A Block-wise-based Fragile Watermarking Hybrid Approach using Rough Sets and Exponential Particle Swarm Optimization Lamiaa M. El Bakrawy 1 , Neveen I. Ghali 1 , Tai-hoon Kim 2 and Aboul ella Hassanien 3 1 Al-Azhar University, Faculty of Science, Cairo, Egypt Email: lamiaabak@yahoo.com, nev ghali@yahoo.com 2 Hannam University, Korea, taihoonn@hannam.ac.kr 3 Cairo University, Faculty of Computers and Information, Cairo, Egypt Email:aboitcairo@gmail.com Abstract In this paper, we propose a fragile watermarking hybrid approach using rough set k- means and exponential particle swarm optimization (EPSO) systems. It is based on a block-wise dependency mechanism which can detect any alterations made to the protected image. Initially, the input image is divided into blocks with equal size in order to improve image tamper localization precision. Then feature sequence is generated by applying rough k-means and EPSO clustering to create the relationship between all image blocks and cluster all of them since EPSO is used to optimize the parameters of rough k-means. Both feature sequence and generated secret key are used to construct the authentication data. Each resultant 8-bit authentication data is embedded into the eight least significant bits (LSBs) of the corresponding image block. We gives experimental results which show the feasibility of using these optimization algorithms for the fragile watermarking and demonstrate the accuracy of the proposed approach. The performance comparison of the approach was also realized. The performance of a fragile watermarking approach has been improved in this paper by using exponential particle swarm optimization (EPSO) to optimize the rough k- mean parameters. The proposed approach can embed watermark without causing noticeable visual artifacts, and does not only achieve superior tamper detection in images accurately, it also recovers tampered regions effectively. In addition, the results show that the proposed approach can effectively thwart different attacks, such as the cut-and paste attack and collage attack, while sustaining superior tamper detection and localization accuracy. Keywords: Fragile Watermarking, rough sets,exponential particle swarm optimization 1 Introduction The rapid expansion of the Internet and the overall development of digital multime- dia content and nonlinear media distribution requires new enabling technologies, beyond traditional approaches such as password-based encryption that are used for safe custody of private keys do not provide adequate security due to very low entropy in user cho- sen passwords. Biometric-based personal identification techniques that use physiological or behavioral characteristics are becoming increasingly popular compared to traditional 1 International Journal of Future Generation Communication and Networking Vol. 4, No. 4, December, 2011 77