Path: SpacecraftEstimation/StarCentroiding
% Center of mass centroiding algorithm.
Pixels below the noise threshold will not be included in the center of mass
calculation. The noise will be subtracted from the remaining pixels. The
threshold is generally some amount, i.e. 3-5 sigma, above the mean noise floor
value.
If 'edge' is true, the pixels beyond the region of interest will be averaged
to compute the 'edge noise'. This value will then be used for options.b. The
maximum edge pixel will be used as the threshold.
If the pixelmap has already had noise removed, pass 'false' for the edge
option and zero for the noise.
--------------------------------------------------------------------------
Form:
options = CentroidCOM
[x0,y0,iC] = CentroidCOM( pixels, iCenter, options )
--------------------------------------------------------------------------
------
Inputs
------
pixels (:,:) Pixel map or array
iCenter (1,n) Index of bright centers in array
options (.) Options data
.b (1,1) Sky-background noise floor
.T (1,1) Sky-background threshold
.nROI (1,1) Size of region of interest (pixels)
.edge (1) Logical for computing edge noise
.minPix (1,1) Minimum number of significant pixels
-------
Outputs
-------
x0 (1,:) x coordinates (column)
y0 (1,:) y coordinates (row)
iC (1,:) Total intensity of the centroids
--------------------------------------------------------------------------
Reference: Liebe, C. "Accuracy Performance of Star Trackers - A Tutorial",
IEEE Transactions on Aerospace and Electronic Systems. Vol. 38, N0. 2,
April, 2002.
--------------------------------------------------------------------------
Back to the SpacecraftEstimation Module page