Multi-pose face recognition method based on face mean and variance energy images

A face recognition and energy map technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem of no periodicity in the face energy map, so as to weaken noise interference, reduce complexity, and reduce feature dimensions. effect of numbers

Inactive Publication Date: 2013-07-24
HARBIN ENG UNIV
View PDF3 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, unlike the gait energy map, the face energy map has no periodicity, and it re

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-pose face recognition method based on face mean and variance energy images
  • Multi-pose face recognition method based on face mean and variance energy images
  • Multi-pose face recognition method based on face mean and variance energy images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] The present invention will be further described below in conjunction with accompanying drawing:

[0042]A multi-pose face recognition method based on face mean value and variance energy map. Firstly, it needs to read multi-pose face images from the face database, and perform face region detection on face images based on AdaBoost algorithm and manual segmentation method , and then construct the face mean energy map based on the face area image, and then construct the face variance energy map by combining the face area image and the face mean energy map, and finally combine the feature extraction method of the K-L transformation and the feature-level fusion strategy to convert the face mean The features are fused with the variance energy map to form a new feature vector, and the face recognition is completed through the nearest neighbor classification.

[0043] 1. Read multi-pose face images and face area detection

[0044] 1.1. Definition of Face Pose Changes

[0045] ...

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 relates to a biological feature identity recognition technique, in particular to a multi-pose face recognition method based on face mean and variance energy images. The method comprises the following steps of detecting face areas to carry out the size normalizing on the images of the face areas; building a narrow face mean energy image and a generalized face mean energy image; building a narrow face variance energy image and a generalized face variance energy image; combining obtained features, so as to obtain a final feature vector; and classifying and recognizing by a nearest neighbor classifier based on Euclidian distance. The method has the advantages that the storage space is well saved, the calculation complexity is reduced, and the noise interference in the single-frame image is weakened; and the face energy images contain face contour information under multiple poses, and have large advantage for recognizing the faces with large-angle pose change, the zero-padding processing is not needed, and the recognition property of the multi-pose face is improved.

Description

technical field [0001] The invention relates to a biological feature identification technology, in particular to a multi-pose face recognition method based on face mean value and variance energy graph. Background technique [0002] The existing solutions to the problem of pose change in face recognition mainly include three categories: three-dimensional recognition methods, pose correction methods, and manifold learning methods. The multi-pose face recognition based on the 3D model is to use the 2D face image to reconstruct the 3D virtual face, rotate the 3D face to form a 3D face database, and then perform 2D projection to construct a 2D multi-pose face database, so that the human Face recognition is converted into a comparison of two photos at a certain angle; or collect 3D faces to build a 3D pose face library, and use 3D faces for recognition. This method can achieve better recognition results for both frontal and quasi-frontal perspectives. , but this method requires s...

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): G06K9/00G06K9/62
Inventor 王科俊邹国锋唐墨付斌杜同春
Owner HARBIN ENG UNIV
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