Improved sift face feature extraction method based on key points

A face feature and extraction method technology, applied in the field of face recognition, can solve the problem of not being able to accurately locate the key points of the face, and achieve the effect of reducing the number of dimensions

Active Publication Date: 2019-01-29
BEIJING UNIV OF CHEM TECH
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, when the SIFT method is used for face images, it cannot accurately locate the key points in the face, because it is mainly suitable for the recognition of general objects with high contrast, and there is a high similarity between face images

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
  • Improved sift face feature extraction method based on key points
  • Improved sift face feature extraction method based on key points
  • Improved sift face feature extraction method based on key points

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] The basic flow of the improved SIFT face feature extraction method based on key points of the present invention is as follows: figure 1 As shown, it specifically includes the following steps:

[0056] 1) Divide the data in the face database into 10 groups for cross-validation experiments, of which 9 groups of data are used as training data, and the remaining 1 group of data is used as test data. Each group contains 300 pairs of face pictures from the same person and 300 pairs of Pictures of faces from different people. For each face picture, use a three-layer deep convolutional network cascade to locate the coordinate positions of five key pixels (the pixel in the middle of the left eye, the pixel in the middle of the right eye, the pixel on the tip of the nose, the corner of the left mouth pixels and pixels at the corner of the right mouth).

[0057] In the first layer of the three-layer deep convolutional network cascade structure, three deep convolutional networks ...

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

Improved SIFT facial feature extraction method based on key points, this method adopts the improved SIFT facial feature extraction method based on key points. By locating five key pixels in the face and using the orientation histogram in the SIFT method to describe these five key points, a robust face image feature vector is formed. The similarity score between two face feature vectors is calculated by combining the bilinear similarity function and Mahalanobis distance. The KELM classifier is used to perform binary classification on the similarity score value. For a type of face image with a higher score value, both face images are judged to be from the same person, and a type with a lower score value Face pictures, both face pictures are judged to be from different people. In the process of face recognition based on the face feature vector, the bilinear similarity function and Mahalanobis distance are combined to calculate the similarity score of the two feature vectors, which enhances the distinguishability between classes.

Description

technical field [0001] The invention relates to an improved SIFT (Scale Invariant Feature Transform) face feature extraction method based on key points, which belongs to the field of face recognition. Background technique [0002] Face recognition is a biometric technology for identification based on human facial feature information. Compared with other biometrics, facial features have the advantages of naturalness, convenience, and non-contact, which make them have great application prospects in security monitoring, identity verification, and human-computer interaction. Therefore, face recognition technology is of great research value. Generally speaking, the face recognition process is divided into two processes: face feature extraction and face similarity score calculation. The face feature extraction process is to extract some key features of the face picture to form a face feature vector. The face similarity score calculation process is to calculate the similarity bet...

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
CPCG06V40/171G06V40/172
Inventor 李伟王璐冯复标
Owner BEIJING UNIV OF CHEM TECH
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