps of firstly establishing a face three-dimensional (3D) shaped recovery model by pin grid array (PGA) technology based on an international standard MAXPLANCK3D face database; secondly, studying the 3D face library by growing self-organizing map (G-SOM), obtaining the distribution rules of 3D characteristics of face gender; at the identification stage, recovering a 3D shape information of a 2D face image on a 3D shape recovering model by firstly using a PGASFS technology, and then inputting the 3D shape information obtained by recovering to the G-SOM, and performing gender identification by using a Soft-KNN algorithm. The method provided by the invention has the advantages that the adverse effect to identification precision caused by changes of face gesture, expression and illumination conditions is reduced; secondly, the mode intrinsic topology can be kept better by the use of G-SOM; and thirdly, the links between the local structure and overall structure of the face is utilized by the gender identification method, thus, the recognition rate is improved.

Description

technical field [0001] The invention relates to technology in the field of face recognition, in particular to a method for face gender recognition based on recovering 3D face shape information from a 2D face image. Background technique [0002] With the rapid development of human economy, society and life, the demand for fast and effective automatic identity verification is increasingly urgent. However, although common forms of identity verification such as magnetic cards, IC cards, and identification numbers are mature and can be protected by various encryption methods, they cannot prevent forgery. Biometrics is the intrinsic biological attribute of human beings, which has strong self-stability and individual differences. Therefore, the use of biometrics for identification has gradually become a research and development hotspot, and face-based biometrics recognition technology is one of them. Compared with other biometrics (iris, palmprint, voiceprint, etc.), face-based bi...

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): G06K9/00G06T17/00
Inventor 於东军戚湧唐振民杨静宇
Owner CHANGSHU RES INSTITUE OF NANJING UNIV OF SCI & 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
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