A Fingerprint Template Encryption Method Based on Double Rotation Feature Descriptor

Abstract

The invention belongs to the technical field of fingerprint recognition and encryption fusion in pattern recognition, and discloses a fingerprint template encryption method based on double-rotation feature descriptors, which expands the data of the minutiae descriptors in combination with the frequency field information construction of fingerprint minutiae points, and improves The matching number of detail point pairs. Not only can the accuracy of fingerprint matching be improved, but also a longer key can be combined to improve security strength. The double-rotation feature descriptor of the present invention can realize the non-compensation-accurate encrypted domain fingerprint identification problem, avoid traditional errors caused by inaccurate registration, and improve the matching accuracy of the fingerprint template encryption technology. The new ridge line counting of the present invention adds elastic deformation δ on the traditional basis, allowing the ridge line of the fingerprint to have a certain deformation range, which can effectively prevent the ridge line counting error caused by the elastic deformation of the fingerprint, and overcomes the fingerprint elasticity. The influence of deformation on the result.

Description

technical field [0001] The invention belongs to the technical field of fingerprint recognition and encryption fusion in pattern recognition, and in particular relates to a fingerprint template encryption method based on double rotation feature descriptors. Background technique [0002] At present, the existing technologies commonly used in the industry are as follows: Biometric encryption is a hot research field that has emerged in recent years. It aims to organically combine the existing biometric identification and cryptography technologies, give full play to their respective advantages, reduce the pressure on key management, and provide stronger security protection and protection for sensitive information. control. Fingerprint features are inherent features of the human body, unique and invariant for life. Once the user fingerprint template in the identification system is lost, it will be permanent. Moreover, fingerprint feature information also includes personal geneti...

AI Technical Summary

Problems solved by technology

This kind of method is difficult to prevent the management personnel from guarding themselves and stealing the user key and encrypted template at the same time to decipher the user's fingerprint biometric information

[0003] In summary, the problems in the prior art are: The ability to protect the user's original biometric features is slightly lacking, and it is difficult to prevent the administrator from stealing the user's key and encrypted template at the same time.

By extracting a double-rotation feature descriptor with stronger anti-deformation ability, the modified descriptor not only has rotation-translation invariance (that is, accurate recognition without compensation) but also transforms and reconstructs fingerprint features without directly using the original feature information of fingerprints.

Images