Image Content Protection in Cloud through Reversible Data Hiding: Theory and Implementation
This tutorial presents the audience with the methods for protecting the shared/stored image content in cloud environment through reversible data hiding process. Cloud systems are currently used to facilitate data sharing capabilities and this provides profuse benefits to the individual users and to the organizations/medical heath care systems etc. Through social networking applications like Facebook, Whatsapp etc., the benefits of sharing data/images are numerous – eg., we can share photos, videos, information and events, etc. Importantly, in modern healthcare environments, healthcare providers are storing and sharing electronic medical records and medical images via the cloud. This has removed the geographical dependence between healthcare provider and patient. The sharing of medical images and data makes the remote monitoring and diagnosis of patients feasible without the patient leaving their house. To quote simply, even the personal photos and videos we capture in our mobile phones are backed-up in the cloud server and are sometimes even shared among our groups.
On the other hand, these clouds are susceptible to huge privacy and security attacks. Though ethically responsible, the cloud providers are not legally liable for securing the shared images. Inspite of the security measures they offer, many privacy and security attacks occur both from external servers, and from within the cloud provider themselves as they usually have direct access to stored data and do espionage to sell to third parties in order to gain profit. Since the Cloud is open to so many privacy and security issues, many users are apprehensive about sharing/storing their most critical data/images in cloud. Hence the real benefit of cloud is not harvested 100%. There do exists some security mechanisms for data sharing in cloud like access control to a group of users, key management, group management, encryption/decryption of data etc. Though they are computationally expensive, they are well suited for data and not for images. Hence a unique security measure for protecting the image content is needed beyond these measures.
In recent years a special kind of data hiding is employed for protecting the images that embeds the secret/confidential data into the original image to yield the stego image, called as Reversible Data Hiding. The confidential data/electronic patient records are hidden inside the images as payload. The original image as well as the embedded payload is recovered from the stego image. Further, the reversible data hiding schemes are robust against the intentional or the unintentional attacks, and are imperceptible to avoid the attraction of attacks. These images can further be encrypted to provide a complete security solution to the images stored on the cloud. Even when they are acquired by the attackers, they don’t provide any sensible information to them and hence the fear of security breaches happening in the cloud could be alleviated to a greater extent.
This tutorial introduces to the novices the basics, the requirements, characteristics, components in the information hiding systems and image sharing in the cloud. Further the special requirements for reversible data hiding, types of algorithms, spatial domain and transform domain schemes, strengths and weaknesses of the schemes are to be presented. Further the open research issues like image privacy preservation, ideal characteristics of image sharing system on cloud, challenges in data hiding schemes, possible directions for further progress will be presented.