Asymmetric watermark embedding/extraction

• US 8,538,066 B2
• Filed: 09/04/2012
• Issued: 09/17/2013
• Est. Priority Date: 04/26/2005
• Status: Expired due to Fees

• Current Assignee: Verance Corporation
• Original Assignee: Verance Corporation
• Inventors: Petrovic, Rade, Tehranchi, Babak, Winograd, Joseph M., Jemili, Kanaan
• Primary Examiner(s): Allison, Andrae S
• Application Number:
  US13/602,844
• Publication Number:
  US 20130011006A1
• Time in Patent Office: 378 Days
• Field of Search: 382/100, 382/232, 382/294, 380/20, 380/28, 380/54, 380/201, 380/210
• US Class Current: 382/100
• CPC Class Codes: G06T 1/0028 Adaptive watermarking, e.g. Human Visual System [HVS]-based watermarking
• CPC Class Codes: G06T 1/005 Robust watermarking, e.g. average attack or collusion attack resistant
• CPC Class Codes: G06T 1/0071 using multiple or alternating watermarks
• CPC Class Codes: G06T 2201/0202 whereby the quality of watermarked images is measured; Measuring quality or performance of watermarking methods; Balancing between quality and robustness
• CPC Class Codes: G09C 5/00 Ciphering apparatus or methods not provided for in the preceding groups, e.g. involving the concealment or deformation of graphic data such as designs, written or printed messages
• CPC Class Codes: G10L 19/018 Audio watermarking, i.e. embedding inaudible data in the audio signal
• CPC Class Codes: H04L 2209/608 Watermarking
• CPC Class Codes: H04N 1/32203 Spatial or amplitude domain methods H04N1/32309 takes precedence
• CPC Class Codes: H04N 1/32245 Random or pseudorandom selection of pixels
• CPC Class Codes: H04N 1/32299 using more than one embedding method
• CPC Class Codes: H04N 1/32315 Selecting a particular method from amongst a plurality of methods
• CPC Class Codes: H04N 2201/3233 of authentication information, e.g. digital signature, watermark

Abstract

Methods and devices are provided to embed imperceptible watermarks into a host content and to subsequently extract embedded watermarks in a manner so as to achieve desired levels of robustness, transparency and security. For a particular host content, a first number of embedding opportunities are used to insert watermarks, which is less than a number of all embedding opportunities within the host content. Watermarks are extracted from a first number of watermark extraction opportunities, which is less than a number of all watermark extraction opportunities associated with the host content. The relationship between the first number of embedding opportunities, the first number of watermark extraction opportunities and the number of all watermark extraction opportunities is such that a circumvention attempt that overwrites watermarks at all extraction opportunities degrades perceptual quality of the host content below a desired perceptual quality level.

Claims

1. A method, comprising:

receiving a host content, the host content comprising embedded watermarks in a first number of embedding opportunities, wherein the first number of embedding opportunities is less than a number of all embedding opportunities within the host content; and

extracting one or more of the embedded watermarks by evaluating no more than a first number of watermark extraction opportunities associated with the host content, wherein

the first number of watermark extraction opportunities is less than a number of all watermark extraction opportunities associated with the host content, and

the relationship between the first number of embedding opportunities, the first number of watermark extraction opportunities and the number of all watermark extraction opportunities is such that a circumvention attempt that overwrites watermarks at all extraction opportunities degrades perceptual quality of the host content below a desired perceptual quality level.

2. The method of claim 1, wherein the first number of embedding opportunities and the first watermark extraction opportunities are selected to produce a predetermined probability of successful detection of an embedded watermark.

3. The method of claim 1, wherein the first number of extraction opportunities is selected to limit a maximum number of computational operations associated with extracting the one or more embedded watermarks.

4. The method of claim 1, wherein the first number of watermark extraction opportunities and a ratio of the first number of watermark extraction opportunities to the number of all watermark extraction opportunities are selected to produce a particular level of security, and wherein a smaller ratio produces a more secure watermarking system.

5. The method of claim 1, wherein the first number of embedding opportunities and a ratio of the first number of watermark extraction opportunities to the number of all watermark extraction opportunities are selected to simultaneously satisfy a particular probability of successful watermark detection and a desired level of watermark imperceptibility.

6. The method of claim 1, wherein the number of all embedding opportunities and the number of all watermark extraction opportunities associated with the content are different.

7. The method of claim 6, wherein the number of all embedding opportunities is smaller than the number of all watermark extraction opportunities.

8. The method of claim 1, further comprising reporting only a subset of all successfully extracted watermarks from the host content.

9. The method of claim 1, wherein

the first number of watermark extraction opportunities corresponds to a first set of watermark extraction locations within the host content,

the first set of watermark extraction locations is selected from a plurality of sets of watermark extraction locations, and

each of the plurality of sets of watermark extraction locations constitutes a subset of all watermark extraction locations within the host content.

10. The method of claim 9, wherein

the first number of embedding opportunities corresponds to a first set of watermark embedding locations within the host content, and

at least one embedding location from the first set of embedding locations coincides with a watermark extraction location within each of the plurality of sets of watermark extraction locations.

11. A device, comprising:

a processor and a memory comprising software code that configures the device to;

receive a host content, the host content comprising embedded watermarks in a first number of embedding opportunities, wherein the first number of embedding opportunities is less than a number of all embedding opportunities within the host content; and

extract one or more of the embedded watermarks by evaluating no more than a first number of watermark extraction opportunities associated with the host content, wherein

the first number of watermark extraction opportunities is less than a number of all watermark extraction opportunities associated with the host content, and

the relationship between the first number of embedding opportunities, the first number of watermark extraction opportunities and the number of all watermark extraction opportunities is such that a circumvention attempt that overwrites watermarks at all extraction opportunities degrades perceptual quality of the host content below a desired perceptual quality level.

12. The device of claim 11, wherein the first number of embedding opportunities and the first watermark extraction opportunities are selected to produce a predetermined probability of successful detection of an embedded watermark.

13. The device of claim 11, wherein the first number of extraction opportunities is selected to limit a maximum number of computational operations associated with extracting the one or more embedded watermarks.

14. The device of claim 11, wherein the first number of watermark extraction opportunities and a ratio of the first number of watermark extraction opportunities to the number of all watermark extraction opportunities are selected to produce a particular level of security, and wherein a smaller ratio produces a more secure watermarking system.

15. The device of claim 11, wherein the first number of embedding opportunities and a ratio of the first number of watermark extraction opportunities to the number of all watermark extraction opportunities are selected to simultaneously satisfy a particular probability of successful watermark detection and a desired level of watermark imperceptibility.

16. The device of claim 11, wherein the number of all embedding opportunities and the number of all watermark extraction opportunities associated with the content are different.

17. The device of claim 16, wherein the number of all embedding opportunities is smaller than the number of all watermark extraction opportunities.

18. The device of claim 11, further configured to produce a report that includes only a subset of all successfully extracted watermarks from the host content.

19. The device of claim 11, wherein

the first number of watermark extraction opportunities corresponds to a first set of watermark extraction locations within the host content,

the first set of watermark extraction locations is selected from a plurality of sets of watermark extraction locations, and

each of the plurality of sets of watermark extraction locations constitutes a subset of all watermark extraction locations within the host content.

20. The device of claim 19, wherein

the first number of embedding opportunities corresponds to a first set of watermark embedding locations within the host content, and

at least one embedding location from the first set of embedding locations coincides with a watermark extraction location within each of the plurality of sets of watermark extraction locations.