Unsupervised human face intelligent accurate recognition method and system

Abstract

The invention discloses an unsupervised human face intelligent accurate recognition method and system. The method comprises the steps of preliminarily classifying extracted human face picture eigenvectors by adopting a density clustering algorithm, and performing model parameter learning on a training set by adopting a logistic regression algorithm, thereby obtaining initialized logic regression parameters; performing logical regression prediction processing on the human face picture eigenvectors in the test set according to the initialized logic regression parameters, and performing probability normalization calculation on obtained predicted values to obtain corresponding probability values; and when the probability values are greater than a preset confidence rate threshold value, allocating the human face picture eigenvectors currently subjected to the logical regression prediction processing to corresponding tags, and forming a new training set. According to the method, the accuracyof human face picture recognition is ensured while the clustering capability of the system is enhanced; and the method is based on an unsupervised classification algorithm, so that the manual taggingis avoided, the manpower and material resources are saved, and the processing speed of human face picture recognition is increased.

Description

technical field [0001] The invention relates to the field of face recognition, in particular to an unsupervised intelligent and accurate face recognition method and system. Background technique [0002] With the vigorous development of video storage, big data analysis and other fields in recent years, face recognition has not only required the ability to detect faces in photos, but also to be able to accurately find photos of the same person from multiple photos. And can be applied to various smart devices. [0003] Nowadays, the commonly used face recognition method is to manually label the face and store it in the database. When the camera equipment captures the face again, compare the newly taken photo with the photo stored in the database, and store it in the database. Find the photo with the highest similarity in the library, and use the tag of the photo as the tag of the newly added photo. [0004] In practical applications, it is common to encounter unsupervised sit...

AI Technical Summary

Problems solved by technology

[0006] In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide an unsupervised intelligent and accurate face recognition method and system, aiming to solve the problems of poor recognition accuracy and slow processing speed of existing unsupervised face recognition methods

Images