Author Topic: PI 1.8.7 Parsing Error During Image Integration  (Read 167 times)

Offline psenness

  • Newcomer
  • Posts: 1
    • View Profile
PI 1.8.7 Parsing Error During Image Integration
« on: 2019 October 06 11:37:04 »
Since upgrading to PI 1.8.7 when I do ImageIntegration I get the following error on each subframe in the Process Console:

*** Error: Parsing ISOSPEED FITS keyword: Parsing 64-bit floating point expression: conversion error:
Default
^
At the end of the integration I get this:

** Warning: Rescaling output image. Integration range: [-3.3802277e-04,1.0251132e+00]

Camera is a ZWO ASI294MC Pro. I have no idea what the error means and the final image looks OK. But error messages and warnings usually mean something isn't right and I'm at a loss.

Advice please.

Offline Juan Conejero

  • PTeam Member
  • PixInsight Jedi Grand Master
  • ********
  • Posts: 6842
    • View Profile
    • http://pixinsight.com/
Re: PI 1.8.7 Parsing Error During Image Integration
« Reply #1 on: 2019 October 06 11:52:27 »
There is no bug here.

Quote
*** Error: Parsing ISOSPEED FITS keyword: Parsing 64-bit floating point expression: conversion error:
Default
^

'Default' is not a valid value for the ISOSPEED FITS keyword. It should be a numeric value equal to the ISO speed used to acquire the image, as specified in ISO 12232. So this is a bug in the software you have used to acquire the images. This error is not fatal; it will be shown on the console but the process will continue normally.

Quote
** Warning: Rescaling output image. Integration range: [-3.3802277e-04,1.0251132e+00]

This warning is telling you that the result after integration of your light frames does not fit in the nominal [0,1] range, where 0 is black and 1 is white. In other words, the integrated image would become slightly saturated if truncated. To prevent saturation, ImageIntegration is rescaling the image linearly to fit in [0,1], so no data at all will be lost. This is the default behavior.

Saturation should not happen under normal conditions with properly calibrated data. When it happens, it is usually because there are some already saturated or nearly saturated pixels that have not been rejected (for example, because they lie at fixed positions across the entire frame set). Another possible cause is improperly applied pedestals. Anyway, this is just a warning and you usually don't need to worry too much about it, unless the integration range is largely beyond the [0,1] range.
Juan Conejero
PixInsight Development Team
http://pixinsight.com/