System and method of watermarking a video signal and extracting the watermarking from a video signal

• US 7,415,127 B2
• Filed: 10/31/2003
• Issued: 08/19/2008
• Est. Priority Date: 10/31/2002
• Status: Expired due to Fees

• Current Assignee: Orange S.A.
• Original Assignee: Orange S.A.
• Inventors: Laurent, Nathalie, Bodo, Yann
• Primary Examiner(s): Wu; Jingge
• Assistant Examiner(s): Krasnic; Bernard
• Application Number:
  US10/697,106
• Publication Number:
  US 20040175017A1
• Time in Patent Office: 1,754 Days
• Field of Search: 382/100, 382/232, 382/107
• US Class Current: 382/100
• CPC Class Codes: G06T 1/0085 Time domain based watermarking, e.g. watermarks spread over several images
• CPC Class Codes: G06T 2201/0053 Embedding of the watermark in the coding stream, possibly without decoding; Embedding of the watermark in the compressed domain
• CPC Class Codes: H04N 19/467 characterised by the embedded information being invisible, e.g. watermarking

Abstract

This method comprises the following steps of applying a watermarking function to motion vectors calculated by estimation of movement between images of a video signal, followed by a step of generating the watermarked video signal by compensating movement with the aid of the watermarked motion vectors. The watermarking function is applied by marking the coordinates of the selected motion vector in a reference space, certain portions of this space constituting a first zone and the other portions of this space constituting a second zone complementary to the first, assigning a binary value to each of the two zones, and, if necessary, modifying the coordinates of the selected motion vector so that it is in the zone whose binary value corresponds to a bit of a marking key associated with the selected motion vector.

Claims

1. A method of watermarking a video signal by applying a watermarking function to motion vectors calculated by estimation of movement between images of the video signal, the method comprising the following steps:

applying the watermarking function to at least some of the calculated motion vectors; and

generating the watermarked video signal by compensating movement with the aid of the watermarked motion vectors,

the watermarking function being applied with the aid of a binary marking key, each bit of which is associated with at least one selected motion vector, wherein, in order to apply the watermarking function, the method further comprises the following steps;

marking a point (V) corresponding to the coordinates (Vx, Vy) of the selected motion vector ({right arrow over (V)}), the marking taking place in a reference space (O, x, y) divided into a plurality of predetermined portions;

defining two complementary zones Z₀ and Z₁ in each portion, one of the two zones being situated inside the other one;

assigning a binary value to each of the two zones; and

if the point corresponding to the coordinates of the selected motion vector is in the zone of the portion to which it belongs, of binary value which corresponds to the bit of the marking key with which the selected motion vector is associated, not modifying the coordinates of the selected motion vector;

if the point corresponding to the coordinates of the selected motion vector is not in the zone of the portion to which it belongs, of binary value which corresponds to the bit of the marking key with which the selected motion vector is associated, modifying the coordinates of the selected motion vector so that the point is in the zone, of binary value which corresponds to the bit of the marking key with which the selected motion vector is associated.

2. The method according to claim 1 for watermarking a video signal, wherein the reference space is a reference grid including blocks with predefined dimensions, each block including first and second zones.

3. The method according to claim 1 for watermarking a video signal, the method including the following steps:

calculating a hierarchical plurality of successive levels of motion vectors, the motion vectors of a given level each being associated with a plurality of motion vectors of the next lower level;

selecting at least some of the motion vectors belonging to the highest level;

applying the watermarking function to each selected motion vector, leading to calculating a modification parameter for said motion vector; and

applying the modification parameter of the selected motion vector to the motion vectors of a lower level associated with said motion vector.

4. The method according to claim 3 for watermarking a video signal, wherein the motion vectors of a given level are each equal to the average of the motion vectors of the next lower level with which they are associated.

5. The method according to claim 3 for watermarking a video signal, the method including a step of calculating a hierarchy of two successive levels of motion vectors, each motion vector of the higher level being associated with four motion vectors of the lower level.

6. The method according to claim 1 for watermarking a video signal, wherein the first and second zones have substantially equal areas.

7. The method according to claim 2 for watermarking a video signal, wherein a sub-block centered inside the block is defined in each block of the reference grid, the first zone being defined by the interior of the sub-block and the second zone being the zone in the block complementary to the first zone.

8. The method according to claim 7 for watermarking a video signal, wherein the blocks and sub-blocks of the reference grid are rectangular.

9. The method according to claim 1 for watermarking a video signal, wherein the modification of the coordinates of the selected motion vector, if any, is a weighted symmetry applied to the point (V).

10. The method according to claim 8 for watermarking a video signal, wherein the modification of the coordinates of the selected motion vector, if any, is either a weighted central symmetry relative to one of the vertices of the sub-block or a weighted axial symmetry relative to one of the sides of the sub-block applied to the point (V).

11. The method according to claim 1 for watermarking a video signal, wherein each bit of the binary marking key is associated with a plurality of selected motion vectors.

12. The method according to claim 1 for watermarking a video signal, wherein some of the bits of the binary marking key are associated with motion vectors calculated by motion estimation between two images of the video signal, and wherein at least one other portion of the bits of the binary marking key is associated with motion vectors calculated by motion estimation between at least two other images of the video signal.

13. A device for watermarking a video signal, the device including means for implementing the method according to claim 1.

14. A computer readable data medium, including means for storing a video signal watermarked with the aid of the method according to claim 1.

15. A method of extracting watermarking from a video signal watermarked by applying the method according to claim 1, which extraction method comprises applying a function for extracting the binary marking key comprising:

selecting the watermarked vectors;

marking a point corresponding to the coordinates of each watermarked motion vector, the marking taking place in the reference space; and

assigning the binary value of the zone in which the point is situated to the bit of the marking key with which the selected motion vector is associated.

16. The method according to claim 15 for extracting the watermarking from the video signal, the video signal further watermarked by the following steps:

calculating a hierarchical plurality of successive levels of motion vectors, the motion vectors of a given level each being associated with a plurality of motion vectors of the next lower level;

selecting at least some of the motion vectors belonging to the highest level;

applying the watermarking function to each selected motion vector, leading to calculating a modification parameter for said motion vector; and

applying the modification parameter of the selected motion vector to the motion vectors of a lower level associated with said motion vector;

wherein, for each motion vector of the highest level selected on application of the watermarking method, the following steps are applied;

extracting the watermarked motion vectors associated with said motion vector;

calculating an average vector equal to the average of the watermarked motion vectors associated with said motion vector; and

applying the marking key extraction function to the calculated average vector.

17. A device for extracting the watermarking from a video signal, the device including means for implementing the method according to claim 15.

18. The method according to claim 9, for watermarking a video signal, wherein the modification of the coordinates of the selected motion vector, if any, is either a weighted central symmetry relative to one of the vertices of the sub-block or a weighted axial symmetry relative to one of the sides of the sub-block applied to the point (V).