A method and system for face detection with polarized light source

A face detection and light source technology, applied in the field of image processing, can solve the problems of inaccurate detection of polarized light source and waste of calculation amount.

Active Publication Date: 2021-04-30
深圳市梦网视讯有限公司
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiments of the present invention is to propose a face detection method with a polarized light source, which aims to solve the problems of inaccurate detection or waste of calculation under the polarized light source 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
  • A method and system for face detection with polarized light source
  • A method and system for face detection with polarized light source
  • A method and system for face detection with polarized light source

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] figure 1 This is a flow chart of a polarized light source face detection method according to a preferred embodiment of the present invention; the method includes:

[0036] Step1: Perform skin color block detection on the blocks in the image that are marked in the detection set Ω, and obtain skin color blocks;

[0037] Specifically: perform skin color block detection on the blocks in the detection set Ω, when a skin color block is detected, set the block as the current skin color block, and then enter Step 2; otherwise, determine that there is no human face in the image, and end;

[0038] Wherein, the detection method of the skin color block can use the block-based skin color detection method disclosed in the industry;

[0039] Among them, Step0 also includes Step0 before Step1: delete the first block line and the last two block lines of the current image, and divide all remaining blocks into the detection set Ω;

[0040] Step2: Perform outer adjacent skin color block ...

Embodiment 2

[0071] Figure 4 It is a structural diagram of a polarized light source face detection system according to a preferred embodiment of the present invention; the system includes:

[0072] The skin color block acquisition module is used to detect the skin color block for the block in the image that is drawn into the detection set Ω, and obtain the skin color block;

[0073] Wherein, the detection method of the skin color block can use the block-based skin color detection method disclosed in the industry;

[0074] Described acquiring the skin color block is specifically: carry out skin color block detection to the block in the detection set Ω, when detecting the skin color block, then set the block as the current skin color block, and then enter the outer adjacent skin color block detection and face acquisition device; otherwise , then it is determined that there is no face in the image, and the process ends.

[0075] The outer adjacent skin color block detection and face acquis...

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 proposes a face detection method and system with a polarized light source. The method of the present invention first adopts conventional face detection to the image, and then further subdivides processing based on the detection results whether it belongs to the side face detection or the face missed detection under the polarized light source, and finally, for the face missed detection under the polarized light source, a special Detect processing mode. At the same time, under the biased light source, the positional relationship between the light source and the face to be detected is used to perform brightness correction and recognition. In this way, it can not only ensure the frontal and side face detection under the forward light source, but also solve the problem of missing face detection under the polarized light source. Face detection based on the positional relationship between the light source and the face to be detected can further reduce the amount of detection calculations for multi-person images.

Description

technical field [0001] The invention relates to the technical field of image processing, and in particular, to a method and system for detecting a face with a polarized light source. Background technique [0002] In practical applications, lighting conditions, such as backlighting, polarized light sources, etc., will have a huge impact on the accuracy of face detection. Among them, under the polarized light source, the face image will appear areas that deviate from the normal range of skin color brightness, so that only partial faces under the smooth light source are detected. Simply expanding the distribution range of skin color brightness values ​​can compensate for some undetected skin color areas, but it will also lead to false detection of non-skin color areas, and it cannot fundamentally solve the problem of skin color detection under unfavorable light source conditions. In the face detection of multi-person images, if the relationship between the light source and the...

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
CPCG06V40/161
Inventor 舒倩
Owner 深圳市梦网视讯有限公司
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