Reversible Watermarking And Its Application In Multi-view Images

Digital watermarking is a technique which can embed the watermark into the digital carriers directly.The difference between the original and the watermarked digital carriers is not perceived by the human vision,thus the watermarked carriers can be used normally.Reversible data hiding is an important branch of digital watermarking.From the watermarked image,it can not only extract the watermark without distortion,but also exactly recover the carrier image,which makes it possible to be applied for the field that the quality of the image must be high,such as medical images,military images and remote sensing images.Multi-view images can provide more natural visual sensation than the traditional images,so they have wide application prospects.This paper introduces some methods about reversible watermarking and the problem in current research in detail.And inspired by the prior works,three novel reversible watermarking algorithms are proposed.(1)A novel reversible watermarking algorithm based on prediction error expansion is proposed.In this paper,we propose a novel reversible data hiding method in the depth map for quality enhancement.The hidden data are the compressed residual error between the actual view and the virtual view.By optimizing the selection of prediction error and embedding data with the low distortion method,the proposed scheme can embed more data with less distortion and the virtual view with higher quality is obtained.(2)A reversible watermarking algorithm based on controllable interval histogram diffusion is proposed.Since a large portion of the depth map is often homogeneous,most prediction errors assume the value 0.In this metlhod,we select the pixel whose prediction error equals zero to embed the watermark by histogram diffusion,which can improve the embedding capacity.A novel overhead embedding method based on least significant bit(LSB)is proposed.Compared with the previous algorithm,the proposed algorithm can obtain higher embedding capacity.(3)A reversible watermarking algorithm for depth map based on saliency is proposed.The residual information is compressed according to the saliency of the virtual view image,and the compression parameters are determined automatically by the K-means classification method.The residual information corresponding with higher saliency use larger quality factor and the residual information corresponding with lower saliency use smaller quality factor.The experimental results show that the synthesized virtual view has higher quality and better subjective visual perception.