Next: , Previous: disk, Up: Position Generators


6.1.4 image( name [, optargs] )

This specifies that the angular distribution of the source is derived from an image, either read from file or one previously created (the latter is currently not supported). name is a string used for identification purposes. Note that the `file' optional argument is for now not optional; it must be specified.

The following optional arguments are recognized:

file
The name of the image file
format
The format of the file. If not specified, B<raygen> will attempt to guess at the format. It may be one of the following:
`ascii'
The pixel values are in ASCII. Values are separated by whitespace (including newlines). The file does not contain any information about the pixel size or dimensions; you must specify these using the pixsz, pixsz_x, pixsz_y, nx, ny arguments.
`double'
The pixel values are stored as binary double precision floating point numbers. The file does not contain any information about the pixel size or dimensions; you must specify these using the pixsz, pixsz_x, pixsz_y, nx, ny arguments.
`float'
The pixel values are stored as binary single precision floating point numbers. The file does not contain any information about the pixel size or dimensions; you must specify these using the pixsz, pixsz_x, pixsz_y, nx, ny arguments.
`fits'
The image is a standard FITS image. The pixel size may be specified by the `CDELT'n FITS keywords, or by the optional pixsz, pixsz_x, pixsz_y arguments. If they are not specified, the default is 0.5 seconds of arc. If the pixel size is specified in the FITS file, the units for the size may be specified with the `CUNIT'n FITS keywords. If not, degrees are assumed. The FITS keywords `CRPIX'n and `CRVAL'n are supported.

If the FITS image has a coordinate with a `CTYPE' whose value begins with `RA', it is taken to be in the standard Astronomical Right Ascension coordinate system, and the X axis flip flag is turned on automatically (as in that system the X pixel increment is negative, which leads to incorrectly flipped images otherwise).

`pbm'
The image is in the Portable Bitmap format. The dimensions of the image are encoded in the file. The user must specify the size of the pixels using the optional pixsz, pixsz_x, pixsz_y arguments
`pgm'
The image is in the Portable Grayscale Map format. The dimensions of the image are encoded in the file. The user must specify the size of the pixels using the optional pixsz, pixsz_x, pixsz_y arguments

nx
ny
The dimensions of the image. Depending upon the image format, this may be specified in the image file.
pixsz
pixsz_x
pixsz_y
The size of the pixels. Pixels need not be square. Depending upon the image format, this may be specified in the image file.
theta
The distance from the optical axis to the center of the rectangle, in radians. If not specified, it defaults to `0'.
phi
The azimuth about the optical axis, in radians. If not specified, it defaults to `0'.
z
The Z coordinate of the source. It defaults to an infinitely distant source.
clip
The minimum value of pixel to raytrace. Pixels with values less than this are ignored.
flip
This takes one of the following values:
x
The image is flipped along the X axis.
y
The image is flipped along the Y axis.
xy
The image is flipped along both the X and Y axes.

point_idx_x
point_idx_y
The pixel index (possibly fractional) of the position in the image which corresponds to the angular source position specified by theta and phi. You may also specify this indirectly via point_wcs_x/point_wcs_y and ref_idx_x/ref_idx_y and ref_wcs_x/ref_wcs_y.
point_wcs_x
point_wcs_y
The WCS position which corresponds to the angular source position specified by theta and phi. This is used in conjunction with ref_idx_x/ref_idx_y and ref_wcs_x/ref_wcs_y.
ref_wcs_x
ref_wcs_y
The WCS position of a reference point in the image. This is used in conjunction with ref_idx_x/ref_idx_y and point_wcs_x/point_wcs_y.
ref_idx_x
ref_idx_y
The pixel index (possibly fractional) of the reference point in the image. This is used in conjunction with ref_wcs_x/ref_wcs_y and point_wcs_x/point_wcs_y.