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

Face recognition method and device based on multispectral fusion

A face recognition and multi-spectral technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem of low accuracy and achieve the effect of reducing difficulty and speeding up speed and efficiency

Inactive Publication Date: 2018-05-29
BEIJING EASY AI TECHNOLOGY CO LTD
View PDF3 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to fuse multiple spectra, overcome the shortcomings of visible light image acquisition through infrared image acquisition, solve the problem of low accuracy of existing face recognition in dark light and occlusion situations, and provide a human face recognition system based on multi-spectral fusion. Face recognition method and device

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 recognition method and device based on multispectral fusion
  • Face recognition method and device based on multispectral fusion
  • Face recognition method and device based on multispectral fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Such as figure 1 Shown, a kind of face recognition method based on multispectral fusion is characterized in that, comprises the following steps:

[0044] S1: collecting RGB images and infrared images;

[0045] S2: preprocessing the collected RGB image and infrared image;

[0046] S3: registering the preprocessed RGB image and the infrared image;

[0047] S4: Fusing the registered RGB image and the infrared image;

[0048] S5: Cut out the face from the fused image, and align the face images by detecting key points;

[0049] S6: Extract facial features from the aligned face images.

[0050] The local features of the facial image are extracted through the LBP (Local Binary Pattern) descriptor, which is robust to feature invariance, and the dimensionality of the extracted features is reduced. Finally, the features are compared with the face image repository to complete. face recognition.

Embodiment 2

[0052] On the basis of Example 1, the S1 step specifically includes:

[0053] S201: Perform grayscale image processing on the RGB image, and correct the infrared image;

[0054] S202: Perform WF de-illumination processing on the gray-scale processed RGB image, and perform histogram equalization processing on the corrected infrared image.

[0055] WF refers to the method of extracting light-invariant features from Weber face, which can effectively remove complex light effects.

[0056]

[0057] where A={-1,0,1}

[0058] The histogram equalization processing can improve the quality of the infrared image and highlight the features of the image.

Embodiment 3

[0060] On the basis of Example 1, the S3 step specifically includes:

[0061] S301: Obtain feature vectors of RGB image and infrared image feature points by feature extraction;

[0062] S302: Perform feature matching on the same-named points of the RGB image and the feature vector of the infrared image;

[0063] S303: Concatenate the same-named points to construct a transformation model, and obtain a registered image through affine transformation.

[0064] Even if the distance between the infrared camera and the ordinary camera is very close, the captured images still have a certain parallax. Before the fusion of the two images, the two images need to be registered and converted to the same viewing angle.

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 recognition method and device based on multispectral fusion. The method comprises the steps that collected RGB images and infrared images are preprocessed; the preprocessed RGB images and infrared images are registered; image fusion of the preprocessed RGB images and infrared images is conducted; human faces are cut out from the images on which fusion is conducted, and the human faces are aligned though detection key points; facial feature extraction is conducted on aligned human faces. According to the method, fusion is conducted on infrared image collection andvisible light image collection, advantages of the infrared image collection and the visible light image collection are combined, the influence of the visible light image collection under the conditions of dark light and blockage is prevented, meanwhile the problems that the infrared image collection is easily affected by environment temperature, and glass products such as glasses are hard to penetrate, so that black shadows are formed to influence image collection quality are solved, and face recognition accuracy under special conditions such as dark light, blockage and the like is improved.

Description

technical field [0001] The present invention relates to the technical field of face recognition, in particular to a multispectral-based face recognition method and device. Background technique [0002] Face recognition has been widely used in fields such as finance, medical care, and public security. With the development of technology, the existing face recognition has higher accuracy. However, under some unrestricted conditions, such as: dark light, occlusion, The accuracy rate of existing face recognition is relatively low. [0003] Face recognition is based on image acquisition. Visible light images widely exist in our lives and are the images we use most. However, with changes in visible light, facial features will change dramatically. Illumination and non-uniform lighting conditions will seriously affect face recognition. In addition, for criminals who need to be arrested by public security organs to solve crimes, they often adopt camouflage and occlusion methods to av...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/171G06V10/467G06F18/251
Inventor 徐枫陈建武肖谋
Owner BEIJING EASY AI TECHNOLOGY CO LTD
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