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

A face detection classifier construction method based on adaboost

A classifier and weak classifier technology, applied in the field of face detection, can solve problems such as insufficient information utilization

Active Publication Date: 2019-03-05
中科天网(广东)科技有限公司 +1
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0014] The above-mentioned adjustment of sample weights based on AdaBoost face detection technology in learning and training only uses the weights of the previous time, and the information utilization is not enough

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 face detection classifier construction method based on adaboost
  • A face detection classifier construction method based on adaboost
  • A face detection classifier construction method based on adaboost

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0028] As shown in the figure, the face detection classifier construction method based on AdaBoost of the present embodiment is made up of following steps:

[0029] A: Obtain face samples and non-face samples: In the surveillance video, intercept images containing faces that are bounded by the outer sides of the ears on the left and right, and that are bounded by the chin and hairline on the top and bottom respectively, and non-human images that do not contain faces Face image; set the face image as a positive sample, and the non-face image as a negative sample; the union of positive samples and negative samples constitutes a training data set, expressed as A={(x 1 ,y 1 )……(x n ,y n )}, where n is an integer greater than 1, x i is a face image or a non-face image, if x i is a positive sample, then y i = 1; if x i is a negative sample, then y i = 0;

[0030] B: Preprocessing of all samples: adjust the face image and non-face image to a predetermined size, and use histog...

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 purpose of the present invention is to propose a face detection classifier construction method based on AdaBoost, which can improve the training efficiency of the face detection classifier, and based on the statistics of the training samples, the contribution of each organ of the face to the final detection is calculated. The ability to improve the accuracy of face detection, so as to obtain a face detection classifier with better classification effect. The present invention consists of the following steps: A: Obtaining human face samples and non-human face samples; B: Preprocessing all samples; C: Manually marking the area of ​​each organ; D: Statistics of the appearance of each organ on the face; E: Utilizing improved AdaBoost algorithm training to obtain weak classifiers for face detection; F: A series of weak classifiers are obtained through step E, and strong classifiers are obtained by combining these weak classifiers with the weights of facial organs. The present invention trains a series of weak classifiers by continuously changing the weight of samples, and finally combines a series of weak classifiers to form a strong classifier capable of distinguishing human faces from non-human faces.

Description

technical field [0001] The invention belongs to the technical field of face detection, and in particular relates to a method for constructing a face detection classifier based on AdaBoost. Background technique [0002] Face detection is one of the most closely integrated technologies in the field of computer vision and plays a vital role in face recognition. At present, the main face detection methods can be divided into three categories: face detection based on skin color regions, face detection based on templates, and methods based on statistical learning. The invention is a face detection method based on statistical learning. [0003] Face detection based on statistical learning is to use statistical analysis and machine learning methods to find the features that distinguish between human faces and non-human faces, and then build a classifier based on the features obtained by automatic machine learning to determine whether there is a human face in the image. The methods...

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): G06K9/00G06K9/62
CPCG06V40/172G06F18/2148G06F18/2431
Inventor 李建明李鑫温峻峰杜海江詹心泉薛柯王俊舒林凯泳陈斌
Owner 中科天网(广东)科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com