Abstract. Keywords: Genetic Algorithm, Mean Square Error, Peak Signal to noise Ratio, Image fidelity. 1. Introduction

Transcription

1 A Report on Genetic Algorithm based Steganography for Image Authentication by Amrita Khamrui Enrolled Scholar Department of Computer Science & Engineering, Kalyani University Prof. (Dr.) J K Mandal Professor Department of Computer Science & Engineering, Kalyani Univerasity Abstract Image authentication is done by embedding image/ message in spatial domain or frequency domain. A mask is chosen from the source image in row major order. Bits of the authenticating image are embedded in each byte of the source image block where the position is chosen using hash function. A hash function is a function that takes variable size inputs and return fixed size output. The position will vary from 0 to 4. The hash function is used to find out the position of embedding that vary for every byte of image and is gives more security of the hidden information. The embedding process is repeated throughout the image. The dimension of the authenticating image along with the content are embedded. Genetic Algorithm is used to minimize the difference between the source & embedded image. Keywords: Genetic Algorithm, Mean Square Error, Peak Signal to noise Ratio, Image fidelity. 1. Introduction Steganography is the method of hiding digital information within any computer files/ image. Steganography offers an essential alternative to image integrity and authenticity problem. It is a kind of data hiding technique that provides another way of security protection for digital image data. Steganography can be achieved in two ways. One is spatial domain steganography and another is frequency domain steganography. In spatial domain steganography the hidden information is directly embedded into image pixels. In frequency domain steganography the image pixels are first transformed into frequency domain using discrete fourier transformation/ discrete cosine transformation / discrete wavelet transformation etc. then the information is embedded on it. In addition to hide the information Genetic Algorithm is included in the research work to incorporate another layer of security as well as to minimize the difference between the source and embedded image pixel for more sensitive application like military people, research institute and medical diagnosis etc. The most common method to make this alteration is usage of the least-significant bit (LSB) developed through masking, filtering and transformations on the source image. 2. Existing Work The candidate proposed various techniques for image authentication. The techniques are divided into spatial and frequency domain. In spatial domain steganography the authenticating image pixels are directly incorporated in the source image pixel. In frequency domain steganography the candidate used discrete fourier/ discrete cosine/ discrete wavelet transformation to generate the coefficients. Frequency domain steganography offer better security as compared to spatial domain. In addition genetic algorithm is proposed to enhance a layer of security as well as to minimize the difference between the source and embedded image pixel to prevent RS attack. The main objective is to embed authenticating image pixels as much as possible with a bare minimum distortion to achieve high PSNR and good IF. Embedding is initiated by taking a mask from source image. Authenticating image bits are embedded

2 onto the source image pixel based on some hash function or some predetermined way. In case of frequency domain technique the source image pixels are transformed into transformed domain using DFT/ DCT/ WT etc after which embedding takes place. Genetic algorithm is applied on the embedded stego image to minimize RS attack. Thirty two bits of the stego intermediate image is taken as initial population. New generation followed by crossover and mutation are applied on initial population. Some of the techniques devised by the candidate with result are explained below. 2.1 A framework for block based image authentication (FBIA). This study deals with devising and implementing a novel technique for image authentication and secured message transmission based on hiding large volume of data into gray scale images. The results have been tested through histogram analysis and standard deviation computation of the source image with the embedded image and compared with popular existing steganographic algorithms like S-Tools[13]. 2.2 An authentication technique for image/legal document (ATILD). The paper presents an image/legaldocument authentication and secures message transmission technique by embedding message/image/message-digest into color images. Experimental results are tested with the popular existing steganographic algorithms like S-Tools[13] where the proposed ATILD is capable to hide large volume of data than S-Tools and shows better performance 2.3 A data embedding technique for gray scale image using genetic algorithm (DEGGA). In this proposal image authentication/data hiding technique through steganographic approach using Genetic Algorithm has been studied. Various statistical parameters computed are compared with the existing genetic algorithm based steganographic algorithm Ran et al.[4] which shows that proposed DEGGA obtained better results in terms of PSNR. 2.4 A genetic algorithm based steganography on color images (GASCI).In this approach a Genetic Algorithm based color image authentication/data hiding technique through steganographic approach, termed as GASCI has been proposed. Comparison of PSNR has been made with GASCI and existing approach [5]. 2.5 A data- hiding scheme for digital image using pixel value differencing (DHPVD). In this scheme pixel value differencing based image authentication/data hiding technique through steganographic approach in spatial domain, termed as DHPVD, has been proposed. The numbers of secret bits of the hidden image that are embedded in consecutive two pixels of the source image depend on the difference between the pixels. Experimental results are tested with existing approach [6], [7]. 2.6 A genetic algorithm based steganography in frequency domain (GASFD). Here Genetic Algorithm based color image authentication/data hiding technique through steganographic approach, in frequency domain using Discrete Fourier Transform (DFT) termed as GASFD, has been proposed. 2.7 A novel genetic algorithm based data embedding technique in frequency domain using z transform (ANGAFDZT). In this proposal a transformed domain based gray scale image authentication/data hiding technique using Z transform (ZT) termed as ANGAFDZT, has been proposed. High PSNR obtained for various images compared to existing approach Chin-Chen Chang et al.[8] conform the quality of invisible watermark of ANGAFDZT. 2.8 An image authentication technique in frequency domain using genetic algorithm (IAFDGA). In this paper Genetic Algorithm based image authentication technique in frequency domain using Z

3 transformation (IAFDGA) has been proposed. It is seen from Comparison has been made with existing approach [9]. 3. The Techniques 3.1 A novel genetic algorithm based image authentication technique using wavelet transform (ANGAIAWT). In this study a novel Genetic Algorithm based image authentication technique in frequency domain using Haar Wavelet transform has been proposed. The proposed algorithm obtains better image fidelity and high PSNR as compared to existing approach Huang et al [11]. 3.2 A Genetic Algorithm based Steganography using Discrete Cosine Transformation (GASDCT). In this paper a Genetic Algorithm based steganographic technique in frequency domain using discrete cosine transform has been proposed. The proposed technique is compared with Chang Chin et al [12] approach in terms of PSNR and capacity. 3.3 A wavelets transform based image authentication approach using genetic algorithm (AWTIAGA). In this paper a Genetic Algorithm based image authentication technique in transformed domain using haar wavelet transform is proposed. The proposed technique obtain better PSNR as well as high embedding capacity as compared to the existing [11]. 4. Result A comparative study of PSNR has been made between various techniques. It has been seen that PSNR is in between 35 to 50 (approx) which is satisfactory as the better value of PSNR improves image quality. In the following plot X axis indicates PSNR where as Y axis denotes the different techniques. It has also been notices that PSNR is gradually increasing which indicates improvement of different techniques. Conclusion Fig 1: Comparison of PSNR for various techniques The main objective is to authenticate image to ensure the security against the RS analysis which is the most notable steganalysis algorithm. It detects the steg-message by the statistic analysis of pixel values. The cover image can be either gray scale or color. The cover and authenticating images are

4 both benchmark images. it is clear that the proposed techniques obtained high PSNR ratio along with good image fidelity for various images. The payload may be increased based on the requirement. Future Scope In due course of research it is planned for generating some more techniques based on different transformation to enhance the security and authenticity and to authenticate the image/ document more efficiently. List of publications: 1. Ghoshal N., Mandal, J. K., Khamrui A., A Framework for Block based Image Authentication (FBIA), Proceedings of 4th IEEE International Conference on Industrial and Information Systems, ICIIS09, ISBN: , Peradeniya University, Sri Lanka, pp , December 28th - 31st, Mandal, J. K., Ghoshal N., Khamrui A., An Authentication Technique for Image/Legal Document (ATILD), Journal of Signal Processing Systems for Signal, Image, and Video Technology (formerly the Journal of VLSI Signal Processing Systems for Signal, Image, and Video Technology) Springer Science+Business Media, LLC /s Mandal, J.K., and Khamrui, A, A Data Embedding Technique for Gray Scale Image Using Genetic Algorithm (DEGGA), IEEE Sponsored International Conference on Communication, Computing & Security (Proceedings by Excel India Publishers ISBN ), NIT Rourkela, pp January Mandal, J. K., Khamrui, A, A Genetic Algorithm based Steganography on Color Images (GASCI)" International journal of Signal and Imaging Systems Engineering (IJSISE), Indersciences, accepted may Mandal, J. K., Khamrui, A., A Data- Hiding Scheme for Digital Image using Pixel Value Differencing (DHPVD), International Symposium on Electronic System Design (ISED),Print ISBN: , December, Kochi, India, Khamrui, A., Mandal, J. K, A Genetic Algorithm based Steganography in Frequency Domain (GASFD), IEEE 1st International Conference on Communication and Industrial Applications (ICCIA- 2011), Print ISBN: , 26th 28th December, Science City, Kolkata, Khamrui, A., Mandal, J. K, et. al. A Novel Genetic Algorithm Based Data Embedding Technique in Frequency Domain Using Z Transform (ANGAFDZT), DPPR- 2012, Advances in Intelligent Systems and Computing, 177, ISSN No.: , Springer, pp , July 13-15, Chennai, India(2012). 8. Mandal, J. K., Khamrui, A, An Image Authentication Technique in Frequency Domain using Genetic Algorithm (IAFDGA), International Journal of Software Engineering & Applications(IJSEA), ISSN : (Online); (Print), Indexed on DOAJ,EBSCO,INSPEC,ProQuest etc.,doi: /ijsea , Vol. 3, No. 5, pp.39-46, Sept ( 9. Khamrui, A., A Novel Genetic Algorithm based Image Authentication technique using Wavelet Transform (ANGAIAWT), Zonal Seminar, The Institute of Electronics and Telecommunication

5 Engineers Kolkata Centre, ISBN No , pp , Salt lake, August 30-31, Khamrui, A., Mandal, J. K, A Genetic Algorithm based Steganography using Discrete Cosine Transformation (GASDCT), accepted on The First International Conference on Computational Intelligence: Modeling, Techniques and Applications, CIMTA-2013, ISBN: , University of Kalyani, India, Sept , Khamrui, A., Mandal, J. K, A Wavelet Transform based Image Authentication approach using Genetic Algorithm (AWTIAGA), accepted by Computer Society of India, 48 th Annual Convention, Dec 13-15, Visakhapatanam, References [1] Lifang Yu et al., Improved adaptive LSB steganography based on chaos and genetic algorithm EURASIP Journal on Advances in Signal Processing, Volume 2010, Article ID doi: /2010/ [2] Elham Ghasemi et al., High Capacity Image Steganography Based on Genetic Algorithm and Wavelet Transform, Intelligent Control and Innovative Computing, Lecture Notes in Electrical Engineering 110, DOI / [3] Shen Wang et al., A Secure Steganography Method based on Genetic Algorithm Journal of Information Hiding and Multimedia Signal Processing 2010 ISSN Volume 1, Number 1, Jan [4] Ran-Zan Wang, Chi- Fang Lib, and Ja- Chen Lin, Image hiding by optimal LSB substitution and Genetic algorithm, Pattern Recognition Society, Published by Elsevier Science Ltd [5] Wang, R-Z et Al, Image hiding by Optimal LSB substitution and Genetic Algorithm,Pattern Recognition Society, Published by Elsevier Science Ltd, [6] Cheng- Huing Yang et al., Adaptive Data Hiding in Edge Areas of Images With Spatial LSB Domain Systems, IEEE transactions on information forensics and security, VOL. 3, NO. 3, Sept, pp , ISSN: , [7] Chang and H. W. Tseng, A steganographic method for digital images using side match, Pattern Recognit. Lett., vol. 25, no. 12, pp , [8] Chin-Chen Chang et al Reversible hiding in DCT- based compressed images, ScienceDirect Information Science vol :177, Pages: , July 2007.