Watermark embedding method, watermark reading method and watermark system based on desensitized data

Abstract

The invention discloses a watermark embedding method, watermark reading method and watermark system based on desensitized data. The watermark embedding method based on desensitized data comprises thesteps: sorting data, setting the sequence of data columns, and recording the sequence of rows and columns of the data; performing desensitization processing on the sensitive field of the data to obtain a desensitization field; encrypting and encoding the traceability information to obtain watermark information; and embedding watermark information into the desensitization field through a preset rule, the preset rule being that one or more bits in the desensitization field are changed into preset symbols to serve as carriers of the watermark information. According to the technical scheme, data can be traced back effectively, more data information cannot be lost when watermarks are added in the encryption method, the watermark removing cost is low, sensitive information can be effectively protected, leakers can be traced back, and privacy rights and interests of enterprises and individuals are protected.

Description

technical field [0001] The invention relates to the technical field of data security, in particular to a watermark embedding method based on desensitized data, a watermark reading method and a watermark system. Background technique [0002] In the era of big data, while data brings great value, it also brings difficulties in user privacy protection. The leakage of sensitive information will give some criminals an opportunity to seriously damage the legal rights of enterprises and individuals. How to protect user privacy and prevent sensitive information leakage in the process of big data development and application has become a new challenge. The Cybersecurity Law of the People's Republic of China has stricter data privacy protection requirements, and data protection technology is becoming a popular research direction for universities and enterprises. [0003] In the prior art, when the sensitive data is exported, the only way to protect the data security is to desensitize ...

AI Technical Summary

Problems solved by technology

[0004]Because of the reproducibility of data, once a problem occurs, the data owner needs to prove his innocence and prove where the data output was leaked from

However, the existing desensitized data cannot be watermarked. Once the data spreads out of control, it is impossible to prove the source of the data leakage, which is not conducive to data security protection.

Images