Eigen Value Decomposition Utilizing Method for Data Hiding Based on Wavelet Multi-resolution Analysis

Eigen value decomposition utilizing method for data hiding based on wavelet Multi-Resolution Analysis (MRA) is proposed. It is possible to improve an invisibility of the hidden information by using eigen value decomposition as a preprocessing of the conventional wavelet Multi-Resolution Analysis (MRA) based method. In the proposed method, the information of the key image is protected by the existence of the eigenvector. That is, the key image information can be restored only if the true original image information is known. The coefficient of eigen value decomposition differs for each original image and is composed of the eigenvectors of the original image. In a 3-band color image, a method involving Hue Saturation Intensity (HIS) conversion, or the like can be considered for the purpose of protecting the information of the key image, but the conversion coefficient such as HSI conversion is a well-known coefficient. Since the conversion coefficient of the method involving HSI conversion is well known, there is a possibility that a third party may obtain the information of the key image. The effectiveness of the proposed method is confirmed from the viewpoint of protecting the information of the key image.