IJARCCE ISSN (Online) 2278-1021 ISSN (Print) 2319 5940 International Journal of Advanced Research in Computer and Communication Engineering ISO 3297:2007 Certified Vol. 6, Issue 4, April 2017 Copyright to IJARCCE DOI10.17148/IJARCCE.2017.64158 847 Hardware Implementation of Robust Digital Image Watermarking using Neural Network– A Survey Sneha Nyamagoud 1 , Suresh Kuri 2 , Praveen Kalkundri 3 M.Tech. Student, Dept. of E & C, KLS, GIT, Belagavi, India 1 Professor, Dept of E & C, KLS, GIT, Belagavi, India 2 Professor, Dept. of E & C, KLS, GIT, Belagavi, India 3 Abstract: Watermarking is the process of hiding the digital information, such as image, audio and video. It is typically used for authentication and copyright. Communication is a major part in our day-to-day life, now a day’s exchanging of information in the form of image, audio and video takes place through internet. There are some issues or threads associated with internet transformation with respect to intellectual property rights (IP rights) like forgery, authentication and copyrights. In this paper we limit the survey to images only and we present a detailed survey of existing and newly existing watermarking techniques. Keywords: Classification of Watermarking, Properties, Neural network, Performance matrices, Applications and Watermarking techniques. I. INTRODUCTION In general information hiding techniques are stenography, cryptography and watermarking. Watermark is a target image, impressed onto a paper which provides evidence of authenticity. Digital image watermarking is the process of superimposing a watermark into a original image or host image, for the copyright [1, 2]. According to working domain, watermarking can be divided as spatial and frequency domain. Least significant bit (LSB) and patch work are comes under spatial domain and discreet wavelet transform and discreet cosine transform comes under frequency domain [3]. A. Classification of Digital Watermarking Watermark technique can be classified as visible invisible, robust & fragile, asymmetric& symmetric, public& private and stenography & non-Stenography watermarking techniques. Digital watermarking describes methods and technologies that hide information, for example a number or text, in digital media, such as images, video or audio. The hiding process has to be such that the modifications of the media are imperceptible [5]. For images, this means that the modifications of pixel values have to be invisible. The watermark must be either robust or fragile, depending on the applications. Robust means the capability of the watermark to resist manipulations of the media, such as lossy compression, scaling and cropping among others. And for fragile the watermark should not resist tampering, or would resist only up to a certain, predetermined extent [15]. B. Properties of Digital Watermarking Some properties of digital watermarking are, effectiveness- this determines the probability that message in a watermarked image will be correctly detected. Image fidelity- it checks for the transparency of the image. Payload size- determines the size of watermark and original image [4]. False positive rate- this property determines the number of a digital works that are identified to have a watermark embedded when in a fact they have no watermark embedded. Robustness- determines that watermark can easily embeddable and can easily retrievable. In this paper we provide latest digital image watermarking techniques [16]. C. Requirement of Digital Watermarking  Robustness: This is important requirement of digital watermarking. It is used to measure a correlation between extracted watermark and original watermark. It can be measured by using normalized correlation (NC). Value of NC must be as high as possible.  Security : Security in watermarking determines that Watermark should be difficult to remove or alter without damaging the original image as all watermarking system seek to protect watermark information without loss of generality.