Bi-visual-angle face identification method based on multi-local correlation characteristic learning

Abstract

The invention discloses a bi-visual-angle face identification method based on multi-local correlation characteristic learning. Multiple local information in correlation characteristic learning is used to better know real non-linear structures among data, so that the face identification accuracy is improved. The realization process includes the steps of determining a plurality of local blocks of each training sample, constructing a uniform optimization framework of related characteristic learning and multi-local fusion, solving the correlation projection direction and multi-local fusion coefficient through alternative iteration, conducting feature extraction and characteristic fusion for training and testing samples, and using the nearest neighbor classifier for identification. Compared with the prior art, the bi-visual-angle face identification method is more effective and robust.

Description

technical field [0001] The invention relates to the fields of pattern recognition, image processing, etc., and in particular to a dual-view face recognition method based on multi-local correlation feature learning. Background technique [0002] Face recognition is a kind of biometric identification, which has the characteristics of non-contact collection, concealable collection, long-distance transmission, fast tracking and network monitoring. It has more advantages than other biometric recognition such as fingerprint recognition, iris recognition and gene recognition. Face recognition has a wide range of application values. It can be applied to customs monitoring, criminal tracking, file management, customer identification, human-computer interaction system certification, job inspection, multimedia data retrieval, information security, machine vision, virtual reality technology, etc. , has significant economic value and social benefits. [0003] Human faces have rich expre...

AI Technical Summary

Problems solved by technology

However, all these CCA methods based on local neighbor information only consider a neighbor relationship of samples

In the practical application of face recognition, the actual collected face data often has a complex nonlinear structure, and it is difficult to reflect the real nonlinear structure of the data only by relying on a neighbor relationship, which in turn affects the accuracy of face recognition.

Images