Human image verification method and device

Abstract

One or more embodiments of the invention disclose a portrait verification method and device, which are used for realizing open type portrait storage and management so as to improve the efficiency andaccuracy of portrait verification. The method comprises the following steps: in response to a portrait verification instruction for carrying out portrait verification on a user, obtaining a first portrait comparison record which corresponds to user information and meets a preset condition from a portrait management system according to the user information of the user contained in the portrait verification instruction; Determining a first reference portrait corresponding to the user according to the first portrait comparison record; Comparing the first reference portrait with a currently collected portrait to be verified of the user to obtain a comparison result; And determining whether the to-be-verified portrait passes verification according to the comparison result, Wherein the portraitmanagement system stores portrait comparison records of a plurality of users.

Description

technical field [0001] This specification relates to the technical field of face recognition, and in particular to a method and device for face verification. Background technique [0002] With the increasing maturity of face recognition technology, face, as an important biological feature of personal identity, has a wide range of applications in finance, travel, security and other fields. Internet products such as face login, face payment, face authentication, and face authorization are also becoming more and more popular. Among them, the face comparison algorithm has been very mature after long-term and multi-scenario verification, including some open source algorithms. However, there is no effective solution for the management of the portrait basemap that the face comparison algorithm relies on, especially when the base number exceeds 100 million or the time span exceeds ten years, the portrait basemap is even more difficult to manage. [0003] Among the multiple portrai...

AI Technical Summary

Problems solved by technology

This leads to capacity and performance problems when the service organization has high concurrency. In addition, the profile of the public security network Figure 1 Generally, it is only updated when the ID card is updated, and the portrait information entered during registration may vary greatly with the growth of natural persons and the comparison with the current living portrait; or, due to registration conditions, the quality of the entered portrait is low. not even entered

These factors may lead to users being unable to use the organization's portrait recognition service, and thus unable to use other products, and even have security risks such as fraudulent use and misidentification

Moreover, the portrait server mechanism can only obtain updated portraits through polling access, which has defects in efficiency and cost.

Images