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

Method for improving accuracy of face identification

A face recognition, correct rate technology, applied in the field of image processing, to achieve the effect of improving the correct rate

Inactive Publication Date: 2013-04-03
SHANGHAI DIGIVISION
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, since this matching algorithm only considers SIFT features, it can be found that this matching still has a certain degree of blindness in actual use.

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 improving accuracy of face identification
  • Method for improving accuracy of face identification
  • Method for improving accuracy of face identification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] According to the idea of ​​the embodiment of the present invention, an algorithm that can quantify whether a feature point is in the face area of ​​a person is proposed, and use this to improve the result of face recognition.

[0034] The face recognition method in the embodiment of the present invention is based on a matching algorithm, and at the same time, the spatial positions of the feature points are jointly considered. It can be understood that the two feature points for effective matching should be in the same part of the face. Therefore, on the basis of conforming to the closest Euclidean distance of the feature vector, it must also be ensured that the two matching feature points need to be in the same part of the face.

[0035] In an embodiment, the human face area can be divided into areas such as eyes, nose, mouth, cheeks, and forehead. Calculate the matching points at the corresponding positions in the corresponding areas in the corresponding face images. ...

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 method for improving accuracy of face identification, which comprises the following steps: a), acquiring a face part from a reference image as a target matching image; b), calculating the positions of three central points of the target matching image; c), calculating the scale-invariant feature transform (SIFT) feature point set of the target matching image; d), calculating the positions of three central points and SIFT feature set of an image to be matched; e), finding the matching points of the image to be matched and the target matching image by using an SIFT matching algorithm; and f), judging if each pair of matching points is positioned in the same area of a face, and if each pair of matching points is positioned in the same area of the face, receiving thepair of matching points, or rejecting the pair of matching points if each pair of matching points is not positioned in the same area of the face.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method for improving the accuracy of face recognition. Background technique [0002] The SIFT algorithm was proposed by D.G.Lowe in 1999 and perfected and summarized in 2004. The algorithm obtains the basic feature parameters by extracting local features, finding extreme points in the scale space, and extracting invariants such as position, scale, and rotation. The SIFT feature parameter is the local feature of the image, which remains unchanged for rotation and scaling, and also maintains a certain degree of stability for brightness changes, viewing angle changes, affine transformations, and noise. [0003] At present, some people have successfully used SIFT features for face recognition. The core of this type of method is the SIFT feature matching algorithm, which is to consider the feature points of one of the images one by one after obtaining the SIFT feature points in 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): G06K9/00
Inventor 孙宏明
Owner SHANGHAI DIGIVISION