Image recognition method, device and system and computer storage medium

Abstract

The invention provides an image recognition method, device and system and a computer storage medium. The method includes: acquiring a to-be-identified image and a target image; executing a full-imagerecognition step, wherein the to-be-recognized image is compared with the target image; executing feature position comparison when a full-image comparison result does not meet a preset condition, wherein feature positions of the to-be-recognized image are compared with feature positions of the target image; and when a feature position comparison result meets a preset condition, outputting the comparison result obtained by feature position comparison. The method greatly improves extraction efficiency of valid information.

Description

technical field [0001] The present invention relates to the field of image recognition, and more particularly relates to an image recognition method, device, system and computer storage medium. Background technique [0002] At present, image recognition systems, especially face recognition systems, are increasingly used in scenarios that require identity verification in the security and financial fields, such as remote bank account opening, access control systems, and remote transaction operation verification. In these high-security application fields, in addition to ensuring that the face similarity of the person being verified conforms to the base library stored in the database, it is first necessary to verify that the person being verified is a legal living creature. In other words, the face recognition system needs to be able to prevent attackers from using photos, videos, 3D face models, or masks to attack. [0003] In image recognition, especially face recognition tec...

AI Technical Summary

Problems solved by technology

[0004] However, the face recognition in the above technologies is based on extracting features from a single face image, and then comparing based on the face features. The two images used for comparison do not interact when extracting features.

In this way, the importance of facial features cannot be determined according to the actual situation when extracting features. For example, one of the pictures is a person wearing a mask, and the other picture is a person not wearing a mask. The mouth features are extracted from the two pictures separately, then the comparison of the features will be meaningless, so the extraction efficiency of effective features will be reduced

Images