| Parameter | Mand | Type | Default | Constraints |
| imagesets | yes | string | none | |
List of input image datasets to be mosaiced
|
| mosaicedset | yes | string | none | |
Name of output mosaiced image dataset
|
| correlset | no | string | none | |
Name of output correlation file dataset
|
| binaxis | no | real | | 0 : 2 |
Image binning factor. 0 means take the largest
binning factor of all the input images and compute
the output image using this value.
1 means that the output image will be a high-resolution image, 2 a low-resoloution one.
|
| mincorr | no | real | | -1 to +1 |
Minimum acceptable correlation successful image alignment
If 0 then no alignment attempted.
If < 0 then images stacked even if the correlation is less than abs(mincorr)
|
The following parameters are for use in the image-alignment algorithm- the default
values have been set by trial and error- changing any of them may significantly increase
the cpu time without improving anything.
| nsigma | no | real | 2 | 0.5 : |
Image pixel used in cross-correlation if the pixel value
is at least background + max(1.0, nsigma * sqrt(background) ),
where background is the computed background level at the pixel
|
| minfraction | no | real | 0.5 | 0.05 : 1 |
Minimum acceptable area overlap of 2 images before
alignment is attempted
|
| maxdx | no | int | 5 | 1 : 50 |
Maximum x and y offsets (in image pixels) between 2 images refereence pixels to
be used in cross-correlation
|
| numintervals | no | int | 2 | 1 : 100 |
The number of intervals to split an x or y offset value into. Ie the
total number of x or y offsets to be used to compute the initial correlation
coefficients will be (2xdx+1)*numintervals and (2xdy+1)*numintervals,respectively.
|
| di | no | int | 10 | 5 to 100 |
The correlation image will be 2*di+1 pixels square.
|
| minnumpixels | no | int | 100 | 20 : |
The minimum number of image pixels that lie above the background than can be used
in the cross-correlation.
|
| Parameter | Mand | Type | Default | Constraints |
