Iris feature hiding method based on random noise mechanism

Abstract

The invention discloses an iris feature hiding method based on a random noise mechanism. The iris feature hiding method comprises the following specific steps of extracting and preprocessing iris features by using a wavelet-change-based zero-crossing detection and coding method; realizing iris biological feature hiding based on a differential privacy algorithm of a controllable noise adding mechanism; and verifying the similarity between the iris image after differential privacy protection and the original data based on a hash algorithm. The invention provides a privacy protection method basedon differential privacy measurement, so as to protect the privacy of iris images, improve the safety of iris image publishing and avoid the risk of privacy leakage. The hash algorithm-based iris image similarity verification method is used for comparing the similarity between an iris image to which a random noise mechanism is applied and an original iris image.

Description

Technical field: [0001] The invention relates to the field of iris image information security, in particular to an iris feature hiding method based on a random noise mechanism. Background technique: [0002] With the increasing popularity of the concept of the Internet in recent years, the security of personal information has become more and more important. Biometric technology is an important way to protect personal information by analyzing the inherent physiological or behavioral characteristics of human beings. The biological characteristics of the human body mainly include fingerprints, face, iris, voice, gait, signature, etc. The iris feature is the highest level of biometric authentication technology at present. The academic and business circles are paying more and more attention to the identification technology based on iris feature extraction. Iris recognition is gradually used in some departments that require high security performance, such as banking systems and ...

AI Technical Summary

Problems solved by technology

For example: relevant scholars have proposed a block cipher based on a two-dimensional chaotic cryptosystem, which uses a two-dimensional chaotic system, such as Baker mapping, standard mapping, etc., to achieve effective encryption of image data; however, iris recognition is mainly based on the similarity measurement Hamming distance, the time complexity of the program is high

Not only that, the above encryption method is limited to iris privacy protection in semi-honest mode, but in real life, there may be malicious attackers stealing information from the server or tampering with the server

In addition, in the current popular cloud concept, if the cloud is used to store the user's iris, the iris information may be compromised

Images