A novel audio color watermarking scheme based on self-organizing map

In this paper, a novel audio color watermarking scheme based on self-organizing map (SOM) neural network is proposed. The SOM neural network is introduced for preprocessing color watermark image and producing a codebook on which a series of operations are performed to obtain the hidden information. Without the perceivable distortion, such hidden information is embedded in the approximate wavelet coefficient of the host audio signal by a linear instantaneous mixing model. The process of extracting hidden information from the marked audio signal, which is the reverse process of the watermark embedding, is fulfilled by the inverse matrix of the linear instantaneous mixing matrix. The experimental results demonstrate the validity of the proposed color watermarking scheme.