Compact star-field sensor (SFS)

1. A star-field sensor (SFS), comprising:

a lens;

an image sensor board comprising an image sensor, the image sensor configured to receive light from the lens and produce an image from the received light; and

a processing board configured to communicate with the image sensor board, the processing board comprising:

a processor configured to control the SFS, and

a field programmable gate array (FPGA) configured to provide interfaces to communicate with devices that are external to the SFS and to provide clocks and power to the image sensor board.

2. The SFS of claim 1 , wherein the FPGA is further configured to provide watchdog timers that reboot the processor when a system fault occurs.

3. The SFS of claim 1 , wherein the image sensor board is connected to a digital camera interface (DCMI) on the processor that reads images produced by the image sensor board.

4. The SFS of claim 1 , wherein the processor is configured to:

determine a noise threshold of an image by finding a median pixel value of a representative portion of the image as a noise floor;

set a noise threshold at a predetermined level above the noise floor;

scan the image for sources and pixels of the image that are below the noise threshold; and

classify pixels above the noise threshold as source pixels.

5. The SFS of claim 4 , wherein the noise threshold is set at least 3-sigma above the noise floor.

6. The SFS of claim 4 , wherein when a source pixel is adjacent to a pixel from a previously detected source, the processor is further configured to add the adjacent source pixel to the previously detected source.

7. The SFS of claim 4 , wherein when a source pixel borders two existing sources, the processor is further configured to:

merge the two existing sources; and

add the source pixel that borders the two existing sources to the merged source.

8. The SFS of claim 4 , wherein when a source pixel does not border a known source, the processor is further configured to create a new source.

9. The SFS of claim 4 , wherein the processor is further configured to:

for each resulting source, determine image coordinates of a centroid using a weighted mean given by:

x

=

∑

(

x

i

*

c

i

)

∑

(

c

i

)

y

=

∑

(

y

i

*

c

i

)

∑

(

c

i

)

where x i and y i are individual pixel coordinates and c i is source counts above the image noise floor for that individual pixel; and

produce an object list that comprises image coordinates for each extracted star, total source counts above the noise threshold, and a number of pixels that the source occupied.

10. The SFS of claim 4 , wherein the processor is further configured to:

use a compilation of star catalogs; and

select a predetermined number of brightest stars available to attain a best signal-to-noise ratio (SNR) and improve a chance that the brightest stars available will be observed in the image.

11. The SFS of claim 10 , wherein the predetermined number of brightest stars available are determined by a magnitude of brightness, measured on a negative logarithmic scale according to:

M V =M V,1 −M V,0 =−2.5 log( F 1 /F 0 )

where M V is a reported magnitude of a star, M V,1 is an overall magnitude, M V,0 is a reference magnitude with Vega=0, F 1 is an observed flux, and F 0 is a reference flux for optical setup.

12. The SFS of claim 4 , wherein the processor is further configured to:

employ a two-dimensional 3 rd order polynomial fit to correct optical distortions and place detected objects on a normalized tangential sky projection.

13. The SFS of claim 4 , wherein the processor is further configured to:

derive attitude using a combination of algorithms as a calibration algorithm to generate a fast and precise attitude solution.

14. A method for determining a star-field implemented by a star-field sensor (SFS), comprising:

determining, by the SFS, a noise threshold of an image by finding a median pixel value of a representative portion of the image as a noise floor;

setting a noise threshold at a predetermined level above the noise floor, by the SFS;

scanning the image, by the SFS, for sources and pixels of the image that are below the noise threshold; and

classifying pixels above the noise threshold as source pixels, by the SFS.

15. The method of claim 14 , wherein

when a source pixel is adjacent to a pixel from a previously detected source, the method further comprises adding the adjacent source pixel to the previously detected source,

when the source pixel borders two existing sources, the method further comprises merging the two existing sources, by the SFS, and adding the source pixel that borders the two existing sources to the merged source, by the SFS, and

when the source pixel does not border a known source, the method further comprises creating a new source, by the SFS.

16. The method of claim 14 , further comprising:

for each resulting source, determining, by the SFS, image coordinates of a centroid using a weighted mean given by:

x

=

∑

(

x

i

*

c

i

)

∑

(

c

i

)

y

=

∑

(

y

i

*

c

i

)

∑

(

c

i

)

where x i and y i are individual pixel coordinates and c i is source counts above the image noise floor for that individual pixel; and

producing an object list, by the SFS, that comprises image coordinates for each extracted star, total source counts above the noise threshold, and a number of pixels that the source occupied.

17. The method of claim 14 , further comprising:

using a compilation of star catalogs, by the SFS; and

selecting a predetermined number of brightest stars available, by the SFS, to attain a best signal-to-noise ratio (SNR) and improve a chance that the brightest stars available will be observed in the image, wherein the predetermined number of brightest stars available are determined by a magnitude of brightness, measured on a negative logarithmic scale according to:

M V =M V,1 −M V,0 =−2.5 log( F 1 /F 0 )

where M V is a reported magnitude of a star, M V,1 is an overall magnitude, M V,0 is a reference magnitude with Vega=0, F 1 is an observed flux, and F 0 is a reference flux for optical setup.

18. The method of claim 14 , further comprising:

employing a two-dimensional 3 rd order polynomial fit, by the SFS, to correct optical distortions and place detected objects on a normalized tangential sky projection.

19. A star-field sensor (SFS), comprising:

a processor configured to:

determine a noise threshold of an image by finding a median pixel value of a representative portion of the image as a noise floor;

set a noise threshold at a predetermined level above the noise floor;

scan the image for sources and pixels of the image that are below the noise threshold; and

classify pixels above the noise threshold as source pixels, wherein

when a source pixel is adjacent to a pixel from a previously detected source, the processor is further configured to add the adjacent source pixel to the previously detected source,

when the source pixel borders two existing sources, the processor is further configured to merge the two existing sources and add the source pixel that borders the two existing sources to the merged source, and

when the source pixel does not border a known source, the processor is further configured to create a new source.

20. The SFS of claim 19 , wherein the processor is further configured to:

for each resulting source, determine image coordinates of a centroid using a weighted mean given by:

x

=

∑

(

x

i

*

c

i

)

∑

(

c

i

)

y

=

∑

(

y

i

*

c

i

)

∑

(

c

i

)

where x i and y i are individual pixel coordinates and c i is source counts above the image noise floor for that individual pixel; and

produce an object list that comprises image coordinates for each extracted star, total source counts above the noise threshold, and a number of pixels that the source occupied.