Using Classification Techniques in Digital Watermarking

• US 20100074466A1
• Filed: 03/20/2009
• Published: 03/25/2010
• Est. Priority Date: 06/02/2000
• Status: Active Grant

• Current Assignee: Digimarc Corporation
• Original Assignee: Digimarc Corporation
• Inventors: Brunk, Hugh L.
• Granted Patent: US 7,958,365 B2
• Time in Patent Office: Days
• US Class Current: 382/100
• CPC Class Codes: G06T 1/005 Robust watermarking, e.g. average attack or collusion attack resistant
• CPC Class Codes: G06T 2201/0065 Extraction of an embedded watermark; Reliable detection

Abstract

A classification scheme assigns samples of the watermarked media to classes based on classification criteria indicating a likely presence of a watermarked signal. Once classified, the scheme determines statistical characteristics of the media signal for the classes, and assigns a figure of merit to the samples in each class. A watermark decoder (e.g., reader and/or detector) use the figure of merit to adapt a watermark decoding function to the statistical features of the classes, such as weighting to samples to improve watermark decoding.

Claims

1. A method for reading a digital watermark in a media signal comprising:

using a computer programmed to perform the acts of;

assigning sets of media signal samples into classes;

computing statistical characteristics of the media samples for the classes; and

using the statistical characteristics to adapt a digital watermark decoding function used to detect or read a watermark in the media signal.

2. The method of claim 1 wherein the media signal is an audio signal.

3. The method of claim 1 wherein the media signal is an image signal.

4. The method of claim 3 wherein the image signal samples are expressed in a frequency domain.

5. The method of claim 4 wherein the image signal samples are spatial frequency coefficients.

6. The method of claim 1 wherein the samples are in a spatial or temporal domain.

7. The method of claim 1 wherein using the statistical characteristics includes:

assigning a figure of merit to a sample indicating a likelihood that the sample includes a recoverable portion of a watermark signal; and

using the figure of merit in a read operation.

8. The method of claim 7 wherein assigning a figure of merit includes assigning a weight to the sample indicating an extent to which the sample is likely to reflect valid watermark data.

9. The method of claim 1 wherein using the statistical characteristics includes:

assigning a figure of merit to a sample indicating a likelihood that the sample includes a recoverable portion of a watermark signal; and

using the figure of merit in a watermark decoding operation.

10. The method of claim 9 wherein assigning a figure of merit includes assigning a weight to the sample indicating an extent to which the sample is likely to reflect valid watermark data.

11. A computer readable medium on which is stored instructions, which when executed by a computer, perform the method of claim 1.

12. A method for reading a digital watermark in a media signal comprising:

using a computer programmed to perform the acts of;

assigning transformed samples of the media signal into classes using characteristics computed from the samples to group the samples into the classes;

computing statistical characteristics of the samples in each of the classes to compute a statistical feature of each of the classes; and

using the statistical feature to compute a detection metric to decode a watermark from the samples.

13. The method of claim 12 wherein the characteristics comprise signal activity of the samples, and the signal activity of the samples is evaluated and the samples are assigned to the classes based on signal activity.

14. A computer readable medium on which is stored instructions, which when executed by a computer, perform the method of claim 12.

15. A method for reading a digital watermark in a watermarked signal comprising:

using a computer programmed to perform the acts of;

assigning samples of the watermarked signal into classes using characteristics computed from the samples to group the samples into the classes;

computing a statistical analysis of the samples in each of the classes; and

using the statistical analysis to compute a decoding function for decoding a watermark from the watermarked signal.

16. The method of claim 15 wherein sets of samples are assigned to classes based on a signal characteristic of the samples in the sets.

17. The method of claim 16 wherein the signal characteristic is a measure of signal energy.

18. A computer readable medium on which is stored instructions, which when executed by a computer, perform the method of claim 15.

19. A method for estimating a watermark signal from a media signal suspected of containing the watermark signal, the method comprising:

using a computer programmed to perform the acts of;

assigning samples of the suspect signal into classes based on a signal characteristic of the samples;

calculating statistical characteristics of the classes; and

estimating the watermark signal based on the suspect signal, the statistical characteristics of the classes, and a characteristic of the watermark signal.

20. A computer readable medium on which is stored software for performing the method of claim 19.