Human face automatic identifying method based on data flow shape

Abstract

This invention has provided an automatic identification method for human face, basing on the shape of data stream; it identifies the unknown human face sample by non-supervision or hemi-supervision method, and using the non-linear structure information of the data of human face. It will carry out the non-supervision identification by extracting the spectrum characters of human face, when any label information can't be gained. And it will carry out hemi-supervision identification for human face by the method to reset the labels among the linear neighbor, when only a part of information of human face label information can be gained.

Description

technical field

[0001] The invention relates to a two-dimensional human face automatic recognition method, in particular to a non-supervised and semi-supervised human face automatic recognition method based on human face data manifold. Background technique

[0002] Face recognition refers to predicting the identity of a new face through existing face data. In recent years, face recognition has become a major direction of biometric recognition technology due to its potential use in military, police and civilian applications. Compared with other biometrics, face recognition has many advantages such as proactiveness, non-invasiveness and user-friendliness. Most of today's mainstream face recognition technologies are based on subspace methods, but the accuracy of these methods is greatly limited due to the influence of illumination, posture, and expression changes. So far, the establishment of a robust face recognition The system is still a very difficult problem.

[0003] In...

Images