Astrometry Settings Issue - Limit Max Peaks in 8-bit Images

[mketi]

Hi there,

I've been working on a little project hat requires plate solving images produced in a small refractor and a sensor with fairly large pixels. I'm not sure if it's the right thing to do, but to counter star blockiness and keep out saturated reference stars, I selected the "Centroid Near Peaks" and "Limit Max Peaks" options in the Astrometry Settings area.

Based on the plate solving animations, it looks like centroid option works. I'm not sure about the "Limit Max Peaks", however. AIJ often seems to circle stars that are over my set limit. In my 8-bit images, if I set the max peak limit to 250, saturated stars with peak values of 255 may still be selected. When I lower the max peak limit all the way down to 30, it seems to stop choosing the saturated stars. Is the program behaving correctly, or am I perhaps doing something wrong?

Thanks very much for your help!

Matt

[karenacollins]

The astrometry feature is not well-tested using 8-bit images. If you'd like to send send us an example of an image that gives the behavior you describe below, we can look into when we get a chance. If you have trouble attaching the image to a post, you can send it to karenacollins atsymbol outlook dotsymbol com.

Karen

On 3/22/2021 5:00 AM, mketi [via AstroImageJ] wrote:

Hi there,

I've been working on a little project hat requires plate solving images produced in a small refractor and a sensor with fairly large pixels. I'm not sure if it's the right thing to do, but to counter star blockiness and keep out saturated reference stars, I selected the "Centroid Near Peaks" and "Limit Max Peaks" options in the Astrometry Settings area.

Based on the plate solving animations, it looks like centroid option works. I'm not sure about the "Limit Max Peaks", however. AIJ often seems to circle stars that are over my set limit. In my 8-bit images, if I set the max peak limit to 250, saturated stars with peak values of 255 may still be selected. When I lower the max peak limit all the way down to 30, it seems to stop choosing the saturated stars. Is the program behaving correctly, or am I perhaps doing something wrong?

Thanks very much for your help!

Matt

---

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

http://astroimagej.1065399.n5.nabble.com/Astrometry-Settings-Issue-Limit-Max-Peaks-in-8-bit-Images-tp1530.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[mketi]

Hi Karen,

It looks like the forum environment does not accept FITS files, so I'll send it to your email.

I've annotated the saturated objects that are circled in the plate solve animation. Hopefully the annotations show up on other computers, though I do see some entries in the FITS header. The troubled plate solving setting is Limit Max Peaks = 250. I'll also include a screenshot of everything that I used.

Thank you again!

Matt

[karenacollins]

Hi Matt,

Thanks for the image. After running a plate-solve on your image, I realized what is causing the stars with greater than 250 peak counts to be included in the plate-solve. As it turns out, the median filter is applied before the max peak count cut, and I expect you have that option enabled. If you really want to cut on the actually pixel values, you can disable the median filter option and see that the stars you circled are not included in the plate solve (i.e. will have no blue circle). That said, the plate-solve seems to complete fine with or without the brighter stars included.

This sequence of operations could be considered backwards, but it's a limitation of the implementation because the median filter, if enabled, needs to run before the source detection routine runs, so I think we are stuck with the awkwardness, unless the median filter is disabled.

Karen

On 3/26/2021 9:14 AM, mketi [via AstroImageJ] wrote:

Hi Karen,

It looks like the forum environment does not accept FITS files, so I'll send it to your email.

I've annotated the saturated objects that are circled in the plate solve animation. Hopefully the annotations show up on other computers, though I do see some entries in the FITS header. The troubled plate solving setting is Limit Max Peaks = 250. I'll also include a screenshot of everything that I used.

Thank you again!

Matt

---

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

http://astroimagej.1065399.n5.nabble.com/Astrometry-Settings-Issue-Limit-Max-Peaks-in-8-bit-Images-tp1530p1532.html

To start a new topic under AstroImageJ, email [hidden email]
To unsubscribe from AstroImageJ, click here.
NAML

[mketi]

Hi Karen,

Ah, that explanation makes sense. Yes, I had the median filter enabled, as you thought. I can work around this issue now that I understand what is happening. Really appreciate your help!

Matt