Digital Watermarking is the process of hiding a message into the digital file like image, audio, video, etc. for the protection and authentication of the content. Sometimes it is related with steganography, as both are used for hiding the message but difference is that in steganography there is no relation between digital file and message, it is only used to hide the message existence, on the other hand with watermarking one hides a message related to the digital content. Digital Watermarking consist of several characteristics like Invisibility, Robustness, Readability and Security from any unauthorized user. Keywords:Digital Watermarking, PSNR,Watermarking Attacks I. INTRODUCTION In this paper discrete Cosine transform is used for insertion and extraction of watermark in original image. We will divide the image into blocks and watermark is added into the first bit of each block. This DCT technique is much simpler and robust than others. Just like Discrete Fourier transform, DCTalso transforms a signal or image from the spatial domain to the frequency domain. DCT helps in separating the image into spectral sub-bands having different visual quality. Mohamed A. Suhail et al.[1] developed a watermarking algorithm based on discrete cosine transform and image segmentation. In this, a image is segmented into different portions based on the voronoi diagram and features extraction points. After, in the DCT domain of each image segment, a pseudorandom sequence of real numbers is added. Soumitra Roy et al. [2] proposed a color multiple watermarking based on DCT and repetition code. In this, different watermark embedding is done over green and blue component, zigzag scanning of each DCT block is done to ensure watermarked host image imperceptibility. Rajput, Tiwari [4] generated digital watermarking using 2-DWT and 2-DCT by extracting RGB elements and applying for more quality and security. In this research, they have used red component for embedding of secret image into the original image. In this, low frequency component is achieved by applying 2-level DWT over the frequency component of the original image. Then Inverse 2-level Inverse DCT is applied to form resultant secret image. Later the 2-level DCT is applied over the RGB elements. For calculating PSNR value, blurred attack is done over the image. Priyanka, Maheshkar [3] noted that most of the visual part that is important reside in the low frequency sub band. Two images are taken i.e a cover image and a binary watermark, then cover is divided into non overlapping blocks and then block wise DCT is applied. Middle frequency sub-band of DCT block is used for watermark embedding for reducing loss during compression process and to reduce quality degradation. Yu Yang et al [8] proposed technique based on DWT and DCT in which a Zero-Watermark is embedded by comparing adjacent frames mean absolute values coefficients and DWT-DCT applied on the host signal frames. The proposed algorithm is aslo robust to common signal processing operations. S. Fazli, M. Moeini [10...
