The Effect of Wavelet Families on Watermarking Evelyn Brannock, Michael Weeks, Robert Harrison Department of Computer Science, Georgia State University, Atlanta, GA, USA Email: {evelyn, mweeks, rharrison}@cs.gsu.edu Abstract— With the advance of technologies such as the Internet, Wi-Fi Internet availability and mobile access, it is becoming harder than ever to safeguard intellectual property in a digital form. Digital watermarking is a steganographic technique that is used to protect creative content. Copyrighted work can be accessed from many different computing platforms; the same image can exist on a handheld personal digital assistant, as well as a laptop and desktop server computer. For those who want to pirate, it is simple to copy, modify and redistribute digital media. Because this impacts business profits adversely, this is a highly researched field in recent years. This paper examines a technique for digital watermarking which utilizes properties of the Discrete Wavelet Transform (DWT). The digital watermarking algorithm is explained. This algorithm uses a database of 40 images that are of different types. These images, including greyscale, black and white, and color, were chosen for their diverse characteristics. Eight families of wavelets, both orthogonal and biorthog- onal, are compared for their effectiveness. Three distinct watermarks are tested. Since compressing an image is a common occurrence, the images are compacted to determine the significance of such an action. Different types of noise are also added. The PSNR for each image and each wavelet family is used to measure the efficacy of the algorithm. This objective measure is also used to determine the influence of the mother wavelet. The paper asks the question:“Is the wavelet family chosen to implement the algorithm of consequence?” In summary, the results support the concept that the sim- pler wavelet transforms, e.g. the Haar wavelet, consistently outperform the more complex ones when using a non-colored watermark. Index Terms— watermark, digital watermarking, image wa- termarking, wavelet, discrete wavelet transform, I. I NTRODUCTION In the past, the U.S. has attempted to utilize copyright and trademarking law to protect creative digital content. However, legal means have not proven to be sufficient, and it can be an almost impossible task to hold the ownership of these works in the author’s control and curb the illegal theft of these labors. Watermarks embed a symbol of the owner of the works into the item itself, still allowing innocent consumers to continue to enjoy creative content. However, the hope is that the capability to identify those who nefariously copy, for example, music files from a CD to sell on the black market, can This paper is based on “Watermarking with Wavelets: Simplicity Leads to Robustness,” by E. Brannock, M. Weeks, and R. Harri- son, which appeared in the IEEE Proceedings SoutheastCon 2008, Huntsville, AL, USA, April 2008. c  2007 IEEE. be identified, and perhaps more successfully prosecuted. Businesses lose unknown profits from those who are willing to inexpensively reproduce artistic digital data, such as movies on DVDs and music on CDs in great volumes and instantaneously distribute worldwide to sell, without the right to do so. Therefore, creators and owners of the work are concerned that unauthorized copying and redistribution of their copyrighted works causes their economic returns to decline. As a result it has become of more significant consequence to study and find the most effective approaches to solve this problem. Watermarks serve to identify the source of the content and thus aid in investigating abusive duplication. Obviously, it is vital that the identifying marks which unassailably establish the true owner of the data are legible and identifiable: otherwise what is the use of embedding these proprietorships? Therefore, the impact of the size and nature of the data on the robustness of the embedded watermark will be investigated, in an extension of [1]. For the purposes of this study, an uncomplicated key will be used. However, as in other cryptographic systems, for commercial applications it should be large enough to make hacking attempts as unachievable as possible. Also, the effect of extensive search attacks on the watermark will be given a detailed examination. In this paper, the next section will cover the background and digital watermarking principles, section III will cover wavelets, and then section IV will discuss the method used. Section V presents results, and section VI concludes the paper. II. DIGITAL WATERMARKING A. Definition A digital watermark is “a digital code unremovably, robustly, and imperceptibly embedded in the host data and typically contains information about origin, status, and/or destination of the data”, according to Hartung and Kutter [2]. It is a form of steganography, because it hides the embedded data, often without the knowledge of the viewer or user. Since the purpose of steganography is the secret communication between two persons, the watermark can be considered to have been successfully attacked if its existence is determined. When contrasting with steganography, watermarks add the property of ro- bustness, which is the ability to withstand most common attacks [1]. The two common categories of attacks are 554 JOURNAL OF COMPUTERS, VOL. 4, NO. 6, JUNE 2009 © 2009 ACADEMY PUBLISHER