Digital watermarking technique using dual tree complex wavelet transform

With the growing advent of data in day to day life, prevention of illegal data distribution is becoming very important. Digital watermarking is an effective way of ensuring the data is not distributed illegally and its being used effectively to stop the malpractices. Hence, the watermark needs to be imperceptible as well as robust from attacks like rotation, cropping and compression. We propose a method of digital watermarking using a novel technique of Dual Tee- Complex Wavelet Transform. The watermark is embedded into the Chrominance levels by converting the RGB image into YUV channels. This is done because the chrominance levels are less perceptible to the human visual system. A perceptual mask is generated using a predefined low pass filter, which adds to the imperceptibility of the watermark. During the watermark extraction process, the inverse of the perceptual mask generated is used. By using DTCWT it is possible to exploit features like directionality and shift invariance which is not possible in other frequency domain techniques like DWT, DCT and DFT.