Method for identifying human face subspace based on geometry preservation

A space recognition and subspace learning technology, applied in the field of face subspace recognition based on geometry preservation, can solve the problem of not being able to find the essential structure well, and achieve the effect of strong discriminating ability.

Inactive Publication Date: 2007-09-19
SHANGHAI JIAO TONG UNIV
View PDF0 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Neither linear discriminant analysis nor principal component analysis can well discover the underlying essential structure embedded in high-dimensional face image data.

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
  • Method for identifying human face subspace based on geometry preservation
  • Method for identifying human face subspace based on geometry preservation
  • Method for identifying human face subspace based on geometry preservation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] The embodiments of the present invention are described in detail below: this embodiment is implemented under the premise of the technical solution of the present invention, and detailed implementation methods and processes are provided, but the protection scope of the present invention is not limited to the following embodiments.

[0023] The embodiment adopts a public face database: YALE database. The YALE database contains 15 people with 11 images for each person. Each image varies in pose, expression, lighting, etc. First stack the rows (columns) of the face image into a one-dimensional vector set: X=[x 1 ,...,x N ]. For each person, randomly select p (=3, 5) samples for training, and the rest for recognition. For each given p, 20 sets of random training-recognition sample sets are generated, and the average recognition rate is calculated on this basis. Next, the dimensionality of the face image is reduced to N-1 by principal component analysis.

[0024] Find a...

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

A human face subspace identification method based on geometry storage, which is applied to the field of human face identification technique. The steps are that: stacking the row or column of the human face image into a one-dimensional sector set, using the principal component analysis for reducing dimension of the high dimensional human face image data in advance, applying the subspace study method on the training sample so as to obtain a low-dimensional subspace and a correspondent projection matrix, mapping the projection matrix of each test sample got from training to the low-dimensional subspace, conducting identification in the low-dimensional subspace by using an adjacent classifier. The invention trustily keeps the inter-class geometrical property and class-within geometrical property, conquers the defects in processing the non-linear problems of the traditional principle component analysis and linear discriminant analysis, moreover, the class information of the human face data is utilized fully, therefore the invention is suitable for solving the classification problem.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to a face subspace recognition method based on geometry preservation. Background technique [0002] Face recognition can be defined as identity verification from images of faces. Compared with other biometric methods, such as fingerprint recognition, iris recognition, palm recognition, etc., face recognition is more user-friendly and convenient. In recent years, face recognition has been applied to many fields, such as human-computer interaction, real-time monitoring, face-based video retrieval and so on. In the past 20 years, a large number of face recognition methods have emerged. Among these methods, the so-called appearance-based methods are the most commonly used and the most effective. The so-called appearance-based methods regard the face image as a whole pattern and operate on it directly. Generally speaking, a face image with a resolution of m×n pix...

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
Inventor 杨杰张田昊杜春华吴证袁泉
Owner SHANGHAI JIAO TONG 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