Investigation and Analysis of Vulnerability of Attacks on Watermarked Image and Its Enhancement

This paper demonstrate the attempt to evolve feasible solution to the scenario in which different possible attacks on embedded watermark can be avoided along with the enhancement of the content retrieved from noisy image post watermarking. With the revolution of information technology and wide area networking, data has become less private where in the admittance of media as well as the attempts to change and manipulate the contents of information has become a universal issue. Watermarking techniques have to be used to protect the copyright of the media and for the digital management but without compromising on the visual front. A universal DWT technique is used in this work. The algorithm for embedding watermark into the original image has been developed with novelty for better results than existing mechanisms. The various attacks such as image resizing, image cropping and image filtering are employed on the watermarked image to investigate the reliability of embedding algorithm. It is reported that Peak Signal to Noise Ratio (PSNR) value obtained before and after the attack has been in the ratio of 0.01 to 0.07. The extracted watermark is addressed again to enhance the PSNR value with various de-blurring techniques such as DCT compression, image cropping with noise treatment, Normal and Wiener filtering.