Non Oblivious Watermarking Technique for JPEG2000 Compressed images using Arnold Scrambling of Unequal Size Watermark Blocks

Geeta Kasana (Computer Science and Engineering Department Thapar University Patiala)
Kulbir Singh (Associate Professor Electronics and Communication Engineering Department Thapar University Patiala)
Satvinder Singh Bhatia (Professor School of Mathematics Thapar University Patiala)

International Journal of Science and Engineering (IJSE)
Vol 9, No 1 (2015)

In this paper, a watermarking technique for JPEG2000 compressed image is proposed. Scrambling of secret message is performed block-wise using Arnold Transform. Secret message is divided into non-overlapping blocks of unequal size and then Arnold transform is applied on each block and secret key is generated based on the periodicity of each block. Scrambled secret message is embedded into qualified significant wavelet coefficients of a cover image. After embedding the secret message into wavelet coefficients, the remaining processes of JPEG2000 standard are executed to compress the watermarked image at different compression rates. Scaling Factor (SF) is used to embed watermark into wavelet coefficients and the value of SF is stored into COM box of the code stream of JPEG2000 compressed image and this SF value and secret key are used to extract the embedded watermark on the receiver side. The performance of the proposed technique is robust to a variety of attacks like image cropping, salt and pepper noise, and rotation. Proposed technique is compared with the existing watermarking techniques for JPEG2000 compressed images to show its effectiveness.