Watermarking in Safe Region of frequency domain using complex-valued Neural Network

It has been discovered by computational experiments that Complex Back-Propagation (CBP) algorithm is well suited for learning complex pattern, and it has been reported that this ability can successfully be applied in image processing with complex values. In this paper, a watermarking scheme based on Complex-Valued Neural Network, CVNN trained by CBP in transform domain is proposed. Fast Fourier Transform, FFT is used to obtain the complex values (real and imaginary part) of the host image. The complex values form the input data of CVNN. Neural networks performs best on detection, mapping, classification, learning and adaption. These features are employed to simulate the Safe Region (SR) to embed the watermark, thus, watermark are appropriately mapped to the safe region of selected coefficients. The implementation results have shown that this watermarking algorithm has high level of imperceptibility.