Abstract
Digital watermarking research has generally focused upon two classes of watermarks, fragile and robust. Robust watermarks
are designed to be detected even after attempts are made to remove them. Fragile watermarks are used for authentication
purposes and are capable of detecting even minute changes of the watermarked content. Unfortunately, neither type of
watermark is ideal when considering "information preserving" transformations (such as compression) which preserve the
meaning or expression of the content and "information altering" transformations (such as feature replacement) which change
the expression of the content. In this paper we describe a semi-fragile watermark for still images that can detect information
altering transformations even after the watermarked content is subjected to information preserving alterations.
