Extracting digital watermarks using logarithmic sampling and symmetrical attributes

摘要

A digital watermark reader performs a logarithmic sampling of a watermarked media signal to produce a sampled signal. It then analyzes the sampled signal to detect attributes associated with a watermark signal. It extracts a digital watermark based on detected attributes associated with the watermark signal. As an alternative or in addition to these operations, the watermark reader performs a polar sampling of the media signal. For example, polar sampling may be used to address rotational distortion. Another digital watermark reader transforms the media signal to a transform domain, and analyzes the transformed media signal to detect a symmetrical attribute associated with a watermark signal. The method extracts the digital watermark based on detected attributes associated with the watermark signal.