Unlock instant, AI-driven research and patent intelligence for your innovation.

3D structured light face recognition method

A face recognition and structured light technology, applied in the field of 3D structured light face recognition, can solve the problem of low accuracy of single visual feature recognition

Active Publication Date: 2021-12-17
福建平潭瑞谦智能科技有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Provide a method of multi-modal fusion to improve the accuracy of lip language recognition, and solve the problem of low accuracy of single visual feature recognition in the prior art

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
  • 3D structured light face recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0048] Examples, see figure 1

[0049] Step 100: Project structured light to the upper part of the target user's body, and receive the information and audio information of the upper part of the body through the sensor; the projected structured light in step 100 adopts the forward projection method; the upper part of the body information in step 100 includes Color image data and depth data.

[0050] It should be further explained that this embodiment uses KinectV1 to project structured light and extract data; the principle is that the infrared emitter that comes with KinectV1 continuously radiates light to the target user through the grating, and projects random light on the upper half of the target user's body. speckle, and then use its built-in infrared camera to record the distribution of speckle in the space, and then realize the acquisition of depth data; at the same time, KinectV1 has its own RGB camera, which can simultaneously extract color image data; what needs to be...

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 3D structured light face recognition method, and belongs to the technical field of face recognition. The method comprises the following steps: projecting structured light, and receiving information through a sensor; performing information preprocessing to obtain information during action; voice recognition; and performing feature fusion processing according to the mouth information, the hand information and the audio data to realize face lip language recognition. According to the invention, fusion of sign language, voice and lip language modal information is realized through face recognition and voice of structured light, and a multi-element auxiliary fusion method for deaf, dumb and deaf-dumb people is provided; the problem of how to generate characteristics which better reflect the visual information of the speaker in the lip movement process is solved, and the recognition accuracy is further improved; the problem of visual ambiguity is solved through mapping from different modes to visual positions; and meanwhile, the multi-mode fusion also solves the problem that the result cannot be output when the single mode cannot be identified, so that the feasibility of the method in practical application is ensured to the greatest extent.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and in particular relates to a 3D structured light face recognition method. Background technique [0002] 3D structured light face recognition is based on infrared, adding an infrared dot matrix emitter. The output image includes an infrared IR image and a depth image. Both the depth image and the IR image are light sources using infrared light. , so it can be used in dark environments and is not easily disturbed by natural light environments. The live detection of face recognition with 3D structured light is through the introduction of a depth map, which uses two face images of IR and depth for 3D live detection. Moreover, on the basis of the attack algorithm of infrared face recognition defense, the depth map can be added to carry the depth information, which can effectively prevent planar attacks, such as photos, videos, paper masks and other material attacks, and can also be combine...

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): G06K9/00G06K9/34G06K9/62G06N3/04G06N3/08G06T17/20
CPCG06T17/20G06N3/08G06N3/045G06F18/253
Inventor 程超超王书琪许晓东
Owner 福建平潭瑞谦智能科技有限公司