Hi Colin,
The parametric PSF is a
variable shape function defined as:
where the parameters are:
B: Local background.
A: Amplitude, which is the maximum value of the function, and also the function's value at the centroid coordinates.
x0,
y0: Centroid coordinates in pixel units. This is the position of the center of symmetry of the function.
sigmax,
sigmay: Standard deviation of the function's distribution on the horizontal and vertical axes, measured in pixels.
As implemented in tools such as Convolution and Deconvolution, the following parameters have fixed values:
B=0,
A=1,
x0=
y0=0.
The parameter
k > 0 is the
shape parameter. When
k = 2, the function is Gaussian; when
k < 2 the function is leptokurtic (peaked profile), and when
k > 2 it is platykurtic (flat profile). The variable shape function is useful because one can fine tune the PSF's profile with a single parameter, taking a Gaussian profile as reference.