Abstract
In accordance with the property of Human Visual System (HVS) that is insensitive to edged and dark areas, a novel steganographic solution based on self-organizing competitive NNs and wavelet is proposed. An image is divided into many small blocks. Transform each block into one-level wavelet in order to obtain the wavelet contrast and the count of nonzero wavelet coefficients. Classify all blocks by self-organizing NNs and embed secret information with steganography based on modulus. The experiments result in that the proposed method can hide much more information while the vision of original stego-image is good remained.