01-08-2012, 02:49 PM
Robust Secure SVD Based DCT – DWT Oriented Watermarking Technique for Image Authentication
[attachment=29889]
Abstract
The digital multimedia content authentication and copyright protection has become an important issue in the recent years. Digital Watermarking Technology has been proposed for the implementation of Digital Right Management. Digital watermarking is the process of embedding information into digital multimedia content such that the information can later be extracted or detected for a variety of purposes including copy prevention and authentication proof.
INTRODUCTION
The rapid increase in exchange and transmission of digital information through internet in the past two decades had demanded the need of technique which prevents the unauthorized users to tamper and distribute the data illegally and to provide the ways to authenticate users to prove their ownership on digital content. The technique used to achieve aforesaid objectives is Digital Watermarking. Digital watermarking is a technique of embedding one digital information into another digital information without changing the content of original information. The technology offers a means of conveying information inside a digital media file (for example, inside a photo, movie, or song). In Digital Image Watermarking the inserted watermark should not degrade the visual perception of an original image and must be robust. Therefore, it must withstand the attacks like JPEG compression.
BACKGROUND
Discrete Cosine Transform is used to convert the Time domain signal into the Frequency domain signal. The 2-dimensional DCT of a given matrix gives the frequency coefficients in form of another matrix. Left topmost corner of the matrix represents the lowest frequency coefficients while the right bottom most corner represents the highest frequency coefficients. Using the DCT, an image is easily split up into pseudo frequency bands, and in this work watermark is inserted into the middle band frequencies because in all frequency domain watermarking schemes, there is a conflict between robustness and transparency. If the watermark is inserted in perceptually most significant components, i.e. low frequencies; the scheme tends to be robust to attacks but it is difficult to hide the watermark.
EXPERIMENTAL RESULTS AND DISCUSSION
To evaluate the performance of the proposed technique, the test case is being simulated by taking Lena 512 × 512 grayscale image (shown in Fig. 1 (a)) with 8 bits / pixel resolution for watermarking. The watermark image is of 128 × 128 size which is also grayscale and of 8 bits / pixel resolution. The value of the constant ‘α’ used is 0.03. The metric used to evaluate the performance of proposed technique is NCC (‘Normalized cross co relation’) which is defined as below.
[attachment=29889]
Abstract
The digital multimedia content authentication and copyright protection has become an important issue in the recent years. Digital Watermarking Technology has been proposed for the implementation of Digital Right Management. Digital watermarking is the process of embedding information into digital multimedia content such that the information can later be extracted or detected for a variety of purposes including copy prevention and authentication proof.
INTRODUCTION
The rapid increase in exchange and transmission of digital information through internet in the past two decades had demanded the need of technique which prevents the unauthorized users to tamper and distribute the data illegally and to provide the ways to authenticate users to prove their ownership on digital content. The technique used to achieve aforesaid objectives is Digital Watermarking. Digital watermarking is a technique of embedding one digital information into another digital information without changing the content of original information. The technology offers a means of conveying information inside a digital media file (for example, inside a photo, movie, or song). In Digital Image Watermarking the inserted watermark should not degrade the visual perception of an original image and must be robust. Therefore, it must withstand the attacks like JPEG compression.
BACKGROUND
Discrete Cosine Transform is used to convert the Time domain signal into the Frequency domain signal. The 2-dimensional DCT of a given matrix gives the frequency coefficients in form of another matrix. Left topmost corner of the matrix represents the lowest frequency coefficients while the right bottom most corner represents the highest frequency coefficients. Using the DCT, an image is easily split up into pseudo frequency bands, and in this work watermark is inserted into the middle band frequencies because in all frequency domain watermarking schemes, there is a conflict between robustness and transparency. If the watermark is inserted in perceptually most significant components, i.e. low frequencies; the scheme tends to be robust to attacks but it is difficult to hide the watermark.
EXPERIMENTAL RESULTS AND DISCUSSION
To evaluate the performance of the proposed technique, the test case is being simulated by taking Lena 512 × 512 grayscale image (shown in Fig. 1 (a)) with 8 bits / pixel resolution for watermarking. The watermark image is of 128 × 128 size which is also grayscale and of 8 bits / pixel resolution. The value of the constant ‘α’ used is 0.03. The metric used to evaluate the performance of proposed technique is NCC (‘Normalized cross co relation’) which is defined as below.