Apparatus and method for inserting and detecting watermark based on stochastic model

• US 6,853,676 B2
• Filed: 12/14/2000
• Issued: 02/08/2005
• Est. Priority Date: 06/15/2000
• Status: Active Grant

• Current Assignee: Korea Telecom (KT Corporation)
• Original Assignee: Korea Telecom (KT Corporation)
• Inventors: Kim, Hongseok
• Primary Examiner(s): Patel, Jayanti
• Assistant Examiner(s): GOSHTASBI, JAMSHID
• Application Number:
  US09/735,922
• Publication Number:
  US 20020138730A1
• Time in Patent Office: 1,517 Days
• Field of Search: 382/100, 380/207, 380/54, 380/238, 713/176, 375/130, 375/150, 375/216, 375/240, 375/240.18, 375/340
• US Class Current: 375/130
• CPC Class Codes: G10L 19/018 Audio watermarking, i.e. embedding inaudible data in the audio signal
• CPC Class Codes: G11B 20/00086 Circuits for prevention of unauthorised reproduction or copying, e.g. piracy indicating unauthorised use of record carriers in general G11B23/28; scrambling for television signal recording H04N5/913; network architectures or network protocols for network security H04L63/00; cryptographic mechanisms or cryptographic arrangements for secret or secure communication H04L9/00
• CPC Class Codes: G11B 20/00884 involving a watermark, i.e. a barely perceptible transformation of the original data which can nevertheless be recognised by an algorithm
• CPC Class Codes: H04H 20/31 using in-band signals, e.g. subsonic or cue signal

Abstract

A method for inserting and detecting a watermark based on a stochastic model is disclosed. The apparatus for inserting a watermark based on a stochastic model, includes: a generator for generating a spread spectrum signal by using a user key from outside; an estimator for estimating an autocorrelation matrix of an original signal from outside as an object for inserting a watermark; an eigenvalue decomposer for decomposing the autocorrelation matrix to an eigenvector matrix and an eigenvalue matrix; a watermark shaper for shaping a watermark from the spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix; and an adder for adding the watermark to the original signal.

Claims

1. An apparatus for inserting a watermark based on a stochastic model, said apparatus comprising:

means for generating a spread spectrum signal by using a user key from outside;

means for estimating an autocorrelation matrix of an original signal from outside as an object for inserting a watermark;

means for decomposing the autocorrelation matrix to an eigenvector matrix and an eigenvalue matrix;

means for shaping a watermark from the spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix; and

means for adding the watermark to the original signal.

2. The apparatus as recited in claim 1, wherein said means for shaping a watermark uses a watermark transformation coefficient in the process of making a watermark from a spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix.

3. The apparatus as recited in claim 1, wherein said means for shaping a watermark controls the strength of a watermark by using a scaling coefficient.

4. The apparatus as recited in claim 3, wherein said means for shaping a watermark controls by a scaling coefficient the power of an watermark in order to get a desirable SNR, the diagonal elements of the eigenvlaue matrix are exponentially multiplied by the watermark transformation coefficient, and a spread spectrum signal is transformed to a watermark signal by using a transformation coefficient, a scaling coefficient, an eigenvector matrix and an eigenvalue matrix.

5. The apparatus as recited in claim 3, wherein said means for estimating an autocorrelation determines in a time domain autocorrelation coefficients by using Ergodic theorem, and then a Toepliz autocorrelation matrix by using the autocorrelation coefficients.

6. An apparatus for detecting a watermark based on a stochastic model, said apparatus comprising:

means for generating a spread spectrum signal by using a user key inputted from outside;

means for estimating an autocorrelation matrix of an original signal from a watermark-inserted signal received;

means for decomposing the autocorrelation matrix to an eigenvector matrix and an eigenvalue matrix;

means for shaping a watermark from a spread spectrum signal in reference to the eigenvector matrix and the eigenvalue matrix;

in order to detect a watermark without using an original audio signal, means for filtering for whitening the watermark-inserted signal by using both the eigenvector matrix and the eigenvalue matrix from the means of decomposing;

means for measuring correlation to find correlation between a watermark signal from said means for shaping a watermark and a whitening-filtered signal from said means for filtering for whitening; and

means for determining whether there is a watermark by referring to a correlation value from said means for measuring correlation, whether or not a received signal includes a watermark.

7. The apparatus as set forth in claim 6, wherein said means of shaping a watermark uses a watermark transformation coefficient in the process of making a watermark from a spread spectrum signal in reference to both the eigenvector matrix the eigenvalue matrix.

8. The apparatus as recited in claim 6, wherein said means for shaping a watermark controls the strength of a watermark by using a scaling coefficient.

9. The apparatus as recited claim 8, wherein said means for filtering for whitening performs filtering for whitening after calculating an inverse eigenvalue matrix and obtaining a whitening filter matrix by using a transforming coefficient, an error-tolerant coefficient, and diagonal elements of an inverse eigenvalue matrix.

10. A method for inserting a watermark based on a stochastic model, said method comprising the steps of:

a) generating a spread spectrum signal by using a user key inputted from outside;

b) estimating an autocorrelation matrix of an original signal from outside as an object for inserting a watermark;

c) decomposing an autocorrelation matrix to an eigenvector matrix and an eigenvalue matrix;

d) shaping a watermark from the spread spectrum signal produced in the step a) in reference to both the eigenvector matrix and the eigenvalue matrix; and

e) adding a watermark produced by said step d) to the original signal from outside.

11. The method as recited in claim 10, wherein said step d) uses a watermark transformation coefficient in the process of making a watermark from a spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix.

12. The method as recited in either claim 10, wherein said step d) controls the strength of a watermark by using a scaling coefficient.

13. The method as recited in claim 12, wherein said step d) controls by a scaling coefficient the power of an watermark in order to get a desirable SNR, the diagonal elements of the eigenvlaue matrix are exponentially multiplied by the watermark transformation coefficient, the spread spectrum signal is transformed to a watermark signal by using a transformation coefficient, a scaling coefficient, an eigenvector matrix and an eigenvalue matrix.

14. The method as recited in claim 10, wherein said step b) includes the steps of:

b1) determining in a time domain autocorrelation coefficients by using Ergodic theorem; and

b2) the seventh step to determine a Toepliz autocorrelation matrix by using the autocorrelation coefficients.

15. A method for detecting a watermark based on a stochastic model, said method comprising the steps of:

a) generating a spread spectrum signal by using a user key inputted from outside;

b) estimating an autocorrelation matrix of an original signal from a watermark-inserted signal received;

c) decomposing an autocorrelation matrix estimated at said step b) to both an eigenvector matrix and an eigenvalue matrix;

d) shaping a watermark from the spread spectrum signal produced in said step a) in reference to both the eigenvector matrix and the eigenvalue matrix;

e) in order to detect a watermark without using an original audio signal, filtering for whitening a watermark-inserted signal by using both the eigenvector matrix and the eigenvalue matrix from said step c);

f) measuring correlation between a watermark signal from said step d) and a whitening-filtered signal from said step e); and

g) determining, by referring to a correlation value from said step f), whether or not a received signal includes a watermark.

16. The method as recited in claim 15, wherein said step d) to shape a watermark uses a watermark transformation coefficient in the process of making a watermark from a spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix.

17. The method as recited in claim 15, wherein said step d) controls the strength of a watermark by using a scaling coefficient.

18. The method as recited in claim 17, wherein said step e) performs filtering for whitening after calculating an inverse eigenvalue matrix and obtaining a whitening filter matrix by using a transforming coefficient, an error-tolerant coefficient, and diagonal elements of an inverse eigenvalue matrix.

19. A computer readable medium storing a program for executing method for inserting a watermark based on a stochastic model, wherein the computer readable medium is in apparatus for inserting a watermark having a processor, the program comprising the functions of:

a) generating a spread spectrum signal by using a user key inputted from outside;

b) estimating an autocorrelation matrix of an original signal from outside as an object for inserting a watermark;

c) decomposing an autocorrelation matrix estimated at the function b) to an eigenvector matrix and an eigenvalue matrix;

d) shaping a watermark from a spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix; and

e) adding the watermark to the original signal from outside.

20. A computer readable medium storing a program for executing method for detecting a watermark based on a stochastic model, wherein the computer readable medium is in apparatus for detecting a watermark having a processor, the program comprising the functions of:

a) generating a spread spectrum signal by using a user key inputted from outside;

b) estimating an autocorrelation matrix of an original signal from a watermark-inserted signal received from outside as an object for detecting a watermark;

c) decomposing an autocorrelation matrix estimated at said second function to an eigenvector matrix and an eigenvalue matrix;

d) shaping a watermark from a spread spectrum signal in reference to both the eigenvector matrix and the eigenvalue matrix;

e) in order to detect a watermark without using an original audio signal, filtering for whitening a watermark-inserted signal by using both the eigenvector matrix and the eigenvalue matrix from the function c);

f) measuring correlation between a watermark signal from the function d) and a whitening-filtered signal from the function e); and

g) determining whether or not a received signal includes a watermark by referring to a correlation value from the function f).