Face acquisition system based on edge intelligent perceptron and privacy protection method

Abstract

The invention relates to a face acquisition system and privacy protection method based on an edge intelligent perception body, and the method comprises a face encryption process and a face decryption process, and the face information encryption process comprises the steps: obtaining a face image through the edge perception body, and obtaining a plurality of face regions in the face image; performing noise processing by using the feature extraction result of the face reference image, and performing superposition to protect the semantic features of the face; carrying out fuzzy processing on the face region and then storing the face region; in the face information decryption process, the visual terminal recognizes the face semantic features of a terminal user to form a decryption key, the face feature cloud database is traversed through a spatial-temporal index to carry out key matching, face feature self-decryption face pairing is achieved, then the encrypted video is presented through the generative adversarial network associated with the semantic features, and the face information is decrypted through the generated adversarial network associated with the semantic features. And multi-face targeted decryption in the image is realized. Compared with the prior art, the method has the advantages of image localization distributed storage, face feature cloud edge collaborative encryption, low-delay semantic transmission and the like.

Description

technical field [0001] The present invention relates to the technical field of image processing and face recognition, in particular to a face collection system and privacy protection method based on edge intelligent sensors. Background technique [0002] With the rapid development of artificial intelligence and computer vision, vision-based intelligent applications gradually occupy the market, and related application directions based on image classification, object detection and other sub-fields emerge in an endless stream, and computer vision technology has gradually integrated into our lives. [0003] With the continuous progress and development of big data application technology in modern society, the application of face recognition technology in the public domain has become an inevitable trend. While bringing convenient applications, it will also cause the loss of private information of face images. The combination of the information sharing and interactivity of the Inte...

AI Technical Summary

Problems solved by technology

This technology only protects from the perspective of user privacy information, which is not conducive to the application background of face images; if it does not touch the feasibility of face recognition, it is necessary to improve the feature extraction and storage scheme of face images, without affecting At the same time as the feature extraction effect of the network, noise processing and feature fusion are performed on the feature map, so that the feature cannot reflect the real face information, forming the protection of face features. On the other hand, for the situation of large-screen display, the large-screen Users should only be able to see their own face area, blur other faces captured by the camera, and then display them on the screen

Images