Optimal Distortion Estimation for Prediction Error Expansion Based Reversible Watermarking

详细信息    查看全文

• 刊名：Lecture Notes in Computer Science
• 出版年：2017
• 出版时间：2017
• 年：2017
• 卷：10082
• 期：1
• 页码：265-279
• 丛书名：Digital Forensics and Watermarking
• ISBN：978-3-319-53465-7
• 卷排序：10082

文摘

Reversible Image Watermarking is a technique to losslessly embed and retrieve information (in the form of a watermark) in a cover image. Prediction Error Expansion based schemes are currently the most efficient and widely used reversible image watermarking techniques. Estimation of the minimum achievable (optimum) distortion for a given payload and a given cover image, is an important problem for reversible watermarking schemes. In this paper, we first show that the bounded capacity distortion minimization problem for prediction error expansion based reversible watermarking schemes is NP-hard, and that the corresponding decision version of the problem is NP-complete. We then propose a low computational overhead heuristic to estimate the minimal distortion. Our estimation technique first estimates (for a given image) the prediction error distribution function without explicit use of any prediction scheme, and then minimizes the distortion for a given payload by solving a convex optimization problem to estimate an embedding threshold parameter. Experimental results for common benchmark images closely match the predictions from our technique, and thus verify the consistency of our approach.