Identity verification method based on face recognition

Abstract

The embodiment of the invention discloses an identity verification method based on face recognition. The method comprises the steps that an identity verification device collects a face picture of a user through a camera, whether the collected face picture meets a preset face picture requirement or not is detected, and if yes, the collected face picture meeting the face picture requirement is uploaded to an identity verification server; The identity authentication server obtains a picture of the certificate with the reticulate pattern of the user according to the user identifier of the user; The identity authentication server inputs the certificate photo with the reticulate pattern of the user into a pre-trained reticulate pattern removal neural network model to obtain a certificate photo with the reticulate pattern removed; And the identity verification server compares the face picture uploaded by the identity verification equipment with the netted certificate picture and calculates amatching degree, and if the matching degree between the face picture and the netted certificate picture reaches a preset threshold, the identity verification of the user is confirmed to be passed. According to the invention, the passing rate of identity verification is improved.

Description

technical field [0001] The invention relates to the communication field, in particular to an identity verification method based on face recognition. Background technique [0002] At present, many industries, such as the financial industry, have relatively strict requirements on the authenticity, security and credibility of customers, and the identity verification technology is also constantly improving. It is possible to use the ID card to take a photo of the user's face and the user's profile captured by the camera. Face photos are compared for identity verification, but the pass rate of this comparison method is generally low, because the ID photos called face photos are textured, which will affect the recognition, and when the user is shooting in real time , there may be phenomena such as the face photo taken is not a frontal photo of the face, and the complete facial features are not exposed, and there are also phenomena such as mistakenly pressing, misaligning the posit...

AI Technical Summary

Problems solved by technology

[0002] At present, many industries, such as the financial industry, have relatively strict requirements on the authenticity, security and credibility of customers, and the identity verification technology is also constantly improving. It is possible to use the ID card to take a photo of the user's face and the user's profile captured by the camera. Face photos are compared for identity verification, but the pass rate of this comparison method is generally low, because the ID photos called face photos are textured, which will affect the recognition, and when the user is shooting in real time , there may be phenomena such as the face photo taken is not a frontal photo of the face, and the complete facial features are not exposed, and there are also phenomena such as mistakenly pressing, misaligning the position, etc. to take pictures and upload them, and there will be errors when clicking to submit photos manually. Or blurred photos, etc. In this way, the key features of the face cannot be captured well, resulting in the inconsistency between the face photo taken and the face photo in the ID photo, making the user need to repeat the operation to pass the identity verification

Images