Media signal filtering for use in digital watermark reading

• US 7,076,082 B2
• Filed: 11/18/2002
• Issued: 07/11/2006
• Est. Priority Date: 12/18/2000
• Status: Expired due to Term

• Current Assignee: Digimarc Corporation
• Original Assignee: Digimarc Corporation
• Inventors: Sharma, Ravi K.
• Primary Examiner(s): Johns, Andrew W.
• Application Number:
  US10/299,244
• Publication Number:
  US 20030072468A1
• Time in Patent Office: 1,331 Days
• Field of Search: 382/100, 382/232, 382/260, 382/261, 382/262, 382/263, 382/264, 382/265, 380/210, 380/287, 380/713, 713/176
• US Class Current: 382/100
• CPC Class Codes: G06T 1/0064 Geometric transfor invariant watermarking, e.g. affine transform invariant
• CPC Class Codes: G06T 2201/0052 Embedding of the watermark in the frequency domain
• CPC Class Codes: G10L 19/018 Audio watermarking, i.e. embedding inaudible data in the audio signal
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence
• CPC Class Codes: H04N 1/32144 embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
• CPC Class Codes: H04N 1/32154 Transform domain methods H04N1/32309 takes precedence
• CPC Class Codes: H04N 1/32352 Controlling detectability or arrangements to facilitate detection or retrieval of the embedded information, e.g. using markers for decoding, detection or retrieval operations H04N1/32149 and subgroups take precedence

Abstract

A curve fitting method is used to synchronize a reader of embedded data in a media signal. A circular reference signal is embedded in the media signal. After geometric distortion of the media signal, the reference signal is distorted, yet still detectable. The amount of distortion is derived by detecting the reference signal, and applying a curve fitting method from which the distortion is calculated.

Claims

1. A filtering method for reading a digital watermark comprising:

applying a filter to an array of samples of a media signal, wherein the filter compares a sample of interest with each of a plurality of neighboring samples, for each comparison generates a result based on the comparison, sums the result, and replaces the sample of interest with the sum; and

decoding a digital watermark from the filtered media signal.

2. The method of claim 1 wherein the samples comprise transform domain samples of the media signal.

3. The method of claim 1 wherein the comparison is implemented using a look up table.

4. The method of claim 1 wherein the plurality of neighboring samples includes samples in a neighborhood around the sample of interest outside of immediately adjacent samples of the sample of interest.

5. The method of claim 1 wherein the decoding includes detecting the digital watermark in frequency domain characteristics of the filtered media signal.

6. The method of claim 1 wherein the filter is used to estimate the digital watermark, and the decoding extracts a hidden message carried in the digital watermark.

7. The method of claim 1 wherein the filter is used to estimate the digital watermark, and the digital watermark comprises a pseudorandom signal.

8. The method of claim 1 wherein the comparison comprises a difference of the sample of interest and a neighboring sample.

9. A filtering method for reading a digital watermark comprising:

applying a filter to an array of samples of a media signal, wherein the filter computes functions of a sample of interest and a plurality of neighboring samples to generate a plurality of results corresponding to the functions applied to the sample of interest and the neighboring samples, combines the results into an output value, and replaces the sample of interest with the output value; and

decoding a digital watermark from the filtered media signal.

10. The method of claim 9 wherein the function is implemented using a look up table.

11. The method of claim 9 wherein the plurality of neighboring samples includes samples in a neighborhood around the sample of interest beyond immediately adjacent samples of the sample of interest.

12. The method of claim 9 wherein the decoding includes detecting the digital watermark in frequency domain characteristics of the filtered media signal.

13. The method of claim 9 wherein the filter is used to estimate the digital watermark.

14. The method of claim 9 wherein the filter is used to estimate the digital watermark and the digital watermark comprises a pseudorandom signal.

15. The method of claim 9 wherein the functions comprises linear functions of the sample of interest and a neighboring sample.

16. The method of claim 9 wherein the functions comprise a comparison of the sample of interest and a neighboring sample.

17. The method of claim 9 wherein the functions comprise a difference of the sample of interest and a neighboring sample.

18. The method of claim 9 wherein combining the results comprises summing the results.

19. A pre-filter for processing a media signal before decoding a digital watermark comprising:

an operator for computing a function of a sample of interest and each of selected neighboring samples of the media signal to generate a plurality of results; and

a summer for summing the results into an output value;

the output value replacing the sample of interest;

wherein the digital watermark is decoded from output values computed by applying the filter to the media signal.