An information hiding algorithm of gray images based on suboptimum weighted matrix

In order to promote the imperceptibility of the host image after it was embedded the secret information, a gray image was divided into some blocks, each with n pixels. The upper bound capacity of each block was analyzed under the condition that not more than 2 pixels were modified and just the least significant bit was changed in every modified pixel. It was proved that such an optimum weighted matrix couldn't be constructed. Therefore a way to construct a suboptimum weighted matrix, which was a natural sequence, was proposed. The size of the modified pixel was enlarged certain times accordingly by the suboptimum matrix, thus the most hiding capacity could be brought by the least modification. At the same time, this method could be used in color images. Compared with the optimum weighted matrix mentioned above, one less bit could be hidden by the suboptimum weighted matrix in the worst condition that n was not more than 25 million. The better balance between the hiding capacity and the imperceptibility was reached.