Method for quickly recognizing face

A face recognition and face technology, applied in the field of face recognition, can solve problems such as the inability to guarantee the real-time requirements of face recognition, and achieve the effect of meeting real-time requirements, improving the speed of face recognition, and reducing the amount of calculation.

Inactive Publication Date: 2018-06-01
SHANDONG UNIV
View PDF10 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in this patent, there is a method of calculating similarity scores one by one during the face recognition comparison process. Each picture to be tested needs to be compared with all pictures in the target 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 quickly recognizing face
  • Method for quickly recognizing face
  • Method for quickly recognizing face

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] A fast face recognition method, such as image 3 shown, including:

[0044] A. Perform steps (1)-(3) for each target picture in the training set to establish a target library

[0045] (1) Extract the key feature points of the face, and construct a standard feature vector with size, rotation and displacement invariance according to the key feature points of the face;

[0046](2) select several components in the standardized feature vector as reference variables, divide into blocks, and carry out binary coding to the blocks;

[0047] (3) Perform PCA dimension reduction processing on each block in step (2) to form several dimension-reduced blocks, and use the encoding of the block in step (2) to name several corresponding blocks after dimension reduction ;

[0048] The amount of calculation for face recognition can be roughly described by a simple example. Suppose there are 1000 face pictures of 100*100 in the target database. We want to judge whether the face captured ...

Embodiment 2

[0055] A kind of fast face recognition method according to embodiment 1, its difference is that,

[0056] Described step (1), comprises:

[0057] a. Extract the key feature points of the face. The key feature points of the face include the corners of the eyes, the tip of the nose, the two corners of the mouth, and the roots of the ears, a total of 9 key feature points;

[0058] b. According to 9 key feature points, 10 distance feature values ​​are formed, including: left eye width d1, vertical distance between the tip of the nose and the line between the eyes d2, the distance between the two ears and the root of the ear d3, the width of the mouth d4, the width of the right eye d5, two The horizontal distance d6 from the outside of the eye, the distance between the outer corner of the right eye and the tip of the nose d7, the distance between the inner corner of the left eye and the tip of the nose d8, the vertical distance between the midpoint of the mouth and the tip of the n...

Embodiment 3

[0062] A kind of fast face recognition method according to embodiment 2, its difference is that,

[0063] Described step (2), such as figure 2 As shown, the setting selects m components in the standardized feature vector as reference variables, 1≤m≤10, including:

[0064] d. Select 1≤i≤m, obtain the standardized feature values ​​corresponding to the face pictures of all people obtained in step (1) The normalized feature values ​​corresponding to each person's face picture is a fixed value;

[0065] e. Calculate the standardized eigenvalues ​​corresponding to the face pictures of all people median value of

[0066] f. According to the median value calculated in step e, perform binary classification on the standard feature vector of all people obtained in step (1), and perform binary coding at the same time, if in the standard feature vector If it is greater than or equal to the median, it is coded as 1; otherwise, it is coded as 0.

[0067] in turn for each Perfo...

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 quickly recognizing a face, and the method comprises the steps: (1), extracting face key feature points, and constructing a standard feature vector; (2), selecting a plurality of features in the standard feature vector as reference variables, carrying out the segmenting, and carrying out the binary coding; (3), carrying out the PCA dimension reduction processing of each block, forming a plurality of blocks after dimension reduction, and naming a plurality of corresponding blocks through the codes of the blocks; (4), reading a to-be-detected face image, extracting the to-be-detected face key feature points, and constructing a standard feature vector; (5), carrying out the binary coding of the standard feature vector, and searching the corresponding blocs according to the codes after dimension reduction; (6), carrying out the PCA dimension reduction processing of the to-be-detected face image; (7), determining the similarity between the to-be-detected face image and the images of each type in the blocks: determining that the to-be-detected face image and the image of the type belong to the same person if the similarity exceeds a set threshold value, and completing the recognition process.

Description

technical field [0001] The invention relates to a fast face recognition method, which belongs to the technical field of face recognition. Background technique [0002] Face recognition technology is a kind of biometric recognition technology, which is widely used in security, finance, telecommunications, transportation and other fields. The related technology uses the common features of the face to extract the face picture from the video stream, and then compares the face image with the template image stored in the face database in advance, and uses the difference between the face features to determine the face to be tested. identity information. [0003] The face recognition process can be roughly divided into three steps: face acquisition and positioning, image preprocessing, and face recognition. Among them, face recognition is the most critical step. Feature-based recognition algorithms include methods based on PCA, methods based on LDA, Based on LBP methods, methods b...

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): G06K9/62G06K9/00
CPCG06V40/165G06V40/171G06V40/172G06F18/2135G06F18/22
Inventor 王洪君胡才胜申大雪王娜
Owner SHANDONG 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