Wavelet Based Image Watermarking Using Arithmetic Coding

Abstract

The recent growth in computer networks, and more specifically, the World Wide Web, has allowed multimedia data such as images to be easily distributed over the Internet. However, many publishers may be reluctant to show their work on the Internet due to a lack of security. The data which is distributed can be replicated easily without error, putting the rights of their owners at risk. Digital watermarks have been proposed as a way to tackle this issue. A digital watermark is an invisible signature embedded inside an image to show authenticity and ownership. An effective digital watermark should be perceptually invisible to prevent obstruction of the original image. It should be statistically invisible to prevent detection, and it should also be robust to many image manipulations, such as filtering, additive noise, and compression. In this project a watermarking algorithm is implemented for digital images using discrete wavelet transform (DWT) and arithmetic coding. The watermark which is a string of characters is compressed using arithmetic coding because of its high coding efficiency. This compressed watermark is then embedded into the host cover image. The various performance factors like peak signal to noise ratio (PSNR) are analyzed by changing the sub band decomposition level, gain factor etc and applying different attacks which gives different results on different values.