Looking for advice: Top pick judging criteria
What are the judges looking for to qualify an image for Top pick or TPN?
Looking for advice: Top pick judging criteria
What are the judges looking for to qualify an image for Top pick or TPN?
The IotD guidelines are a great place to start. Everyone also has their own standards too, which is not a bad thing because it produces variety.
For instance, I am a stickler for proper resolution, that is to say, there’s no point going for an insane pixel scale if it adds nothing to your image. If I am going to zoom in, I should see the details continue to improve, but a lot of the time it never does and zooming in just reveals flaws and a worse noise profile.
On the whole though, a lot of what we notice, especially of images that sneak into the reviewer queue, is an excessive use of AI tools.
Since I’m here I will also mention planetary nebulae. The nebulae itself tends to get all the focus, so you end up with a good looking planetary nebula surrounded by a poorly processed background. This has sunk many a PN image in the IotD process.
For anyone who feels like they are struggling to get awards, I would suggest checking out awarded images and comparing them to your own. Not just in aesthetics of course, but integration time or any useful information they give in the description. You’ll get there eventually!
For me in my submitter queue (TPN) I scroll and pause for a a brief second at each image. On average I’d say there are 140 or so images in this queue, and they’re only in there for up to two days, so they rotate often. Some I can tell just from that that I’m not going to promote them. Usually it’s because of background neutralization issues, color balance, the overall sharpness (or lack of it) in the image, or if its a pretty easy/common object it just doesn’t stand out.
Also, we can only see the image title, the telescope, the camera, and how long ago the image was published. So I’m a bit more strict when it comes to extremely high end equipment. If you’re imaging with an ASA Telescope and an FLI camera, your image should look good. There are a non-negligible amount of images that come through our queues with extremely high end equipment similar to that, and they just aren’t very good.
Then a lot of the images will mouse over it which gives us a center crop. If I like what I see I’ll click on it and zoom/pan around it. First looking for glaring technical issues and then looking more at the detail. I generally don’t zoom much past 1.00x. If I still like what I see I’ll back out of the zoom, and take in the overall photo again. If I still like it I’ll promote it. If I’m not sure, I’ll leave it and come back if I have promotions left for that day once I get through the rest of the images for that day.
I’m also a reviewer (TP), and this queue is pretty similar but much smaller, however images in this queue stay for longer. I do a lot of the above, but also as I’m panning and zooming around an image I’m looking more closely at the details. I’m also being more critical of noise and artifact issues that might be there. Since the images are in the queue longer, I will often leave them in my queue in order to look at them the next day with fresh eyes to see if my feelings on it have changed.
Using my latest image IC1396 as an example. Does it not meet any of the requirements? Would like to improve in areas where needed. Thanks!
The main thing going against it is that IC1396 is imaged fairly often so it’s a bit harder to stand out. The past couple weeks there have been a lot of good images, so it’s been a bit hard to decide which 6 to promote some days. If it came up in my queue I’d definitely consider a promotion, but its hard to say yes or no without the context of the other images.
I like the color. The stars are good. They’re there without being overpowering. I like the wider field that you show compared to many who focus all on the elephant trunk itself, at the same time, a slightly tighter crop appeals more to my eye. Mainly the nebulosity in the bottom right taking my attention away from the center of the image.
Thank you so much for the feedback Quinn! Greatly appreciated. 🙏
Spencer · Jul 20, 2026, 12:50 PM
Using my latest image IC1396 as an example. Does it not meet any of the requirements? Would like to improve in areas where needed. Thanks!
The first thing I noticed is the color mix. You shot with three filters but the colors look mostly duotone in yellow and blue. More color variation and depth is a good thing (within reason). As for how to get there; there many paths.
One thing I’ve done a lot of lately is to use an RGB image as the base and add continuum subtracted H, S and O to the RGB color channels in various strengths till I get the color mix I like. I often split the S between Red and Green to let that present as yellow. I also add a little H to the Blue channel to preserve a magenta hue.
So many options…
Kevin
Thank you Kevin. I will most definitely add RGB and CS method will be done. There is more than yellow and blue tones in that image, but you are right, very slight. This is great to know for future images I will present. Appreciate the feedback🙏
Spencer · Jul 20, 2026, 12:50 PM
Using my latest image IC1396 as an example. Does it not meet any of the requirements? Would like to improve in areas where needed. Thanks!
I think this is just one of those things… It’s weird sometimes, but because we don’t have the context of all the other submitted images, we can’t really assess how we will do.
I have submitted images that I thought were a sure thing that got next to no promotions, and I’ve submitted other images that were good, but I didn’t hold too much hope for them, and they’ve landed a Top Pick..
If you don’t get a TPN or a TP, you don’t get any real context about what other images it was competing with - but I can tell you this, I’ve got a TP recently, and when I took a look at the other TP’s for the day, I’ve looked at my image and wondered how it landed amongst such incredible company.. Perhaps I am over critical of my own images sometimes.
In any case, I wouldn’t put too much concern into how your image goes in the TPN/TP/IOTD queue… They are a nice bonus or spot of recognition for your image, sure. But, if you enjoyed acquiring and processing your data, and you’re happy with your final result enough to choose to submit it for consideration - you’ve already won!
Thank you for explaining Alex. I’m not worried about the award so much, but rather the acknowledgment from experienced eyes that I’m doing things right with processing. It’s always a learning platform and always looking to improve. Hearing the different inputs really helps out.
SemiPro · Jul 20, 2026, 08:49 AM
Since I’m here I will also mention planetary nebulae. The nebulae itself tends to get all the focus, so you end up with a good looking planetary nebula surrounded by a poorly processed background. This has sunk many a PN image in the IotD process.
It depends. I do a lot of those, some quite small. Some have substantial background and others have little or none, especially if the FOV is small, which it often is on small planetaries, which are numerous. There only two ways to know what the background actually is:
1) The evaluator is already familiar with the object.
2) They have compared the image to other images of the same object.
As one would expect, #1 is seldom true for the smaller and less imaged planetaries and as far as I know #2 is seldom done due to time constraints.
I have always lobbied for a system that enables easy comparison to other images of the same object since that would seem to be the only truly valid metric.
Bill McLaughlin · Jul 21, 2026 at 03:22 AM
SemiPro · Jul 20, 2026, 08:49 AM
Since I’m here I will also mention planetary nebulae. The nebulae itself tends to get all the focus, so you end up with a good looking planetary nebula surrounded by a poorly processed background. This has sunk many a PN image in the IotD process.
It depends. I do a lot of those, some quite small. Some have substantial background and others have little or none, especially if the FOV is small, which it often is on small planetaries, which are numerous. There only two ways to know what the background actually is:
1) The evaluator is already familiar with the object.
2) They have compared the image to other images of the same object.
As one would expect, #1 is seldom true for the smaller and less imaged planetaries and as far as I know #2 is seldom done due to time constraints.
I have always lobbied for a system that enables easy comparison to other images of the same object since that would seem to be the only truly valid metric.
What I am more talking about is backgrounds that would never be acceptable on any other image somehow becoming okay on images of PNs. Dust motes, artifacts on the stars, gradients, stuff like that.
To your second point (comparison with other images), that is actually recommended as part of the process in the submitters tutorial. However I think I advised looking at images from 2020 onwards because that is when the high QE, low read noise sensors really took to the mainstream. We might even have to move that up to the post-release era of BlurX and NoiseX just because of how much that also changed the game. Although pre-AI images are very useful to find out what is real and what is not.
Sorry I thought this was the noise thread, so I had to edit all that out haha
SemiPro · Jul 21, 2026, 03:36 AM
Bill McLaughlin · Jul 21, 2026 at 03:22 AM
SemiPro · Jul 20, 2026, 08:49 AM
Since I’m here I will also mention planetary nebulae. The nebulae itself tends to get all the focus, so you end up with a good looking planetary nebula surrounded by a poorly processed background. This has sunk many a PN image in the IotD process.
It depends. I do a lot of those, some quite small. Some have substantial background and others have little or none, especially if the FOV is small, which it often is on small planetaries, which are numerous. There only two ways to know what the background actually is:
1) The evaluator is already familiar with the object.
2) They have compared the image to other images of the same object.
As one would expect, #1 is seldom true for the smaller and less imaged planetaries and as far as I know #2 is seldom done due to time constraints.
I have always lobbied for a system that enables easy comparison to other images of the same object since that would seem to be the only truly valid metric.
What I am more talking about is backgrounds that would never be acceptable on any other image somehow becoming okay on images of PNs. Dust motes, artifacts on the stars, gradients, stuff like that.
To your second point (comparison with other images), that is actually recommended as part of the process in the submitters tutorial. However I think I advised looking at images from 2020 onwards because that is when the high QE, low read noise sensors really took to the mainstream. We might even have to move that up to the post-release era of BlurX and NoiseX just because of how much that also changed the game. Although pre-AI images are very useful to find out what is real and what is not.
Sorry I thought this was the noise thread, so I had to edit all that out haha
Haha! I was just on the noise thread and getting it confused with this one! But I do think I’d stick to the most recent 3-4 years to find awarded images that still relate to the current expectations. I know I cringe when I look at my old images…
Kevin Morefield · Jul 22, 2026, 01:00 AM
Haha! I was just on the noise thread and getting it confused with this one! But I do think I’d stick to the most recent 3-4 years to find awarded images that still relate to the current expectations.
Yes, I agree that any comparisons should be of the most recent images although I would not always limit that to those with awards since some of the objects do not have any awards (or very few) due to their unspectacular and/or tiny nature. That is sort of an illustration of part of the problem, really.