Global Journal of Advanced Engineering Technologies and Sciences INFRANGIBLE APPROACH OF DIGITAL IMAGE WATERMARKING USING DCT IN FULL COUNTER PROPAGATION NEURAL NETWORK
Poornima Rawat, Jitendra Badole, Sheetesh Sad · 2014
In recent years, internet revolution resulted in a rapid growth in multimedia applications. The rapid advancement of internet has made it easier to send the image accurate and faster to the destination. Besides this, it is easier to modify and misuse the valuable information through hacking at the same time. Digital watermarking is one of the best solutions for copyright protection of multimedia data. A watermark is a form, image or text that is impressed onto paper, which provides evidence of its authenticity. In this paper, an invisible watermarking technique is implemented. Keyword: Image watermarking, Full counterpropagation neural network, DCT I Introduction Watermark is a piece of information, which provides ownership of the document. Digital Image watermarking is idea of embedding such a watermark image into the cover image.A different digital image watermarking technique is proposed, In which we have used three watermarked images embedded in encoded cover image and operation is performed in full counter propagation neural network. We have also given an overview of image watermarking and different security issues. Various attacks are also performed on watermarked images and their impact on quality of images is also studied .Quality factors like PSNR,NC and a Error Rate are also discussed. Image Watermarking in transform domain using counter propagation algorithm has been used for embedding the watermark images into the the cover image. This work is implemented through MATLAB. A Digital water marking technique based on full counter propagation neural network for improving the robustness , imperceptibility and security based on comparison of error metrics is proposed [1].In this , digital image watermarking has done transform domain. In which the Error metrics or parameter like PSNR, NC are compared at extraction of watermarked image. Gunjal and Mathalkar [2] given the overview an overview of transform domain robust digital image watermarking algorithms. They have focused on Discrete Wavelet Transform as per ISO norms JPEG2000 has replaced Discrete Cosine Transform by Discrete wavelet Transform. Pooja and Kavita [3] has implemented three different digital watermarking techniques each from Spatial Domain (LSB) and Transform Domain (DCT and DWT) for evaluating their performance using various parameters such as PSNR, MSE, similarity ratio(SR), correlation(CORR) and BCR against robustness for attacks.