International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 351 ISSN 2229-5518 IJSER © 2014 http://www.ijser.org Image Encryption Using Chaotic Based Artificial Neural Network Minal Chauhan 1 , Rashmin Prajapati 2 Abstract— Cryptography is the science to transform the information in secure way. Encryption is best alternative to convert the data to be transferred to cipher data which is an unintelligible image or data which cannot be understood by any third person. Images are form of the multimedia data. There are many image encryption schemes already have been proposed, each one of them has its own potency and limitation. This paper presents a new algorithm for the image encryption/decryption scheme which has been proposed using chaotic neural network. Chaotic system produces the same results if the given inputs are same, it is unpredictable in the sense that it cannot be predicted in what way the system's behavior will change for any little change in the input to the system. The objective is to investigate the use of ANNs in the field of chaotic Cryptography. The weights of neural network are achieved based on chaotic sequence. The chaotic sequence generated and forwarded to ANN and weighs of ANN are updated which influence the generation of the key in the encryption algorithm. The algorithm has been implemented in the software tool MATLAB and results have been studied. To compare the relative performance peak signal to noise ratio (PSNR) and mean square error (MSE) are used. Index Terms— Chaotic maps, Image encryption, Chaotic cryptosystems, Artificial neural network- ANN, Peak signal to noise ratio- PSNR, Mean square error- MSE, Cipher text, Plain text ——————————  —————————— 1 INTRODUCTION In recent years, more and more electronic services and de- vices are available, like mobile phones and personal digital assistant PDAs, also started to provide the additional func- tions like saving and exchanging multimedia data. The preva- lence of multimedia technology has promoted digital images and videos to play a more significant role than the traditional dull texts, which demands a serious protection of users’ priva- cy and so the protection of multimedia data is becoming very important. There are so many different techniques should be used to protect confidential image data from unauthorized access. To fulfill such privacy and security needs in various applications, encryption of images is very important to aggra- vate malicious attacks from the unauthorized parties. This paper is organized as follows In Section 1; general guide line about cryptography is presented. In Section 2, Pro- pose method is presented. Finally, concluded in section 3. Here is a quick introduction of chaotic systems and ANN systems. 1.1 What is Image Encryption? Image encryption is a hiding of information. An original important and confidential plaintext is converted into cipher text that is apparently random nonsense. This cipher text can be saved or transmitted over the network. At the receiver, the cipher text can be transformed back into the original plaintext by using a decryption algorithm [2]. 1.2 What is Chaotic Systems? Chaos communications is application of chaos theory which is to provide security in the transmission of infor- mation. Chaos theory [2], [5], [9] describes the behavior of cer- tain nonlinear dynamic system that under specific conditions exhibit dynamics that are sensitive to initial conditions. Prop- erties of chaotic systems are the sensitivity to initial condi- tions. A large number of uncorrelated, random, deterministic signals can be generated by changing initial values to system. These sequences generated are called chaotic sequences. One of the simplest and most widely studied nonlinear dynamic chaos systems is the logistic map. 1.3 What is ANN? A neural network [2], [12] is a machine that is designed to represent the way in which brain performs any particular task. The network is implemented by using electronic components or is simulated in software on a digital computer. The input images are encrypted with an encryption function on Artificial Neural Network and that images are decrypted using another ANN at receiver side in order to attain the initial image. 2 PROPOSE SYSTEM There are many algorithms to encrypt image data may be simple symmetric key encryption algorithms or chaotic based or ANN based, each of them have their own advantages and limitations. So in proposed system we have combined these two approaches- chaotic cryptosystems and ANN based cryp- tosystems to make CNN--chaotic based ANN systems. Now why these two systems only that review research has already been done by us and available in our review paper [2]. Here is architecture of proposed system. Fig. 1. At sender side IJSER