Reversible Image Watermarking Using Interpolation Technique

Abstract

Watermarking embeds information into a digital signal like image. Reversible image watermarking can restore the original image without any distortion after the hidden data is extracted. Texts are embedded using by Rijndael Algorithm and Images are embedded using by SNDR (Signal to Noise Disortion Ratio). Here present a novel reversible watermarking scheme using an interpolation technique, which can embed a large amount of convert data into images or image into image with imperceptible modification. Then utilize the interpolation-error, the difference between interpolation value and corresponding pixel value, to embed bit ??1?? or ??0?? by expanding it additively or leaving it unchanged. Image embedded function will take Target Image and Cover Image as input. It will output the encoded watermarked image. Due to the slight modification of pixels, high image quality and text quality are preserved. Experimental results also demonstrate that the proposed scheme can provide greater payload capacity and higher image fidelity compared with other state-of-the-art schemes.

Existing System

 Since the earliest reversible watermarking scheme has complex computation and limited capacity.  Here, we can obtain the interpolation-errors just through calculating the difference between interpolation values and original values.  However, a quarter of pixels, which are exploited to reconstruct the image, cannot be utilized to embed the data.  Consequently, the capacity of the algorithm is constrained, which is a problem we must deal with.

Proposed System

 Reversible watermarking scheme based on additive interpolation-error expansion, which features very low distortion and relatively large capacity have proposed.  Different from previous watermarking schemes, an interpolation technique was utilized to generate residual values named interpolation-errors and expand them by addition to embed bits. The strategy is efficient since interpolation-errors are good at decorrelating pixels and additive expansion is free of expensive overhead information.  Different from the latest schemes using prediction or wavelet techniques, the proposed scheme uses an interpolation technique to generate residual values named interpolation-errors, which are demonstrated to be of greater decorrelation ability.

Architecture