Self associative memory face identification method based on cellular neural network

Abstract

The invention discloses a self associative memory face identification method based on a cellular neural network. The method comprises the steps of S1, randomly acquiring m face pictures, and performing numbering; S2, through setting a binary picture brightness threshold, respectively processing the m face pictures in the step S1 for obtaining binary face pictures, and obtaining a self associative memory input matrix and a self associative memory output matrix which are composed of the binary face pictures; S3, establishing a cellular neural network face picture identification model frame which comprises unknown model parameters; S4, by means of the unknown model parameters, determining a final cellular neural network face picture identification model; and S5, based on a self associative memory principle, performing identification matching on the random face picture. The self associative memory face identification method has beneficial effects of digitalized storage, high safety coefficient, high identification efficiency and high credibility of a matching result.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to a face recognition method based on cellular neural network auto-associative memory. Background technique [0002] With the development of the big data era, people have identity information checks during travel, such as face recognition. Through identity recognition, identity verification is realized, the security performance of the system is improved, and different user identity information is confirmed. [0003] When checking the face information, it must include the face information saved in the database and the face information waiting for verification. For the face information saved in the database, in the prior art, there are the following defects: [0004] First: face information is often directly stored, which makes identity information easy to leak and has a low safety factor. Once leaked, it is easy to be copied and has poor reliability. [0005] Second: In t...

AI Technical Summary

Problems solved by technology

[0004] First: Face information is often stored directly, which makes identity information easy to leak and has a low safety factor. Once leaked, it is easy to be copied and has poor reliability.

[0005] Second: In the process of face image recognition, it is necessary to retrieve and compare a large amount of image data in the database, and it takes a long time to retrieve images, resulting in low recognition efficiency and prone to errors

[0006] Third: In the process of image recognition, when similar images are found, there is no image matching, the reliability is poor, and verification errors are more likely to occur

Images