Mode Combine in AIJ ?

[sengle01]

Hello,

I'm sorry if this option already exists - I've looked around a good bit and haven't been able to find it.

If I'm correct and it doesn't exist, are there any plans to create an option to Mode Combine (as opposed to Average, Median...) images or a stack of images?

Best,
Scott

[karenacollins]

Hi Scott,

You are correct that Mode Combine is not currently an option. However, I'm adding that option to the feature development
list.

Karen

[sengle01]

Karen,

Glad I wasn't mistaken! Thank you for the quick reply and for adding the feature to the list.

Best,
Scott

[karenacollins]

Scott,
I looked into adding a "mode" combine option to DP. However, it is not clear to me which value to choose if there are two more values with the same number of occurrences (i.e. the distribution is multi-modal). Do you have a preference on how that case is handled?
Karen

[sengle01]

Karen,

I'm sorry, I completely missed this reply when it first arrive and just saw it now.

It's a good point - my quick, short answer would be that I have no real preference.

The most common application that I personally see for the mode combine is producing sky flats. Unless you're unlucky, observe densely packed regions of space, or simply aren't taking enough exposures, the sky values should be the most represented.

Perhaps, if you were combining some images and a star fell on the same pixels in 10 out of 20 images, then you might get 10 star values and 10 sky values, so two peaks. In this case, I would expect the star peak to be narrow and the sky peak to be broader. In that case, is there any way to (relatively easily) look at the neighboring values for each peak, and choose the broadest of the peaks?

Best,

Scott

---

From: karenacollins [via AstroImageJ] <ml+[hidden email]>
Sent: Monday, December 4, 2017 7:10:50 AM
To: Scott Engle
Subject: Re: Mode Combine in AIJ ?

 

Scott,
I looked into adding a "mode" combine option to DP. However, it is not clear to me which value to choose if there are two more values with the same number of occurrences (i.e. the distribution is multi-modal). Do you have a preference on how that case is handled?
Karen

---

If you reply to this email, your message will be added to the discussion below:

http://astroimagej.1065399.n5.nabble.com/Mode-Combine-in-AIJ-tp917p928.html

To unsubscribe from Mode Combine in AIJ ?, click here.
NAML

[karenacollins]

Hi Scott,

I'm under a crunch the next few weeks, but I can add this to the development to-do list. Unfortunately, the GUI
interface change is more work than implementing the underlying mode calculation. Unless you let me know there is a
particular problem with median combine for flats as a temporary substitute, I'd like to treat this as a medium priority
issue. I'll try to remember to send an update to this thread when it is implemented.

Karen