Reversible Color Image Watermarking Scheme Based on Contourlet Transform and Principal Component Analysis

It became an important issue to protect the ownership and prevent unauthorized access. A blind and reversible watermarking technique is introduced in this work. The watermark will be generated from the host image and owner identity. Gray Level Co-occurrence Matrix can be used for key generation. Contourlet Transform is used to partition the image into multiple sub-bands. Principal Component Analysis is used to identify the appropriate bands that can be used for embedding the watermark. For the extraction of watermark, secret key and ownership id are required. This technique ensures superior performance in contrast with other techniques while calculating Bit Error Rate and Peak Signal to Noise Ratio.