Signal processing to hide plural-bit information in image, video, and audio data
AssigneeDigimarc Corporation
InventorGeoffrey B. Rhoads
Life of this patent
Estimating the term…
Abstract
An identification code signal is impressed on a carrier to be identified (such as an electronic data signal or a physical medium) in a manner that permits the identification signal later to be statistically discerned and the carrier thereby identified. The method and apparatus are characterized by robustness despite degradation of the encoded carrier, and by holographic permeation of the identification signal throughout the carrier.
Claims27
In an image processing method that includes steganographically encoding an input two-dimensional image to embed a multi-bit message code therein, the image comprising a plurality of rows of picture data, the method including processing said input image in accordance with said multi-bit message code to produce an output image having the message code steganographically encoded therein, an improvement including:
- encoding the message code throughout the output image so that the message code can be recovered from first and second non-overlapping excerpts of the output image;
- and representing a given bit of the encoded message code differently, both in absolute and percentage terms, in said first and second excerpts to reduce image degradation and increase message security;
- wherein subtraction of the input two-dimensional image from the output image yields a difference frame comprised of difference values, said difference frame having a snow-like appearance in which at least half of said difference values have non-zero values.
In an image processing method that includes steganographically encoding an input image signal to embed a multi-bit message code therein, the input image signal including a plurality of picture data, the method including processing said input signal in accordance with said multi-bit message code to produce an output signal having the message code steganographically encoded therein, an improvement including:
- encoding the message code throughout the output signal so that the message code can be recovered from first and second non-overlapping excerpts of the output signal, said encoding including changing a plurality of said picture data in accordance with said multi-bit message code, and also in accordance with local scaling data to control the amplitude of the multi-bit message code at different locations in the image, said local scaling data being computed from said picture data using a non-linear function.
In an audio processing method that includes steganographically encoding an input signal to embed a multi-bit message code therein, the input signal including a plurality of audio data, the method including processing said input signal in accordance with said multi-bit message code to produce an output signal having the message code steganographically encoded therein, an improvement including:
- encoding the message code throughout the output signal so that the message code can be recovered from first and second non-overlapping excerpts of the output signal, said encoding including changing a plurality of said audio data in accordance with said multi-bit message code, and also in accordance with local scaling data to control the amplitude of the multi-bit message code at different locations in the audio data, said local scaling data being computed from said audio data using a non-linear function.
In an image processing method that includes steganographically encoding an input two-dimensional image to embed a multi-bit message code therein, the image being represented by plural elements, the method including processing said input image in accordance with said multi-bit message code to produce an output image having the message code encoded therein, an improvement including:
- encoding the message code throughout the output image so that the message code can be recovered from first and second non-overlapping rectangular excerpts of the output image when same is represented in the pixel domain;
- and representing a given bit of the encoded message code differently, both in absolute and percentage terms, in said first and second excerpts to reduce image degradation and increase message security, said encoding including changing a plurality of said elements in accordance with said multi-bit message code, and also in accordance with local scaling data to control the amplitude of the multi-bit message code at different locations in the image, said local scaling data being computed from said picture elements using a non-linear function.
A tangible storage medium having an image processed in accordance with the method of claim 19 stored thereon.
Dependencies are inferred from the claim text.