Race classification based multi-feature gender judgment method

Abstract

A race classification based multi-feature gender judgment method comprises the following steps of (1) obtaining a face area through an AdaBoost face detection algorithm; (2) cutting the face area from binoculus positions; (3) establishing face and race classifier image training sets of the Mongolian, the Caucasian and the Negroes; (4) performing feature extraction on training set images; (5) training race classifiers through an SVM method; (6) combining face and eye characteristics to be served as gender judgment characteristics; (7) training gender classifiers of difference races through the SVM after characteristic vectors are extracted; (8) firstly judging the race to which a gender unknown face image belongs after face detection, preprocessing and characteristic extraction, judging the gender through the classifier which is corresponding to the race after the race is obtained. The race classification based multi-feature gender judgment method solves gender difference of different races and greatly improves the face gender judgment accuracy rate.

Description

technical field [0001] The invention relates to a gender judging method based on a human face race classification method, in particular to a multi-ethnic face gender judging method based on various features of a human face and using machine learning technology, and belongs to the technical field of image processing. technical background [0002] As the flow of people increases, security monitoring plays an increasingly important role in the safe development of cities, and video surveillance is also increasingly used in public places to ensure the safety of people's daily life. With the increase of application occasions, the market demand for intelligent video surveillance is also increasing. [0003] The human face contains a wealth of biological information. Human beings can identify the identity, gender, race, age, expression and other information of the individual through the face. With the development of computer vision technology, face recognition technology has gradua...

AI Technical Summary

Problems solved by technology

With the globalization of the economy, the development of transportation, and the increasing flow of people, the distribution of various races is not limited to a certain area. There are these three types of races in almost all regions of the world. There are also cases of mixed races among different races. In the case of mixed races, the facial features will still be more inclined to a certain race. Gender differences are also presented in different ways. For example, the gender differences of the Mongolians are mainly reflected in the eye area, while the gender differences of the Caucasians are more obvious, and the gender differences of the Negroes require the comparison of the details of the facial features. Therefore, if you simply Use all races for gender judgment, and the accuracy of gender judgment for some races will be affected

Images