Countering ambiguity attacks against digital image watermarking schemes

Abstract

The ambiguity attack, or invertibility attack, was discovered in 1998 as a simple but a powerful attack against digital watermarking schemes. Identifying the potential threat of this attack, a number of ambiguity attack resistant watermarking schemes have been proposed in the past literature. However, later on some of these schemes were proven to be failing under the ambiguity attack. In this paper, we study the concept of the ambiguity attack, evaluate different ambiguity attack resistant watermarking schemes and the scenarios under which they fail to provide necessary security against ambiguity attacks. We employ the secure spread spectrum method as the underlying watermarking scheme to implement and evaluate these ambiguity attack resistant watermarking schemes.