Open set palmprint recognition system and method based on weighted element metric learning

Abstract

The invention discloses an open set palmprint recognition system and method based on weighted meta-metric learning, which meet the real-time requirement of palmprint recognition application by applying edge calculation to a palmprint recognition community access control system. The community access control system comprises an edge server and a plurality of palmprint recognition access control terminals distributed at the entrance of a community, at present, a palmprint recognition method is mainly concentrated in a closed scene, the invention further provides a novel weight-based meta-metric learning method for non-contact open set palmprint recognition, a network model is trained at a time only by adopting a known category in the training process, so that the model has good generalization ability, the method can be directly applied to the recognition of the types which do not appear in the training set, the recognition model does not need to be updated and trained again, and the robustness and convenience of the recognition system are improved; the palmprint recognition process is non-contact operation, and the sanitation and safety problems in the using process are avoided.

Description

technical field [0001] The invention belongs to the technical field of community access control, and relates to an open-set palmprint recognition system and method based on weighted element metric learning. Background technique [0002] With the development of urbanization, people live in the community as the basic unit. For the sake of safety, identity verification is required for daily access to the community. Most of the existing community access control uses verification methods such as IC cards and passwords to verify the identity information of people entering and leaving. However, since IC cards are easy to lose and cannot be fixedly matched with personnel information, access control passwords are easy to forget and leak. People who are not in the community may enter the community by swiping IC cards or stealing access control passwords, which poses a huge challenge to community security and defense. [0003] At present, some smart access control uses human face for...

AI Technical Summary

Problems solved by technology

In a small-scale scenario, such as a home user scenario, the number of registered users generally does not exceed 50, and the traditional computing model will not have obvious disadvantages; but when it is replaced by a medium-scale or above commercial application, such as community management, In large-scale factories, enterprises, schools and other scenarios with a large number of users, the disadvantages of the traditional computing model will be particularly obvious, that is, the system will take a long time to identify and respond slowly, which will lead to poor efficiency of palmprint recognition access control.

[0007] Most of the traditional palmprint recognition access control is a stand-alone version. All the processes and calculations included in the palmprint recognition system are carried out in the palmprint recognition access control terminal. Limitations, the computing power of the palmprint recognition access control terminal often cannot meet the requirements, resulting in a long time-consuming process of palmprint recognition access control and slow response speed, which cannot meet the low-latency requirements of the access control system

Images