Patent №
US 6,636,616
Granted
2003-10-21
Filed 1999
Owner
XEROX CORPORATION
Lab
—
AI components
1
hardware
Assignment
Recorded
Dataset
AIPD
2023_r1 edition
Application
09451994
A method is described to imbed hidden information in output images that are rendered using error diffusion. The idea is to introduce a bias in the dot placements and then modulate that bias with the pattern to be hidden. The method particularly involves the steps of rendering the output copy using an error diffusion routine that parses through a plurality of stored pixels representing an image; modifying the error diffusion routine with a biasing pattern that influences where on the output copy each of the plurality of pixels is rendered; and further modifying the error diffusion routine with a watermark pattern that modifies the biasing pattern for pixels that are positioned within the watermark.
AI classification
Ownership
XEROX CORPORATION
assignment · 111430954
Assignors
HARRINGTON, STEVEN J.
On an employer assignment, the assignors are typically the inventors.