flatdiv: Divide spectra by flat field

Package: irs

Usage

flatdiv input records

Parameters

input: The root file name for the input records to be divided.

records: The range of spectra to be included in the divide operation. Each range item will be appended to the root name to form an image file name.

output: New spectra are created by the flatdiv operation. This parameter specifies the root name to be used for the created spectra.

start_rec: Each new spectrum created as "output" has its root name and a trailing number appended starting with "start_rec". Subsequent output images will have an incremented trailing number. Note that even if an output image is not created because the input image has already been flattened or the input image is not found the output record number is still incremented.

flat_file: The root name for the sensitivity spectra as produced by FLATFIT. Normally with multi-aperture instruments, FLATFIT will produce a spectrum appropriate to each aperture and the file name will have "flat_file" as the file name root and the aperture number appended.

coincor = )_.coincor

If set to yes, coincidence correction is applied to the data during the division, and the following three parameters are required. For more about this correction see coincor.

ccmode = )_.ccmode: The mode by which the coincidence correction is to be performed. This may be "iids" or "photo".

deadtime = )_.deadtime: The detector deadtime in seconds.

power = )_.power: Power law IIDS non-linear correction exponent.

Description

The input spectra are divided by the flat fields which are represented by spectra produced by FLATFIT.

To avoid possible division by zero, any zeroes in the flat field spectra generated by FLATFIT are replaced by 1.0.

The input spectra may optionally be corrected for coincidence losses.

If the input and output spectra (after appending the record numbers) are the same then the division is performed in-place; i.e. the flattened spectra replace the original input spectra. Note that even if an output image is not created because the input image has already been flattened or the input image is not found the output record number is still incremented. This is to insure that if in-place division is desired that the input and output names remain matched.

Examples

The following example divides a series of spectra to produce 20 new spectra having names nite1.1221 ... nite1.1240.

cl> flatdiv nite1 1201-1220 nite1 1221

The same spectra as above are simultaneously corrected for coincidence losses.

cl> flatdiv nite1 1201-1220 nite1 1221 coincor=yes

The flattened spectra replace the unflattened spectra.

cl> flatdiv nite1 1201-1220 nite1 1201

Note that the input record numbers must be contiguous and the starting output record number must be the same as the first input record number.

Time requirements

Approximately 1 second is required to correct a spectrum of length 1024 points.