My fantastic Webdings Logo

Scalable Secret Sharing of Compressed Multimedia


Shreelatha Bhadravati, Pradeep K. Atrey and Majid Khabbazian


Introduction

Advancements in computer technology has led to the widespread use of multimedia applications. Availability of devices to handle these multimedia data has made it easy to modify, copy and distribute multimedia data illegally. Multimedia data is vulnerable to attacks while being transferred over the network or when being stored for future use. Hence security of digital data is an important concern.

Preserving security of multimedia data is an important concern as, firstly the size of multimedia data is huge and processing in real time is a constraint and computationally expensive. Furthermore, the diversified application areas of digital content makes it more challenging to develop security measures as each application can have different end users and have specific requirements. For example, in the case of digital videos, in one scenario a low quality video is made available to all the users, as a means to promote the content as in pay-TV, HDTV. Whereas in applications like video conferencing and surveillance, it requires that the video is completely unintelligible to unauthorized users, preserving the privacy of the people and objects involved.

The task of securing digital content has been studied extensively in the past. Traditional encryption methods like AES are not suitable for digital data because of the computational complexity. To, overcome this problem selective encryption methods were proposed, in which only certain parts of the digital data is encrypted. But these methods were proved to be insufficient to provide confidentiality to digital data. Further, the encryption methods require the use of cryptographic keys. Storing the key is also trivial problem as it leads to single point failure. Hence it becomes necessary to develop a method to secure digital content

Thus the aim of our research is to provide a computationally secure method for protecting the images and videos in the compressed domain.