Front-end portrait encryption and recognition method for biological characteristic privacy protection

Abstract

The invention discloses a front-end portrait encryption and recognition method for biological characteristic privacy protection. The method comprises the steps: portrait data collection: photographinga video stream with portrait biological characteristics through a front-end camera; portrait preprocessing: attenuating the image noise of different intensities to different extents by using an adaptive fractional order integral algorithm to realize the adaptive denoising effect of the image, and detecting the dynamic portrait position of the video by using a portrait positioning algorithm; portrait encryption: gray scrambling and diffusion are performed on portraits by using a pseudo-random sequence generated by a chaotic system, so that encrypted portraits can be obtained, and an encryptedportrait library can be established; and encrypted portrait recognition: taking the encrypted portrait library as a training set to train a portrait recognition model, and directly recognizing the encrypted to-be-detected portrait. According to the invention, image encryption is applied to portrait recognition, and an encrypted portrait recognition method is utilized, so that the risk of privacy leakage in the decryption process is avoided, and the problem of personal privacy leakage caused when a portrait recognition product is used is avoided.

Description

technical field [0001] The invention relates to the field of deep learning application technology, in particular to a front-end portrait encryption and recognition method for biometric privacy protection. Background technique [0002] With its security and convenience, biometric recognition has been widely used in the field of identity authentication. Identification based on biometrics can solve the problems of insecurity and inconvenience in traditional identification. Among biometric features such as fingerprints, faces, palm prints, iris, retina, voice, and gait, the face is the most widely used for identification because of its high versatility, uniqueness, permanence, availability, and acceptability. One of the broadest biometrics. In recent years, face recognition has achieved very significant research results, and the recognition rate and recognition speed have been greatly improved. [0003] The rise of a new generation of artificial intelligence, while bringing s...

AI Technical Summary

Problems solved by technology

[0004]The controversy about the privacy protection of portrait recognition is extremely controversial and has become a pain point in the development of the industry. From a technical perspective, the methodology of data privacy protection is not mysterious, such as k-anonymity, l-diversity, t-closeness, adding noise to the model training, etc., but technical difficulties often arise in practical applications, such as adding noise and disturbance, the originally usable original information is submerged by the disturbance information, resulting in data failure

Images