Public security view desensitization test data generation method and system and storage medium

Abstract

The invention discloses a public security view desensitization test data generation method and system and a storage medium, and the method comprises the steps: employing a non-sensitive human face view sample to desensitize a sensitive object human face view through a deep learning-based sample camouflage method, reserving the recognition features of a real view, and camouflaging the real view asa non-sensitive human face view; different from traditional encryption, content removal and other desensitization technologies, the scheme provided by the invention has the advantages that recognitionfeatures of partial attention in the original (human) face image are reserved, and the face data of the non-attention area is replaced with the face data of the public sample by utilizing the deep neural network, so that high-simulation view data desensitization is realized. By utilizing the scheme, the desensitization test data can be quickly generated in batches, the data can be shared and transmitted in batches on the premise of not losing the application value, and the leakage of key sensitive data can be effectively prevented. The invention can be widely applied to public safety test data generation and police artificial intelligence remote auxiliary systems.

Description

Technical field[0001]The invention relates to the field of artificial intelligence and information security, and in particular to public security view desensitization technology.Background technique[0002]In the public safety big data application, the data information obtained through collection, aggregation, and sharing often contains sensitive information that needs to be exchanged and analyzed between information platforms. In order not to affect the use, but also to effectively protect such information. Desensitization technology has become one of the hot information security technologies studied in the industry. Desensitization technology is an information security technology widely used in the protection of critical information and important data. It mainly includes methods such as replacement, scrambling, averaging, anti-inference, offset, encryption, and partial shielding. In terms of video image applications, sensitive information mainly comes from specific tasks, scenes and...

AI Technical Summary

Problems solved by technology

[0005] In view of the problems faced by the sensitive information desensitization technology in video images in the existing public security big data application system, a new sensitive information desensitization scheme in video images is needed

Images