Access control method based on image recognition and user level

Abstract

The invention discloses an access control method based on image recognition and user levels. The access control method is characterized in that the access control method comprises the following steps of: obtaining access information; the image owner, the image visitor and the access controller have three roles; and an image processing module, an access control module, a user control module and an image management and distribution module. The method comprises the following steps: firstly defining the image sensitivity of an image owner, then utilizing deep learning to mine a sensitive area in an image, realizing classification and hierarchical management of an access controller on the image area, encrypting the identified image, realizing hierarchical management based on the identity level of a data visitor, and enabling different identity levels to access different image contents. Compared with the prior art, the method has the advantages that the sensitive area of the image is encrypted and stored, even if the sensitive area is maliciously leaked, the obtained sensitive area is only a non-sensitive area, and the security of the sensitive image is effectively guaranteed.

Description

technical field [0001] The invention relates to the technical field of digital image processing and information security, in particular to an access control method based on image recognition and user level. Background technique [0002] With the rapid development of the Internet and big data, data sharing has become an indispensable part, which brings convenience but also poses great challenges to the security of sensitive data. The so-called sensitive information mainly refers to that if it is improperly used, leaked, tampered with, stolen, or disseminated without permission, or the hidden value can be tapped by a third party through correlation analysis and other means, it will cause serious harm to national interests, business operations, and data owners. Seriously damaged data information. In the process of data exchange, if there is a lack of effective management and protection measures, sensitive information will be leaked. How to ensure data privacy and access secur...

AI Technical Summary

Problems solved by technology

The access control of the existing technology is often limited to the authorization protection of data, granting or restricting the user's access to relevant data based on different standards, which no longer meets the protection requirements of sensitive data, and cannot be applied to the effective distribution and access of sensitive data

[0004] In order to realize the protection of sensitive information, the combination of access policy and data encryption is also proposed at present. However, compared with authorization and encryption, defining sensitive information, accurately identifying, discovering, and mastering the location of sensitive information is an important aspect of access control. However, the current access control for sensitive information is often carried out on the premise of knowing the location of sensitive information, ignoring the mining of sensitive information

Data carriers are generally text, images, videos, etc. Text can be filtered and identified by setting sensitive words or topics, while images are different from general face recognition or image classification because of their poor readability, and a complete The image is often treated as an independent object for operation, but there are also sensitive areas and non-sensitive areas in an image, which are stored and managed indiscriminately, resulting in a large waste of resources

Images