Face image feature extraction and comparison method based on deep learning

A picture feature, deep learning technology, applied in the information field, can solve the problem of face image feature interference, reducing the efficiency of recognition, etc.

Active Publication Date: 2016-06-15
CHINACCS INFORMATION IND
View PDF5 Cites 39 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there are still face images affected by various factors such as illumination, expression, posture, makeup, etc.,...

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 image feature extraction and comparison method based on deep learning
  • Face image feature extraction and comparison method based on deep learning
  • Face image feature extraction and comparison method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The present invention proposes a hybrid local feature extraction and comparison method based on deep learning. First, it provides an effective face region segmentation method, which can extract local stable features compared with the entire face. Secondly, it proposes a The new convolutional neural network model based on mixed local features, compared with traditional features such as LBP and HOG, can extract deep face feature information, and use the feature point positioning method to identify 60 different areas, sizes, and color spaces of the face. Perform feature extraction and compare the COS distance of the two faces. The present invention adopts the OFD large-scale oriental face database, which includes two fonts of viewpoint and illumination, and collected and arranged 33,669 face images of 1,247 people. Each person took 19 viewpoint images and 8 illumination images, among which Figure 6 Only 8 of the 19 viewpoint images and 8 lighting images are provided. Sin...

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 discloses a face image feature extraction and comparison method based on deep learning. The method comprises the steps: performing local modularization on the images in a face library; constructing a convolution neural network, and inputting image modules into the convolution neural network to train; obtaining the image modules after performing local modularization on the images in a face library, using the trained convolution neural network model to extract the features of each image module, and obtaining the feature vector corresponding to each image; after performing local modularization on the face images to be compared, utilizing the trained convolution neural network model to extract the features of each image module, and obtaining the feature vectors corresponding to the face images to be compared; and using the cosine similarity to compare the feature vectors corresponding to the face images to be compared with the feature vector corresponding to each image in order. For the face image feature extraction and comparison method based on deep learning, face comparison in the OFD eastern face library ranks top 5, and the accuracy is 95%.

Description

technical field [0001] The invention relates to the field of information technology, in particular to a method for extracting and comparing features of human face pictures based on deep learning. Background technique [0002] Face comparison is dedicated to identifying whether two faces belong to the same person. The difficulty lies in whether stable features can be obtained. Early commonly used face feature extraction algorithms such as: LBP, Gabor, haar, etc., but the accuracy rate is affected by environmental factors and there is a bottleneck. In recent years, more and more deep models, especially convolutional neural networks (CNN), have been used to extract deep visual features. The convolutional neural network model can obtain more abstract and essential visual features. However, there are still face images affected by various factors such as illumination, expression, posture, makeup, etc., which bring great interference to the extraction of facial image features, th...

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
CPCG06V40/171G06V40/172
Inventor 舒泓新蔡晓东李隆泽王爱华
Owner CHINACCS INFORMATION IND
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