Closed form non-iterative watermark embedding
The present disclosure relates generally digital watermarking and data hiding techniques. One claim recites a method comprising: obtaining data representing imagery; using one or more configured processors, transforming the data into a multi-channel color space; obtaining information indicating an impact of adding a color channel encoded signal to a color direction of the data; determining a color space direction for encoded signal detection; obtaining weighting factors based on the information indicating an impact of adding a color channel encoded signal; weighting the color channel encoded signal with weighting factors to yield a modified color channel encoded signal; using one or more configured processors, embedding the modified color channel encoded signal in the data representing imagery. Of course, other claims and combinations are provided as well.
1. A method comprising:
obtaining data representing imagery;
transforming the data representing imagery into a Lab color space;
determining a visibility change for local image areas attributable to adding a color channel encoded signal, said determining a visibility change utilizing a luminance and a color contrast sensitivity function associated with the transformed data in the Lab color space, in which said determining a visibility change yields a determined visibility change, in which said determining utilizes an ellipsoid model of the Lab space, the ellipsoid model comprising a larger radius in the a-b plane relative to a radius in the L plane;
determining luminance (L), a channel (a), and b channel (b) embedding weights to minimize the determined visibility change, in which said determining (L), (a), and (b) embedding weights yields (L), (a), and (b) determined embedding weights, in which embedding weights for the a channel and b channel are determined utilizing an approximation of an intersection of a line through the ellipsoid model origin and an ellipse in the a-b plane; and
applying the (L), (a), and (b) determined embedding weights per pixel in an embedding color space, the (L), (a), and (b) determined embedding weights collectively conveying the encoded signal within the imagery, said applying yielding altered imagery comprising the encoded signal.
2. The method of claim 1 in which said transforming utilizes profiles including sets of data to represent image information.
3. The method of claim 1 in which the data representing imagery comprises data representing pixel information, in which said applying comprises altering the data representing pixel information.
4. The method of claim 1 , and prior to said applying, further comprising transforming the (L), (a), and (b) determined embedding weights into the embedding color space.
5. The method of claim 1 in which the encoded signal comprises digital watermarking.
6. An apparatus comprising:
memory for storing data representing imagery;
one or more electronic processors configured for:
transforming the data representing imagery into a Lab color space;
determining a visibility change for local image areas attributable to adding a color channel encoded signal, the determining a visibility change utilizing a luminance and a color contrast sensitivity functions associated with the transformed data in the Lab color space, in which the determining a visibility change yields a determined visibility change, in which the determining a visibility change utilizes an ellipsoid model of the Lab space, the ellipsoid model comprising a larger radius in the a-b plane relative to a radius in the L plane;
generating luminance (L), a channel (a), and b channel (b) embedding weights to minimize the determined visibility change, in which the generating (L), (a), and (b) embedding weights yields (L), (a), and (b) generated embedding weights, in which embedding weights for the a channel and the b channel are generated with an approximation of an intersection of a line through the ellipsoid model origin and an ellipse in the a-b plane; and
altering the data representing imagery with the (L), (a), and (b) generated embedding weights in an embedding color space, the (L), (a), and (b) generated embedding weights collectively conveying the encoded signal within the imagery, the altering yielding altered imagery comprising the encoded signal.
7. The apparatus of claim 6 in which the transforming utilizes profiles including sets of data to represent image information.
8. The apparatus of claim 6 in which said one or more electronic processors comprise multi-core electronic processors.
9. The apparatus of claim 6 in which the data representing imagery comprises data representing pixel information, in which the altering comprises transforming the data representing pixel information.
10. The apparatus of claim 6 , and prior to the altering, further comprising transforming the (L), (a), and (b) generated embedding weights into the embedding color space.
11. The apparatus of claim 6 in which the encoded signal comprises digital watermarking.
12. A non-transitory computer readable medium on which is stored instructions, which when executed by one or more electronic processors, perform a method comprising:
obtaining data representing imagery;
transforming the data representing imagery into a Lab color space;
determining a visibility change for local image areas attributable to adding a color channel encoded signal, said determining a visibility change utilizing a luminance and a color contrast sensitivity function associated with the transformed data in the Lab color space, in which said determining a visibility change yields a determined visibility change in which the determining utilizes an ellipsoid model of the Lab space, the ellipsoid model comprising a larger radius in the a-b plane relative to a radius in the L plane;
determining luminance (L), a channel (a), and b channel (b) embedding weights to minimize the determined visibility change, in which said determining (L), (a), and (b) embedding weights yields (L), (a), and (b) determined embedding weights, in which embedding weights for the a channel and the b channel are determined with an approximation of an intersection of a line through the ellipsoid model origin and an ellipse in the a-b plane; and
applying the (L), (a), and (b) determined embedding weights per pixel in an embedding color space, the (L), (a), and (b) determined embedding weights collectively conveying the encoded signal within the imagery, said applying yielding altered imagery comprising the encoded signal.
13. The non-transitory computer readable medium of claim 12 in which the transforming utilizes profiles including sets of data to represent image information.
14. The non-transitory computer readable medium of claim 12 in which the data representing imagery comprises data representing pixel information, and in which the applying comprises altering the data representing pixel information.
15. The non-transitory computer readable medium of claim 12 , on which is stored instructions, which when executed by the one or more electronic processors, perform, prior to the applying, transforming the (L), (a), and (b) determined embedding weights into the embedding color space.
16. The non-transitory computer readable medium of claim 12 in which the encoded signal comprises digital watermarking.