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

A personalized preprocessing method, system and terminal for face images

A face image and preprocessing technology, applied in the field of image processing, can solve the problems of inability to personalize processing and low recognition accuracy, and achieve the effects of clear layers, enhanced personalized features, and improved accuracy.

Active Publication Date: 2020-06-26
浙江智贝信息科技有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present invention provides a personalized preprocessing method for face images, aiming to solve the technical problems of low recognition accuracy and inability to perform personalized processing before 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
  • A personalized preprocessing method, system and terminal for face images
  • A personalized preprocessing method, system and terminal for face images
  • A personalized preprocessing method, system and terminal for face images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] figure 2 The flow chart of a personalized preprocessing method for face images provided by an embodiment of the present invention is shown, and is described in detail as follows:

[0037] In step S201, the grayscale image of the human face is analyzed, and the key points of the grayscale image of the human face are identified and extracted.

[0038] The embodiments of the present invention aim at including the converted face grayscale image, and perform preprocessing on the face area before image recognition. That is to say, when the embodiment of the present invention is actually used, the input data stream should be a grayscale image.

[0039] Wherein, the key points include 68 key points of the edge contours of the four regions of eyes, nose, mouth and chin. The embodiment of the present invention uses the dlib database to detect and locate 68 key points of the human face.

[0040] Of course, the user can adopt other key point detection methods such as the ASM (A...

Embodiment 2

[0056] image 3 It shows the process of using the face personalized feature model to fill the edge of the face gray image to obtain the face area image provided by the embodiment of the present invention, and the details are as follows:

[0057] In step S301, according to the key points in the grayscale image of the human face, the edge contour of the human face is extracted, and the grayscale image of the human face is divided into a human face area and a non-human face area.

[0058] In step S302, according to the feature ratio in the personalized face feature model, the non-face area in the gray-scale image of the face is filled according to the following formula to obtain the face area image:

[0059]

[0060] Among them, G i is the filling value of the i-th face grayscale image when performing grayscale filling, is the average gray value of the i-th face gray image.

[0061] The embodiment of the present invention aims at the technical problem that the features are...

Embodiment 3

[0063] Figure 4 It shows the process of normalizing the face region image according to the key points of the face region image provided by the embodiment of the present invention, and performing personalized alignment on the face region image, which is described in detail as follows :

[0064] In step S401, the pupil position of the face area in the face area image is identified and acquired, and an affine transformation is performed on the face area image so that the distance between the left eye pupil and the right eye pupil is a constant D.

[0065] Due to the instability of the distance and posture of the face in the real-time video stream, the embodiment of the present invention firstly performs normalized adjustment on the image of the face area, so that the shape of the face image is balanced and presents a better posture, which is convenient Learning of facial features in face recognition.

[0066] At this time, D can be learned through training, and can be a fixed ...

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 present invention is applicable to the technical field of image processing, and provides a personalized preprocessing method for human face images, including the following specific steps: identifying and extracting the key points of the grayscale image of the human face; establishing a personalized human face A feature model; using the personalized feature model of the human face to perform edge filling on the gray-scale image of the human face; normalizing the image of the human face area, and performing personalized alignment on the image of the human face area. In the embodiment of the present invention, the method of extracting the key points of the face is used to obtain the personalized features of the face image simply and quickly, and the key points are converted into personalized gray values ​​through a mathematical model, which effectively enhances the Personalized features in the gray-scale image of the face, and the personalized alignment method is also used to make the processed gray-scale image of the face clear and the face is located in the center of the image, especially suitable for the prediction of the face image in real-time video. process.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a personalized preprocessing method, system and terminal for face images. Background technique [0002] With the rapid growth of application needs in security access control and financial trade, biometric identification technology has received new attention, and face recognition is one of the most widely used technologies in all biometric identification methods. With the continuous evolution of technology, face recognition technology has been widely used in various fields such as public security, finance, network security, property management and attendance. [0003] In recent years, the recognition method based on real-time video is one of the main development directions of face recognition technology. Since the face area in the real-time video is always changing, that is, face recognition based on real-time video has slightly lower requirements for illumin...

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
CPCG06V40/161G06V40/168
Inventor 曹耀和
Owner 浙江智贝信息科技有限公司
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