International Journal of Computer Applications (0975 – 8887) Volume 73– No.16, July 2013 8 Image Authentication System using LDPC and Watermarking Technique Ankit Bhatnagar M.Tech Scholar Computer Science & Engineering RKDFIST, Bhopal Jasvinder Pal Singh Assistant Professor Computer Science & Engineering RKDFIST, Bhopal Gaurav Shrivastava HOD Computer Science & Engineering RKDFIST, Bhopal ABSTRACT Now a days, the rate of using digital image is increasing exponentially because of its low cost and easy manipulation property. It is obvious that digital images captured from CCD cameras or other digital image devices have to be analyzed and determined before processing it, to keep the integrity and ensure the accuracy and reliability of the digital image. We need some sort of robust methods and standard procedures that guarantee and strengthen the authenticity of digital image. Image authentication is a technique that analyzes a digital image and determines whether it is altered or not. Image authentication technique is very useful for various organizations such as health care, law enforcement agencies and insurance sector etc. Image authentication is also important in content delivery via untrusted intermediaries, such as peer-to-peer (P2P) file sharing. Untrusted intermediaries might tamper the contents of image. Distinguishing the legitimate diversity of encodings from malicious manipulation is the challenge addressed in this paper. We developed a LDPC and watermark based new approach for image authentication. With the help of this approach we can authenticate images effectively. In our technique, we provide LDPC quantized image projection and the Encrypted image as authentication data. As well as watermark image that was embedded into original image to identify legitimate or illegitimate state of image authentication system. These data can be correctly decoded only with the help of an authentic image as side information. This technique provides the desired robustness against legitimate encoding alteration, while detecting illegitimate variations. Keywords: Image Authentication, Low Density Parity Check (LDPC), Digital Watermarking. 1. INTRODUCTION Proliferation of digital media is accompanied by increasing functionality and usability of software for manipulating digital media. For example, advancement in digital imaging technologies have led to the development of low-cost and high-resolution digital still and video cameras and scanners. Digital Videos and images generated by various sources are widely used in a number of applications from medical imaging and law enforcement to banking and daily consumer use. Relying on digital media such as law enforcement and security makes robust techniques for media authentication a must [2]. These techniques are also vital in content delivery via untrusted intermediaries, such as peer-to peer (P2P) file sharing or P2P multicast streaming [1]. In the P2P file sharing applications, intermediaries might tamper the contents of file for a variety of reasons, such as hindering with the distribution of a specific file, piggybacking unauthentic content, or generally discrediting a particular distribution system. Distinguishing the legitimate diversity of encodings from malicious manipulation is the major technical challenge for image authentication systems. Forensics, Watermarks and media hashes have been used in past for image authentication. In digital forensics, the user verifies the authenticity of an image solely by checking the received content [3,5]. Unfortunately, without any information from the original, one cannot entirely confirm the integrity of the received content because content unrelated to the original may pass forensic checking. Another alternative for image authentication is watermarking. A “fragile” watermark can be embedded into the host signal waveform without perceptual distortion [6,7]. Users can confirm the authenticity by extracting the watermark from the received matter. The system design should ensure that the watermark survives lossy compression, but that it “breaks” as a result of a malicious manipulation. Unfortunately, Watermark authentication is not backward compatible with previously encoded, unmarked contents which cannot be authenticated later. Embedded watermarks may also increase the bit-rate required when compressing a media file. Media hashing [8,9] achieves verification of previously encoded media by using an authentication server to supply authenticated data to the user. Media hashes are inspired by cryptographic digital signatures[10], but unlike cryptographic hash functions, media hash functions are supposed to offer the proof of perceptual integrity. Using a cryptographic hash, a single bit difference leads to an entirely different hash value. If two media signals are perceptually indistinguishable, they should have equivalent hash values. A common approach of media hashing is to extract features which have perceptual importance and should outlast compression. The authentication data are generated by compressing these features or generating their hash values. The user analyze the authenticity of the received content by comparing the features or their hash values to the authentication data. We propose a combination of Encryption based on Low Density Parity Check (LDPC) code and watermarking technique for image authentication. In [1], a method for backward-compatible image authentication based on distributed source coding is presented. This method provides a Slepian-Wolf encoded [11] quantized image projection as the authentication data which can be successfully decoded only by using an authentic image as side information. The fixed decoder used in [1,12] can do successful image authentication for JPEG compressed images but image authentication is not possible using fixed decoder if the channel applies contrast