Distributed human face authentication information generation method, and distributed human face authentication method and apparatus

Abstract

The invention discloses a distributed human face authentication information generation method, and a distributed human face authentication method and apparatus. The authentication information generation method comprises the steps of extracting human face feature data from a target human face picture; generating authorized information for the human face feature data according to a preset policy; performing compression processing on the human face feature data and the authorized information according to maximum supported capacity of a two-dimensional code; inputting the compressed data to a two-dimensional code generator; and outputting the two-dimensional code bearing human face authentication information. According to the authentication information generation method, human face features obtained by scanning two-dimensional codes are compared with human face features extracted on site in an one-to-one manner when human face identification is required, and persons with the human face features higher than a threshold are identified as the same person, so that high-precision human face identification can be realized; and only one-to-one comparison is required, so that compared with a system for performing identification in a database with a large amount of human face data, the calculation amount required for performing identification calculation once is small; and in a distributed calculation mode, a network can be flexibly established.

Description

technical field [0001] The invention belongs to the technical field of data processing. Background technique [0002] At present, face recognition technology has made great progress. There are also quite a few types of face recognition systems and related equipment on the market. However, these systems or devices all have one thing in common, that is, they all need to store the sample information of the face to be recognized collected in advance in the memory of the entire system or distributed in different device memories to form a face feature When the database, system or device is running, it is necessary to obtain these sample information through network connection or direct access to local storage, so as to realize face recognition or face comparison and complete the face recognition function. [0003] In addition, with the development of machine learning technology, face recognition systems or devices with better recognition results at this stage mostly use deep neur...

AI Technical Summary

Problems solved by technology

If the database is stored locally, the cost of storage hardware will be increased. If the storage server is placed at the remote end and communicated through a network connection, it will not only increase the cost of storage hardware, but also increase the complexity of system deployment.

[0006] 2) It is difficult to collect a face database containing all targets

In this way, the deployment complexity and maintenance cost of the system are further increased

[0007] 3) Under the conditions of a large number of face recognition databases, it will be difficult to achieve fast recognition and high-precision recognition

[0008] 4) There is a legal and moral risk of infringing on privacy

Facial feature information is personal sensitive privacy information. The current system or device needs to store the face feature information of the person to be identified on the server or the local storage of the device. In this way, the implementation of the system cannot avoid personal privacy issues.

[0009] It is also these practical problems mentioned above that have brought more difficulties to the widespread use of current face recognition systems or devices.

Images