ABSTRACT
Digital watermarking can be used for tracing images that have been illegally distributed. There are two major steps in the digital watermarking process: watermark embedding, in which a watermark is inserted into a host image, and watermark extraction, in which the watermark is pulled out of the image. Because there are a great number of watermarking algorithms being developed, it is important to define some criteria for classifying them to understand how different schemes can be applied. This chapter describes classification based on characteristics, and presents classification based on applications. The classification based on characteristics categorizes digital watermarking technologies into five classes according to the characteristics of embedded watermarks. The five classes are blind versus nonblind, perceptible versus imperceptible, private versus public, robust versus fragile and spatial domain based versus frequency domain based. The classification based on applications categorizes digital watermarking technologies into five classes according to the following applications: copyright protection, data authentication, fingerprinting, copy control and device control.
