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

LBP human face identification method of complex number wavelet transform domain

A face recognition and wavelet transform technology, applied in the field of face recognition, can solve problems such as unseen

Inactive Publication Date: 2018-01-12
YIBIN UNIV
View PDF7 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is no report on using Gabor and DTCWT two kinds of complex wavelets to decompose the local rectangular area of ​​the key points of the face, and extracting LBP features on the decomposed subbands.

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
  • LBP human face identification method of complex number wavelet transform domain
  • LBP human face identification method of complex number wavelet transform domain
  • LBP human face identification method of complex number wavelet transform domain

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0013] The technical scheme of the present invention is described in detail below in conjunction with accompanying drawing, concrete realization has the following steps (see figure 1 shown):

[0014] Step (1) Face image preprocessing. First, use Gamma correction to eliminate part of the illumination. Gamma correction is to replace the index value of the pixel with its own value, expressed as: I=I λ , λ is the correction factor, which is taken as 0.2 here; then select a histogram of a frontal image without the influence of illumination and noise as a reference, and further enhance the image with the histogram specification.

[0015] Step (2) Establish and determine the key point area: use the SDM method to extract 25 feature points on the image. For 25 points, a local rectangular area with a size of M×M pixels is established with each key point as the center, where M=32. These keypoint-based local rectangular regions are denoted as LR i ,i=1,...,25. See figure 2 shown. ...

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 relates to a human face identification technology and in particular to an LBP feature human face identification technology based on multiple complex number wavelet transform domains of akey point local rectangular region. A research proves that the identification accuracy rate can be improved by using multiple features. A Gabor wavelet and a DTCWT domain are two types of the different complex number wavelets through which the image local features are extracted by the different modes, so the expression ability to the image can be improved. The whole solution comprises the following steps: firstly, pre-processing the image, positioning a plurality of feature points on the pre-processed image by using SDM, and extracting the 'local rectangular region' on the feature points, anddecomposing as the feature images of the different directions and different scales by using the Gabor wavelet and DTCWT; secondly, performing the feature extraction on each Gabor sub-band and DTCWT sub-band by using LBP, and calculating an LBP normalized histogram (named as Gabor-DTCWT-LBP feature); and finally, performing the transform to the features by using principal component analysis (PCA)and linear discriminant analysis (LDA), and using for the human face identification. The identification speed of the method is rapid, and the effect of negative factors, such as different illumination, different facial expression, different face angles and human face aging, can be effectively prevented, and the identification effect is good.

Description

technical field [0001] The invention relates to face recognition technology, in particular to an LBP face recognition technology based on multiple complex wavelet transform domains of local rectangular areas of key points. Background technique [0002] In the era of intelligent information, computer-based face recognition is widely used, such as online transactions, automatic attendance and so on. When performing automatic face recognition, different lighting, different facial expressions, different facial angles, face aging and other factors are still difficult problems in face recognition technology. These unfavorable factors often exist together, making identification more difficult. [0003] The present invention proposes a face recognition technology based on feature point areas, which can effectively overcome the above-mentioned unfavorable factors. The local area based on feature points has two advantages: (1) better discrimination; (2) can effectively resist the in...

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
IPC IPC(8): G06K9/00
Inventor 李朝荣张永华许涧
Owner YIBIN UNIV