Method for detecting human face shielding based on structure perception

A detection method and face occlusion technology, applied in the field of face occlusion detection and image processing, can solve the problems of low accuracy, poor performance and low efficiency.

Active Publication Date: 2016-07-20
ZHEJIANG UNIV OF TECH
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to overcome the disadvantages of low efficiency, poor performance, and low accuracy of existing face detection and recognition technologies when processing occluded face pictures, the present invention provides a structure-based Perceptual face occlusion detection method

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
  • Method for detecting human face shielding based on structure perception
  • Method for detecting human face shielding based on structure perception
  • Method for detecting human face shielding based on structure perception

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The method provided by the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0037] refer to Figure 1 ~ Figure 1 2, a kind of face occlusion detection method based on structure perception, described human face occlusion detection method comprises the following steps:

[0038] Step 1. Take the m×n-dimensional occluded face image y on the real number field as the face image to be detected, and select the set A={a of N face images 1 ,a 2 ,...,a N} as a training sample, here, a i is an m×n order face image collected under normal lighting conditions without occlusion in the real number field;

[0039] Step 2 Estimate the occlusion support s of the face image y to be detected based on the structural error measurement and clustering method in the identity transformation domain I( ), and obtain s I , here, I(·) is the identity transformation operator: I(y)=y, and the occlusion support s is defined as an m×n ...

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 provides a method for detecting human face shielding based on structure perception. Shielding can exist in various forms, but structural differences from normal human face image mainly lie in three aspects, such as color, direction and texture, and generally a certain difference is the most explicit. To highlight potential explicit structural differences, 6 characteristic transformation domains are proposed: an identical transformation domain, a logarithm transformation domain, a logarithm-gradient module transformation domain, a gradient direction transformation domain, a Laplace transformation domain, and a differential excitement transformation domain. Based on the 6 characteristic transformation domains, all kinds of structural errors between the shielding and a human face image are measured, and the structural errors are subject to two-value mean-value clustering and morphological filtering, so that 6 possible shielding results can be obtained. For an automatic determination of the best shielding detection result, a structure perception criterion is proposed: the minimum boundary regularization criterion. Experiments show that the method proposed in the invention can effectively detect the position of the shielding, and perceives explicit structures that are shielded, and has a strong practicality.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to the technical field of face occlusion detection, which is mainly used for detecting the position of occlusion in real occluded face images. Background technique [0002] With the rapid development of information technology, face recognition technology has been widely used in real life, such as: various ATM cash machines in banks monitor cashiers, and customs gates monitor passengers entering and leaving. For a reality-oriented face recognition system, a major difficulty is that face images are often occluded (such as sunglasses, masks, scarves, etc.). On the one hand, occlusion will lead to the loss of effective features in the face image; on the other hand, it will form local salient features, so that the existing face recognition system often makes misjudgments based on these local salient features, such as treating a black scarf as a black scarf. beard. Therefore, how...

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): G06K9/00
CPCG06V40/171G06V40/161
Inventor 李小薪李晶晶梁荣华冯远静刘义鹏
Owner ZHEJIANG UNIV OF 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