Data hiding is a way of hiding secret data on cover-media and it is used for a variety of applications. An important of the data hiding is to conceal the data in a secret way without loss of cover-media. Until now, continuous research on absolute moment block truncation coding based data hiding methods have improved a performance on data concealment and image quality. However, the current absolute moment block truncation coding based data hiding technology has a limitation in deriving a method that significantly surpasses existing performance. In this paper, we propose a new method to overcome this problem. To do this, first the original image is transformed to the cover image using absolute moment block truncation coding and is expanded the image using neighbor average interpolation algorithm. The proposed three data hiding methods are based on the generated cover image. The first method is to directly replace the pixel value, which is a component of each block, with the same secret value. The second method is to replace the pixels to match the secret bits only for the extended pixels in each block of the cover image. The third method is to apply Hamming code to each block to minimize the number of replacement pixels for data hiding. Experimental results show that our method is superior in terms of efficiency compared to traditional absolute moment block truncation coding based data hiding methods.
Keywords: AMBTC; BTC; Data Hiding; Hamming code; Neighbor Mean Interpolation.