Encrypted face recognition method based on Ridgelet-DCT and Tent-Henon double chaos

Abstract

The invention discloses an encrypted face recognition method based on Ridgelet-DCT and Tent-Henon double chaos. The method comprises the following steps: generating a first encryption key and a second encryption key by using a double-chaos model, performing Ridgelet transformation on a face image, and performing primary encryption on the face image in a Ridgelet transformation domain and the first encryption key; carrying out DCT on the encrypted image and performing Hadamard product operation on the encrypted image and a second encryption key for secondary encryption, and acquiring an encrypted face image after DCT inverse transformation; finally, extracting encrypted face image features in combination with a PCA algorithm, and training a neural network model to complete encryption recognition of the face image. The geometric attack resistance and the shielding attack resistance of the face image are improved, so that the method has high robustness.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to an encrypted face recognition method based on Ridgelet-DCT transformation and Tent-Henon double chaos. Background technique [0002] Face recognition technology refers to computer technology that uses analysis and comparison of facial feature information for identity identification. In recent years, face recognition technology has become a hot spot in artificial intelligence research, and face recognition products have been widely used in finance, security inspection, medical care and other fields. While face recognition technology brings a series of conveniences, there are also huge security risks. During face recognition, a large amount of original face data is collected and stored and transmitted in the cloud, which may easily lead to the theft, leakage or tampering of face images and personal related information by others, causing certain losses and impacts on indiv...

AI Technical Summary

Problems solved by technology

While face recognition technology brings a series of conveniences, there are also huge security risks

During face recognition, a large amount of original face data is collected and stored and transmitted in the cloud, which may easily lead to the theft, leakage or tampering of face images and personal related information by others, causing certain losses and impacts on individuals and society

[0003] At present, the method used for face image encryption and recognition is mainly the face recognition method based on wavelet dimensionality reduction under homomorphic encryption. Although it has a high recognition effect, it has poor anti-geometric attack and occlusion attack ability and low robustness.

Images