Detecting electronic watermarks in digital video

Electronic watermarking is an active area of research with many applications being foreseen. Watermarks may become an essential tool for copy management in future consumer electronic or PC devices. With simple circuits, detection of watermarks after noise addition, MPEG compression, D/A conversion, pixel shifts appears feasible, but detection after transformations, such as cropping and stretching, remains a challenge. We propose a model to evaluate the effect of scaling on the detector reliability and verify it with experiments.