With the fast evolution of digital data exchange and increased usage of multi media images, it is essential to protect the confidential image data from unauthorized access. In natural images the values and position of the neighbouring pixels are strongly correlated. The method proposed in this paper, breaks this correlation increasing entropy of the position and entropy of pixel values using block shuffling and encryption by chaotic sequence respectively. The plain-image is initially row wise shuffled and first level of encryption is performed using addition modulo operation. The image is divided into blocks and then block based shuffling is performed using Arnold Cat transformation, further the blocks are uniformly scrambled across the image. Finally the shuffled image undergoes second level of encryption by bitwise XOR operation, and then the image as a whole is shuffled column wise to produce the ciphered image for transmission. The experimental results show that the proposed algorithm can successfully encrypt or decrypt the image with the secret keys, and the analysis of the algorithm also demonstrates that the encrypted image has good information entropy and low correlation coefficients.