Face description, recognition method and device

A descriptor and facial image technology, applied in character and pattern recognition, computer components, instruments, etc., can solve the problems of AFR descriptor extraction complexity, achieve the effect of improving recognition or identification performance and reducing extraction complexity

Inactive Publication Date: 2010-04-21
MITSUBISHI ELECTRIC CORP
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0030] As mentioned earlier, the extraction of AFR descriptors is very complicated even when only the Four-domain features (Fourier part) of the descriptors are used.

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
  • Face description, recognition method and device
  • Face description, recognition method and device
  • Face description, recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] Table 1 explains the block symbols in the figure. The basic components are: Fourier transform (T, F); eigenvector transform (discriminant transform) (D) in the form of a linear map (specifically Dual-LDA); normalization (N) and digital transformation (Q). As seen in the block diagram of the structure, there are only 6 input images compared to the 42 images used for feature extraction in the Fourier domain part of AFR: f 1 , f 1，1 , f 1，2 , f 2 , f 2，1 , f 2，2 . figure 2 These images used in the preferred embodiment are shown in and described below. Note that the proposed scheme uses a 3-level discriminative structure, and Fourier facial features are extracted from the overall and central image regions combined in the final level DLDA map.

[0053] Although only at the initial discriminative level ( Figure 9 ) shows normalization, but normalization can also be performed in subsequent discriminative stages.

[0054] Table 1 - Block Symbols

[0055]

[0056]...

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 face recognition method, apparatus and system uses an architecture comprising multiple dual Linear Discriminant Analysis (LDA) transformations applied to coefficients of a frequency-based transformation applied to the facial image and its components. This new method significantly reduces extraction complexity compared to prior art and also offers increased robustness to face pose, illumination conditions and other factors, thus improving the recognition or authentication performance.

Description

technical field [0001] The present invention relates to methods and apparatus for facial description and recognition. The present invention also relates to methods and apparatus for representing images, especially faces in facial images, and the use of such representations. Background technique [0002] Probably the first article describing face recognition in the Fourier domain was "Face Recognition in Fourier Space" (published in Vision Interface '2000:38-44, Montreal, 2000). This article describes a facial recognition system that uses facial descriptions based on Fourier spectra computed from normalized facial images. Facial verification or recognition is accomplished by finding the closest match between feature vectors containing Fourier coefficients at selected frequencies. A subset of 27 frequency coefficients was reported to yield good results (using 15 real and 12 imaginary coefficients). The selection of these coefficients is based on the relative variance of the ...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06K9/00281G06V40/171
Inventor 米罗斯瓦夫·博贝尔克日什托夫·库哈尔斯基瓦迪斯瓦夫·斯卡尔贝克
Owner MITSUBISHI ELECTRIC CORP
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