Method and apparatus for setting a detection threshold given a desired false probability

1. A method for determining a watermark detection threshold value in a watermark detection system, said method comprising:

selecting a desired false result probability; and

determining, by a processor, the watermark detection threshold value, wherein the watermark detection threshold value is determined using the selected desired false result probability given by

P

fp

=

m

!

∏

i

=

1

m

-

1

⁢

⁢

(

T

+

i

)

where P fp is said desired false result probability, T is said watermark detection threshold value and m is a dimension of a feature vector, and wherein the watermark detection threshold value is set when a watermark detection measure is a ratio of a largest normalized correlation to a second largest normalized correlation.

2. The method according to claim 1 , further comprising:

creating a threshold table, wherein

said creating act further comprises:

calculating each false result probability entry in said threshold table using

m

!

∏

i

=

1

m

-

1

⁢

⁢

(

T

+

i

)

storing said calculated false result probability in said threshold table and said threshold value used to calculate said false result probability entry;

selecting said threshold value from said threshold table corresponding to said desired false result probability, or by performing iteratively the steps of:

setting a current threshold value to said first threshold value;

calculating a false result probability using said current threshold value;

calculating a difference value between said calculated false result

probability and said desired false result probability;

comparing said difference value to a first predetermined value;

comparing said difference value to a second predetermined value if a

result of said first comparing act is positive; and

modifying said current threshold value responsive to said comparisons.

3. The method according to claim 2 , wherein said calculating act is performed using a first threshold value, a second threshold value and a threshold interval, iteratively incrementing said first threshold value by said threshold interval and calculating each said false result probability entry of said threshold value is equal to said second threshold value.

4. The method according to claim 3 , wherein said false result probability is a false positive probability and further wherein said first threshold value is a lower bound of a false negative probability and said second threshold value is an upper bound of said false positive probability.

5. The method according to claim 1 , wherein said selecting act retrieves said threshold value from said threshold table, wherein said threshold value is a greatest threshold value having a false result probability less than said desired false result probability.

6. The method according to claim 1 , wherein said false result probability is a false positive probability.

7. An apparatus for determining a watermark detection threshold value in a watermark detection system, comprising:

means for selecting a desired false result probability; and

means for determining the watermark detection threshold value, wherein means for determining the watermark detection threshold value is performed using the selected desired false result probability given by

P

fp

=

m

!

∏

i

=

1

m

-

1

⁢

⁢

(

T

+

i

)

where P fp is said desired false result probability, T is said watermark detection threshold value and m is a dimension of a feature vector, and wherein the watermark detection threshold value is set when a watermark detection measure is a ratio of a largest normalized correlation to a second largest normalized.

8. The apparatus according to claim 7 , further comprising:

means for creating a threshold table, wherein said creating means further comprises:

means for calculating each false result probability entry in said threshold table using

m

!

∏

i

=

1

m

-

1

⁢

⁢

(

T

+

i

)

means for storing said calculated false result probability in said threshold table and said threshold value used to calculate said false result probability entry;

selecting said threshold value from a threshold table corresponding to said desired false result probability; or

means for selecting a first threshold value;

means for setting a current threshold value to said first threshold value;

means for calculating a false result probability using said current threshold value;

means for calculating a difference value between said calculated false result probability and said desired false result probability;

means for comparing said difference value to a first predetermined value;

means for comparing said difference value to a second predetermined value if a result of said first comparing act is positive; and

means for modifying said current threshold value responsive to said comparisons.

9. The apparatus according to claim 8 , wherein said means for calculating is performed using a first threshold value, a second threshold value and a threshold interval, iteratively incrementing said first threshold value by said threshold interval and calculating each said false result probability entry of said threshold table until said first threshold value is equal to said second threshold value.

10. The apparatus according to claim 9 , wherein said false result probability is a false positive probability and further wherein said first threshold value is a lower bound of a false negative probability and said second threshold value is an upper bound of said false positive probability.

11. The apparatus according to claim 7 , wherein said means for selecting retrieves said threshold value from said threshold table, wherein said threshold value is a greatest threshold value having a false result probability less than said desired false positive probability.

12. The apparatus according to claim 6 , wherein said false result probability is a false positive probability.