Unlock instant, AI-driven research and patent intelligence for your innovation.

Pattern characteristic extraction method and device for the same

A technology of pattern features and extraction methods, applied in character and pattern recognition, instruments, computer parts, etc., can solve problems such as difficult to distinguish

Inactive Publication Date: 2007-12-05
NEC CORP
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

thus making it difficult to differentiate
[0026] In the prior art, there is a problem that cannot be avoided by principal component analysis and the technique of removing spaces with smaller eigenvalues ​​in the (total) covariance matrix

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
  • Pattern characteristic extraction method and device for the same
  • Pattern characteristic extraction method and device for the same
  • Pattern characteristic extraction method and device for the same

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0072] Embodiments of the present invention will now be described in detail with reference to the accompanying drawings. FIG. 1 is a block diagram showing a pattern feature extraction device according to the present invention.

[0073] The pattern feature extraction device will be described in detail below.

[0074] As shown in Figure 1, the pattern feature extraction device according to the present invention includes a first linear transformation device 11 for input feature vector x 1 Perform linear transformation; the second linear transformation device 12 is used to input feature vector x 2 performing linear transformation; and a third linear transformation means 13, configured to receive the feature vector transformed and dimensionally reduced by the linear transformation means 11 and 12, and perform linear transformation on it. Each linear transformation means performs basic transformation based on discriminant analysis by using discriminant matrices acquired through tr...

no. 2 example

[0099] According to the above, when different kinds of features such as direction features and density features are combined, discriminant analysis is repeatedly performed on the feature vectors for each of the feature vectors. However, multiple elements corresponding to a feature can be divided into multiple vectors, discriminant analysis can be performed for each element set as an input feature, and the corresponding projection vector further needs to be discriminant analysis.

[0100] A face image feature extraction device will be described in the second embodiment.

[0101] As shown in Figure 4, the face image feature extraction device according to the second invention includes an image feature deconstruction device 41, which is used to deconstruct the density feature of the input face image; a linear transformation device 42, which is used to The discriminant matrix is ​​used to project the feature vector; and the discriminant matrix group storage device 43 is used to sto...

no. 3 example

[0111] Another embodiment of the present invention will now be described in detail with reference to the accompanying drawings. FIG. 7 is a block diagram showing a face image matching system using a face metadata generating device according to the present invention.

[0112] The following describes the face image matching system in detail.

[0113] As shown in Figure 7, the face image matching system according to the present invention includes a face image input unit 71 for inputting a face image; a face metadata generation unit 72 for generating face metadata; Storage unit 73 is used to store the extracted face metadata; face similarity calculation unit 74 is used to calculate face similarity according to the face metadata; face image database 75 is used to store face images; The control unit 76 is used to control the input of images, the generation of metadata, the storage of metadata, and the calculation of the similarity of human faces according to the image storage reque...

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

An input pattern feature amount is decomposed into element vectors. For each of the feature vectors, a discriminant matrix obtained by discriminant analysis is prepared in advance. Each of the feature vectors is projected into a discriminant space defined by the discriminant matrix and the dimensions are compressed. According to the feature vector obtained, projection is performed again by the discriminant matrix to calculate the feature vector, thereby suppressing reduction of the feature amount effective for the discrimination and performing effective feature extraction.

Description

[0001] This application is a divisional application of the patent application with application number 03809032.5 submitted on October 22, 2004. Background technique [0002] Currently in the field of pattern recognition, by extracting feature vectors from the input pattern, extracting feature vectors effective for recognition from the feature vectors, and comparing the feature vectors obtained from each pattern, the relationship between patterns such as characters or faces can be determined. similarity between. [0003] For example, in the case of face verification, the pixel values ​​after normalizing the face image using the positions of human eyes etc. are converted into one-dimensional feature vectors after raster scanning, and by using the feature vectors as input Principal Component Analysis of Eigenvectors (Non-Patent Reference 1: "Probabilistic Visual Learning for Object Representation" by Moghaddam et al., IEEE Transactions on Pattern Analysis and Machine Intelligence...

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 龟井俊男
Owner NEC CORP