Evaluate Darks

[CJ-Wright]

Automatically evaluating the darks to ensure they are "good" and rejecting any that are bad would be extremely helpful. For example, if a dark is taken right after a strongly scattering sample, it will contain a weak diffraction pattern from the previous sample, which will of course interfere with subsequent images processed using that dark. It would be great if this could be avoided automatically by the software.

[CJ-Wright]

Attn: @dooryhee Would it be possible to come up with a metric for dark "goodness"?

[dooryhee]

One immediate thought is to identify whether or not the dark contains ghost (residual) diffraction peaks. Frequency analysis?

From: Christopher J. Wright notifications@github.com Sent: Tuesday, July 31, 2018 11:46 AM To: xpdAcq/mission-control mission-control@noreply.github.com Cc: Dooryhee, Eric edooryhee@bnl.gov; Mention mention@noreply.github.com Subject: Re: [xpdAcq/mission-control] Evaluate Darks (#64)

Attn: @dooryheehttps://github.com/dooryhee Would it be possible to come up with a metric for dark "goodness"?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/xpdAcq/mission-control/issues/64#issuecomment-409268504, or mute the threadhttps://github.com/notifications/unsubscribe-auth/ALyE3d-wcp3cokw4nezEfEd32ROFhAnYks5uMHungaJpZM4QHjo0.

[MehmetTopsakal]

@CJ-Wright @dooryhee

I am trying to understand this issue.

This is one of the recent experiment.

11x11 robot scan. I am able to find the center of sample as denoted by star.

scan-a is the measurement at the center of the sample. scan-b and scan-c are nearby scans. scan-d and scan-e are far from the sample.

If we take scan-e as the "reference", we can compare it with other scans using the Spearman's correlation metric.

• https://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient
• https://docs.scipy.org/doc/scipy-0.15.1/reference/generated/scipy.stats.spearmanr.html

If we have a "good" dark, we can use Spearman as metric to compare subsequent darks with the "good" one.

Spearman is invariant with y-scale and small deviations in the intensities.

Hope this helps