05-08-2011, 10:57 AM
[attachment=15066]
Abstract
Digital watermarking technique emerged as a tool for protecting the multimedia data from copyright infringement. In digital watermarking an imperceptible signal is embedded into the host image, which uniquely identifies the ownership. In the proposed algorithms, DNA sequence is used as a digital watermark, as the DNA sequences are unique and difficult to copy. This paper proposes two algorithms namely content based watermark algorithm using DNA sequence (CBDNA) and user specified watermark algorithm using DNA sequence (USDNA). In CBDNA the DNA sequence is generated from the cover data itself whereas in USDNA input data is chosen by the user and DNA sequence is generated based on the input data. These DNA sequences serve as a watermark for the cover data. The quality of the cover image and extracted watermark is measured using peak signal to noise ratio (PSNR) and normalized correlation (NC) respectively. The calculated values are tabulated and it shows that the proposed algorithm is withstanding many attacks, since watermark is available in all the frequency range of cover data. Keywords: Deoxyribonucleic Acid (DNA), Discrete Wavelet Transform (DWT), Content based watermarking algorithm using DNA sequence (CBDNA), User specified watermarking DNA sequence (USDNA)
1. INTRODUCTION
In recent years the phenomenal growth of the Internet has highlighted the need for mechanisms to protect ownership of digital media. Exactly identical copies of digital information, images, text or audio, can be produced and distributed easily. In such a scenario, we need to differentiate between the artist and the plagiarist [1]. Digital watermarking is a technique that provides a solution to the longstanding problems related to copyrights of digital data. In order to protect multimedia data many techniques are available including cryptography, steganography, digital watermarking and information hiding. Among these techniques digital watermarking is gaining popularity in the present days to protect copyright of multimedia data. The applications of digital watermarking include active ownership verification, copyright protection, finger printing and image authentication. Digital watermarking is defined as a process of hiding a piece of secret information (watermark) in cover data which can be extracted later for ownership verification, Copyright protection and for image authentication [7]. A digital watermark can be any piece of information about users such as date of birth, user name, telephone numbers and any user selected input data. Digital watermarking can be classified according to the domain in which watermark is embedded. In spatial-domain, the watermarking is done by changing the intensity value of pixels [3]. This kind of watermarking is simple and computation complexity is very low, since no frequency transform is needed. In frequency-domain watermarking, it embeds the watermark by changing the frequency components. Many transformation techniques are available to transform images from time domain to frequency domain including discrete Fourier transformation (DFT), discrete cosine transformation (DCT) [5][9], discrete wavelet transformation (DWT) [10] and discrete Hadamard transformation (DHT). Frequency domain transformation approach is more complex when compared to spatial domain transformation but has the advantage of robustness to image processing operations which may be unintentional and intentional. According to how the watermark is being detected and extracted, digital watermarking can be classified into blind watermarking technique and non-blind watermarking technique. In blind watermarking, the watermark detection and extraction does not depend on the availability of original image [2]. The drawback is when the watermarked image is seriously destroyed; watermark detection will become very difficult. In non-blind watermarking, the watermark can be detected if only the original image is available [15]. In [10], DNA sequences are being used in the field of information hiding, where the input text message is encoded into a DNA sequence. This text message is embedded into a host DNA using substitution cipher method. The output obtained is an encoded DNA sequence with the input message embedded into it. In this paper two new algorithms have been proposed for hiding a unique DNA sequence in a cover data. This proposed technique could be used to identify the owner of the cover data in a unique way and similarly it could also be used for authentication and copyright protection. Proposed algorithms could be classified as non-blind watermarking algorithm since it requires original cover data to extract the watermark. The discrete wavelet transformation technique is used to transform an image from time domain to frequency domain. In first algorithm named CBDNA the watermark is generated based on the content of the cover image whereas in second algorithm named USDNA the user specified input is considered for generating unique DNA sequence which is used as watermark.
