Blind and robust images watermarking based on wavelet and edge insertion

TL;DR

This paper introduces a blind, wavelet-based image watermarking method that inserts watermarks via edge addition in the HH sub-band, demonstrating superior robustness against various attacks and adjustable parameters for flexibility.

Contribution

It presents a novel blind watermarking scheme using edge insertion in the wavelet domain, enhancing robustness and flexibility over existing methods.

Findings

Robust against JPEG and GIF compression

Effective against noise and small rotations

Flexible capacity and robustness adjustment

Abstract

This paper gives a new scheme of watermarking technique related to insert the mark by adding edge in HH sub-band of the host image after wavelet decomposition. Contrary to most of the watermarking algorithms in wavelet domain, our method is blind and results show that it is robust against the JPEG and GIF compression, histogram and spectrum spreading, noise adding and small rotation. Its robustness against compression is better than others watermarking algorithms reported in the literature. The algorithm is flexible because its capacity or robustness can be improved by modifying some parameters.