Design, Development of a Watermarking System for Scalable Video Authentication

About The Project

Title: Design, Development of a Watermarking System for Scalable Video Authentication

Sponsoring Agency: Dept. of Information Technology, Govt. of India

Budget: Rs. 102.051 (in lakhs)

Principal Investigator: Dr. Arijit Sur

Co-Principal Investigator: Dr. Arnab Sarkar



Technical Details

With the enormous growth of Internet and advent of media communication technologies, scalable media (image, video etc.) coding is considered to be integral texture of video representation and transmission. In this coding approach adaptation is done in the media content to fulfill the varying transmission and presentation demands of the consumers. A hypothetical scalable coding framework is depicted in the following figure:





Assuring protection such as ownership as well as video content authentication becomes an important part of efficient and secure multimedia communication. This poses challenging research problems especially in their scalable representation. Encryption and cryptographic hashes are proposed for this purpose. But it is observed that the scalability property of the bit stream is lost if the video bit stream is encrypted with conventional cryptographic ciphers.

On the other hand, digital video watermarking is a technique which inserts a digital signature into the video stream. The signature can be extracted or computed to authenticate the ownership of the media or the media itself. The application of video watermarking includes copyright protection, ownership authentication, video authentication, broadcast monitoring etc. Watermarks are designed to be detectable, even when the multimedia content is altered during transmission. This advantage motivates researchers to propose watermarking solutions for security enforcement of the multimedia content. Over last decade, video watermarking becomes a popular research topic. Despite intense research on the video watermarking, very less attention is paid on its application for scalable video authentication.

In this project, a scalable video watermarking system will be developed. The proposed system will have the following salient features:

i) The Robustness of Watermark in the video will be evaluated against Content Adaptation Attack
ii) The perceptual quality of the watermarked video at different scale will be tested to determine the acceptability of viewers, and
iii) Individual bit streams at different scalability level will be authenticated separately


To meet the above goals, the work is divided in three major Phases. The brief description of these Phases is as follows:

PHASE-I:

Design a Framework to Analyze the Robustness of the Watermark with respect to different types of Scalability
The first phase of our project is aimed to design a framework to evaluate the robustness as well as perceptual quality of the watermarking strategies against content adaptation attacks.There are a few existing image watermarking evaluation tools available in the literature for measuring the performance of the watermarking algorithms against different intentional attacks (including cropping, average filtering, scanning etc.) and unintentional attacks (natural image processing tasks, such as compression, rotation, scaling etc.). These include Stirmark [28], Checkmark [29] and Watermark Evaluation Test bed (WET)[30]. Recently Bhowmik [26] has proposed a robustness evaluation bench for image watermarking subjected to content adaptation attack.

PHASE-II:

Design and develop robust separate scalable watermarking modules for different scalability parameters like Quality Scalability, Resolution / Spatial Scalability, Temporal Scalability etc
In this part of the project, robust watermarking solution will be provided against different content adaptation attacks. In our project, three different scalability parameters are considered. In this part of the project, the robustness of the proposed algorithm will be defined separately for different scalability parameters like resolution scalability, quality scalability and temporal scalability.

PHASE-III:

Develop a Watermarking scheme integrating above scalability aspects and simulate a practical simulation of scalable video transmission to analyze the robustness of the developed watermarking system
In this part, firstly three watermarking algorithms developed for individual scalability parameters are merged to get a complete watermarking system against content adaptation attacks.
Secondly, the scalable video transmission system will be simulated for verifying the robustness of the proposed watermarking scheme in real life situation. In this system, watermarked video content will be stored in the server. Different end users will be connected to the network through some access points where a scalable extractor will be installed. This software patch will understand the user’s demand and extract the proper bit stream from the server which will be most appropriate for the end users with respect to the resolution, quality and frame rate etc.