Opened 10 years ago

Closed 10 years ago

Last modified 9 years ago

#250 closed task (invalid)

Create program to rescale dark file

Reported by: benj Owned by: benj
Priority: immediate Milestone:
Component: PML utilities Keywords:
Cc: mggr Other processors:

Description (last modified by mggr)

We need a program that will read in a dark frame file, rescale it to mimic that for a different integration time and then output the new file.

Basic level: Read in data file, multiply by a user-specified scaling factor (except pixels 1 & 2 on band 1 for each line), output new data file.

What we really want: Same, but calculate scaling factor from differential of integration times - check relative data values for existing dark files to try and work out if the result of changing the integration time is just a linear scaling (hopefully) or something more complicated, then implement appropriate algorithm. The user should also be able to specify a scaling factor as above if they prefer.

Gold-plated: Same, but also account for changes in framerate and in spectral/spatial binning. May be worth spending half an hour to check whether the effects of altering these are also just linear scaling, but if they're not then don't include them (assume constant).

Related tickets: #218 #249

Attachments (1)

hawk_tint_darkframes.png (5.9 KB) - added by benj 10 years ago.
Hawk dark frame values varying with integration time

Download all attachments as: .zip

Change History (7)

comment:1 Changed 10 years ago by jatt

Basic level script has been completed and is at:-

~jatt/scratch/scripts/darkframes_scale_basic.py

Tested using:

  • ~arsf/arsf_data/2008/flight_data/vocals/VOC_08-2008_310_Chile/hawk/SWIRshvoc08_31008-4dark.raw
  • ~arsf/arsf_data/2008/flight_data/vocals/VOC_08-2008_310_Chile/hawk/SWIRshvoc08_31008-4dark.hdr

and a scale factor of 2

New output darkframes file is at:

  • ~jatt/scratch/ARSF/newdark_test.raw
  • ~jatt/scratch/ARSF/newdark_test.hdr

Changed 10 years ago by benj

Hawk dark frame values varying with integration time

comment:2 Changed 10 years ago by benj

Jane's script suggests that the Hawk dark frame values are linearly proportional to integration time. Regression suggests that the dark frame value is in the region of 320 times the integration time. Actual values obtained from gnuplot's curve (well, line) fitting:

Final set of parameters            Asymptotic Standard Error
=======================            ==========================

m               = 322.077          +/- 3.197        (0.9925%)
c               = -30.1481         +/- 41.63        (138.1%)

Graph:
Hawk dark frame values varying with integration time

comment:3 Changed 10 years ago by jatt

My script suggests that the values for Eagle are constant within the range of 51 - 62 regardless of binning and integration time.

comment:4 Changed 10 years ago by mggr

  • Owner changed from jatt to benj

Push this code into ~arsf and close the ticket?

comment:5 Changed 10 years ago by benj

  • Resolution set to invalid
  • Status changed from new to closed

We actually wound up not needing this code - Eagle frames didn't need rescaling (dark frame values constant irrespective of integration time), and for Hawk the rescaling can be done within azspec.

Closing ticket.

comment:6 Changed 9 years ago by mggr

  • Description modified (diff)

Added convenience links for later memory assistance ;)

Note: See TracTickets for help on using tickets.