Method and apparatus for scene-based video watermarking
AssigneeRegents of the University of Minnesota
InventorsAhmed H. Tewfik, Mitchell D. Swanson, Bin Zhu
Life of this patent
Estimating the term…
Abstract
A method and apparatus for the scene-based watermarking of video data is disclosed. In one embodiment, each of a number of frames of a scene of video host data undergoes a temporal wavelet transform, from which blocks are extracted. The blocks undergo perceptual masking in the frequency domain, such that a watermark is embedded therein. Once the watermark block is taken out of the frequency domain, a spatial mask of the original block is weighted to the watermark block, and added to the original block to obtain the watermarked block.
Claims14
A computerized method for embedding data representing a watermark into host data relating to video:
- generating the data representing the watermark;
- subjecting the host data to a temporal wavelet transform;
- embedding the data into the host data, as subjected to the temporal wavelet transform, in accordance with a perceptual mask conducted in the frequency domain;
- and,
- subjecting the host data, including the data embedded therein, to a non-frequency mask.
A scene-based computerized method of watermarking host data relating to video comprising:
- segmenting the host data into a plurality of scenes, each scene having a plurality of frames;
- subjecting each frame of each scene to a temporal wavelet transform;
- and,
- embedding each frame of each scene, as has been subjected to the temporal wavelet transform, with a watermark in accordance with perceptual and spatial characteristics of the frame.
A computerized system for watermarking host data relating to video and having a plurality of scenes, each scene having a plurality of frames, comprising:
- a processor;
- a computer-readable medium;
- computer-executable instructions executed by the processor from the computer-readable medium comprising:
- applying a temporal wavelet transform to each frame;
- segmenting each frame of each scene into blocks;
- applying a discrete cosine transform (DCT) to each block to generate a frequency block corresponding to the block;
- generating a perceptual mask for each frequency block;
- applying the DCT to a watermark for each frequency block;
- weighting the perceptual mask for each frequency block with the watermark for the frequency block to which the DCT has been applied to generate a frequency-shaped author block;
- applying an inverse DCT to each frequency-shaped author block to generate a time-domain block;
- generating a spatial mask for each block;
- weighting each time-domain block by a spatial mask to generate a watermark block;
- and,
- adding each block to a corresponding watermark block to generate a watermarked block.
A computer-readable medium having a computer program stored thereon to cause a suitable equipped computer to perform a method comprising:
- applying a temporal wavelet transform to each frame;
- segmenting each frame of each scene into blocks;
- applying a discrete cosine transform (DCT) to each block to generate a frequency block corresponding to the block;
- generating a perceptual mask for each frequency block;
- applying the DCT to a watermark for each frequency block;
- weighting the perceptual mask for each frequency block with the watermark for the frequency block to which the DCT has been applied to generate a frequency-shaped author block;
- applying an inverse DCT to each frequency-shaped author block to generate a time-domain block;
- generating a spatial mask for each block;
- weighting each time-domain block by a spatial mask to generate a watermark block;
- and,
- adding each block to a corresponding watermark block to generate a watermarked block.
Dependencies are inferred from the claim text.