Determining thresholds to filter noise in GmAPD LADAR data

1. A computerized method to automatically determine thresholds, comprising:

receiving data generated from a Geiger-mode avalanche photodiode (GmAPD) laser detecting and ranging (LADAR) sensor of a scene;

determining an overall threshold value for the scene using a binomial distribution;

determining post threshold values for each x-y position in the scene; and

for each x-y position, using a maximum value of the post threshold values and the overall threshold value to filter the data from the scene,

wherein determining the overall threshold value comprises:

determining a noise density;

determining a volume of a correlation matrix, v;

solving a cumulative distribution function for a candidate threshold value;

using the candidate threshold value as the overall threshold value if the candidate threshold value meets minimum requirements; and

using a default value as the overall threshold value if the candidate threshold value does not meet the minimum requirements.

2. The method of claim 1 wherein

the cumulative distribution function is:

1

-

∑

i

=

0



x



⁢

(

n

i

)

⁢

p

i

⁡

(

1

-

p

)

n

-

i

=

C

,

where n is a number of trials, p is the probability of a point falling in the correlation matrix volume based on the noise density, x is the candidate threshold and C is a constant.

3. The method of claim 2 wherein using a default value as the overall threshold value comprises:

using a default value equal to 2*v*(R/S) if n is greater than about 10% of the number of voxels in the volume;

using a default value equal to R*2.5 if the candidate threshold value is less than R*2.5; and

using a default value equal to S*10.0 if the candidate threshold value is less than S*10.0,

wherein R is equal to the average number of reflection received at the LADAR for each x-y position, S is the average number of times the LADAR shines on each x-y position and v is equal to the volume of the correlation matrix.

4. The method of claim 2 wherein C=1×10 −6 .

5. A computerized method to automatically determine thresholds, comprising:

receiving data generated from a Geiger-mode avalanche photodiode (GmAPD) laser detecting and ranging (LADAR) sensor of a scene;

determining an overall threshold value for the scene using a binomial distribution;

determining post threshold values for each x-y position in the scene; and

for each x-y position, using a maximum value of the post threshold values and the overall threshold value to filter the data from the scene,

wherein determining post threshold values for each x-y position in the scene comprises determining a first post threshold value equal to a minimum value of p avg , p std , and p max where p avg is an average voxel value used for an x-y position, P std is a standard deviation of voxel values used for the x-y position and p max is the maximum voxel value used for the x-y position.

6. The method of claim 5 wherein determining post threshold values for each x-y position in the scene comprises determining a second post threshold value equal to (f/10)*(p max −p avg )+p avg where f is a sigma multiplier.

7. An apparatus comprising:

circuitry to determine thresholds comprising circuitry to:

receive data generated from a Geiger-mode avalanche photodiode (GmAPD) laser detecting and ranging (LADAR) sensor of a scene;

determine an overall threshold value for the scene using a binomial distribution;

determine post threshold values for each x-y position in the scene; and

for each x-y position, use a maximum value of the post threshold values and the overall threshold value to filter the data from the scene,

wherein the circuitry to use a default value as the overall threshold value comprises circuitry to:

determine a noise density;

determine a volume of a correlation matrix, v;

solve a cumulative distribution function for a candidate threshold value;

use the candidate threshold value as the overall threshold value if the candidate threshold value meets minimum requirements; and

use a default value as the overall threshold value if the candidate threshold value does not meet the minimum requirements.

8. The apparatus of claim 7 wherein the circuitry comprises at least one of a processor, a memory, programmable logic and logic gates.

9. The apparatus of claim 7 wherein

the cumulative distribution function is:

1

-

∑

i

=

0



x



⁢

(

n

i

)

⁢

p

i

⁡

(

1

-

p

)

n

-

i

=

C

,

where n is a number of trials, p is the probability of a point falling in the correlation matrix volume based on the noise density, x is the candidate threshold and C is a constant.

10. The apparatus of claim 9 wherein the circuitry to use a default value as the overall threshold value comprises circuitry to:

use a default value equal to 2*v*(R/S) if n is greater than about 10% of the number of voxels in the volume;

use a default value equal to R*2.5 if the candidate threshold value is less than R*2.5; and

use a default value equal to S*10.0 if the candidate threshold value is less than S*10.0,

wherein R is equal to the average number of reflection received at the LADAR for each x-y position, S is the average number of times the LADAR shines on each x-y position and v is equal to the volume of the correlation matrix.

11. The apparatus of claim 9 wherein C=1×10 −6 .

12. An apparatus comprising:

circuitry to determine thresholds comprising circuitry to:

receive data generated from a Geiger-mode avalanche photodiode (GmAPD) laser detecting and ranging (LADAR) sensor of a scene;

determine an overall threshold value for the scene using a binomial distribution;

determine post threshold values for each x-y position in the scene; and

for each x-y position, use a maximum value of the post threshold values and the overall threshold value to filter the data from the scene,

wherein the circuitry to determine post threshold values for each x-y position in the scene comprises circuitry to determine a first post threshold value equal to a minimum value of p avg , p std , and p max where P avg is an average voxel value used for an x-y position, p std is a standard deviation of voxel values used for the x-y position and p max is the maximum voxel value used for the x-y position.

13. The apparatus of claim 12 wherein the circuitry to determine post threshold values for each x-y position in the scene comprises circuitry to determine a second post threshold value equal to (f/10)*(p max −p avg )+p avg where f is a sigma multiplier.

14. An article comprising:

a non-transitory machine-readable medium that stores executable instructions to automatically detect buildings, the instructions causing a machine to:

receive data generated from a Geiger-mode avalanche photodiode (GmAPD) laser detecting and ranging (LADAR) sensor of a scene;

determine an overall threshold value for the scene using a binomial distribution;

determine post threshold values for each x-y position in the scene; and

for each x-y position, use a maximum value of the post threshold values and the overall threshold value to filter the data from the scene.

15. The article of claim 14 wherein the instructions causing a machine to determine the overall threshold value comprises instructions causing a machine to:

determine a noise density;

determine a volume of a correlation matrix, v;

solve a cumulative distribution function for a candidate threshold value, x, the cumulative distribution function is:

1

-

∑

i

=

0



x



⁢

(

n

i

)

⁢

p

i

⁡

(

1

-

p

)

n

-

i

=

C

,

where n is a number of trials, p is the probability of a point falling in the correlation matrix volume based on the noise density and C is a constant;

use the candidate threshold value as the overall threshold value if the candidate threshold value meets minimum requirements; and

use a default value as the overall threshold value if the candidate threshold value does not meet the minimum requirements.

16. The article of claim 15 wherein the instructions causing a machine to use a default value as the overall threshold value comprises instructions causing a machine to:

use a default value equal to 2*v*(R/S) if n is greater than about 10% of the number of voxels in the volume.

use a default value equal to R*2.5 if the candidate threshold value is less than R*2.5; and

use a default value equal to S*10.0 if the candidate threshold value is less than S*10.0,

wherein R is equal to the average number of reflection received at the LADAR for each x-y position, S is the average number of times the LADAR shines on each x-y position and v is equal to the volume of the correlation matrix.

17. The article of claim 15 wherein C=1×10 −6 .

18. The article of claim 14 wherein the instructions causing a machine to determine post threshold values for each x-y position in the scene comprises instructions causing a machine to determine a first post threshold value equal to a minimum value of p avg , p std , and p max where p avg is an average voxel value used for an x-y position, p std is a standard deviation of voxel values used for the x-y position and p max is the maximum voxel value used for the x-y position.

19. The article of claim 18 wherein the instructions causing a machine to determine post threshold values for each x-y position in the scene comprises instructions causing a machine to determine a second post threshold value equal to (f/10)*(p max −p avg )+ p avg where f is a sigma multiplier.