Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Human face age estimation method based on deep sparse representation

A technology of sparse representation and sparse representation coefficient, which is applied in the field of face age estimation based on deep sparse representation, which can solve the problem of lack of engineering application of massive age training samples, and achieve the effect of strengthening ability and reducing interference.

Active Publication Date: 2017-11-07
HUBEI UNIV OF SCI & TECH
View PDF9 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, its application faces problems such as the optimal design of the deep network and the optimal processing skills of face images; secondly, the parameters of the existing deep learning structure increase sharply when the number of hidden layers increases, which puts forward higher requirements for computing performance; finally, Facing problems such as lack of massive age training samples and engineering applications

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
  • Human face age estimation method based on deep sparse representation
  • Human face age estimation method based on deep sparse representation
  • Human face age estimation method based on deep sparse representation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0070] A face age estimation method based on deep sparse representation proposed by the present invention mainly includes four steps: dictionary training, deep sparse representation model construction, feature extraction and age estimation.

[0071] Step 1: Dictionary Training

[0072] This step can be done offline in the background. First, collect face samples from 0-80 years old, each age as a class, each class contains 500 face images. Accordingly, the composition age face training set A=[A 1 , A 2 ,L,A 80 ],in is the jth sample face in the i-th category. Then, perform AAM, BIF and Gabor+LBP feature extraction on the face training set A. Finally, the discriminative dictionary learning method is used to train a complete dictionary D based on AAM, BIF and Gabor+LBP AAM ,D BIF ,D GL .

[0073] Step 2: Deep sparse representation model construction

[0074] Use the complete dictionary obtained in the previous step to build a deep sparse representation model, as shown...

Embodiment 2

[0086] A method for estimating the age of a human face based on depth sparse representation is the same as embodiment 1, wherein the method for discriminating dictionary learning in step 1 comprises the following steps:

[0087] (1a) Collect face samples from 0-80 years old, each age as a class, each class contains 500 face images. Accordingly, the composition age face training set A=[A 1 , A 2 ,L,A 80 ],in is the feature vector of the j-th sample in the i-th class. Assume that the training sample set A is represented by a linear combination of a complete dictionary D, and its sparse representation coefficient matrix is ​​X. Then X can be rewritten as X=[X 1 , X 2 ,L,X 80 ], where X i for subset A i coefficient matrix.

[0088] (1b) In order to make the obtained complete dictionary D not only have good sparse reconstruction ability to sample set A, but also have strong discrimination and noise processing ability, the present invention constructs the following dictio...

Embodiment 3

[0109] A method for estimating the face age based on depth sparse representation is the same as embodiment 1-2, wherein the depth sparse representation model building method in step 2 includes the following steps:

[0110] (2a) The first layer design: Since the AAM feature integrates the face texture and shape information and is global, it is suitable for rough estimation of face age. Therefore, the first layer adopts AAM features. First, use the dictionary learning method introduced in step 1 to obtain the complete dictionary D AAM Perform sparse representation on the test face y to obtain the sparse representation coefficient x AAM :

[0111]

[0112] Among them, γ is a constant balance factor. Rewrite sparse representation coefficients Among them, the coefficient vector Corresponds to the sub-dictionary

[0113] Then, according to Define the residuals for each class:

[0114]

[0115] Among them, the first item is the reconstruction error item of the i-th...

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 provides a human face age estimation method based on a deep sparse representation, and belongs to the technical field of image processing and pattern recognition. The method solves the problem that an existing human face age estimation method is unstable. The method mainly comprises the following steps: A, building a distinguishing dictionary learning model; B, establishing a deep sparse representation model based on a distinguishing dictionary; C, building a two-factor analysis model, and removing an identity factor; D, extracting a robustness age characteristic; and E, building a stratified age estimation model, and performing age estimation. The method has the advantages of high anti-interference capability, high accuracy and the like.

Description

technical field [0001] The invention belongs to the technical field of image processing and pattern recognition, and relates to a face age estimation method based on deep sparse representation. Background technique [0002] In the medical field, people mainly determine a person's "biological age" by analyzing blood test indicators such as cholesterol, high-density cholesterol, and albumin, and use this to study the degree of human aging. Unfortunately, this index system is still far from perfect and inconvenient to use. If you can use computer and image processing technology to accurately predict a person's "biological age" by analyzing the apparent image of the face, and compare the "biological age" with the "actual age", you can know whether you are "permanent youth" Or "fading before old age". Then it will greatly improve the research efficiency and reduce the research cost. Estimating age by looking at the "face" can not only be used to quantify aging, but also can be...

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): G06K9/00
CPCG06V40/178G06V40/172
Inventor 廖海斌
Owner HUBEI UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com