OK I'm addicted now.

Just finished my first 200 (as "cents"). Almost relaxing. WAY easier than the Stardust particle search which was too taxing on me.

I asked about my image with 5 objects marked, and it looks like many of those are likely variable stars. Pamela told me it'll be extremely unlikely to see more than 2 real KBOs in a single image. But here's my favorite one so far: Two KBOs, and two asteroids! (three asteroids if you count the additional one in the next image to the right of this image)

Click to view attachmentClick to view attachment


Is it sad that I can sometimes recognize previous Fields of View when shifted or subtracted? Probably!

But still fun!

Does someone want to show me some distinct examples of variable stars verses KBOs? I think my hit-rate is too high.

Whooosh...!

Click to view attachment

Have to be honest, I'm pretty addicted too. Unlike the other projects like this, you actually get a "Yes!" Discovery Moment, which is what makes you keep clicking, and clicking, thinking "Just onemore... just one more..."

Dan, re variable stars: No real idea how to filter them out during the initial screening, and apparently neither does the project at this time; looks like they're going to rely on follow-up observations. I've been thinking that most variables are going to appear more point-like than a KBO, but this might not be a valid assumption based on the occasional glimpses of unprocessed imagery around the edges of the main images...they ALL look like KBOs there!

I would assume that all light sources in these images are points of light -- remember that we can't even resolve relatively big Pluto as a disk, so the size of KBO you're looking for wouldn't be resolveable as a disk with even the most powerful of telescopes. They are spread out not because they are disks but because the pointlike light is spread out across the detector a little bit for various reasons.

Yeah, you're right, Emily...it was my best guess.

I & obviously many others would still be very interested in knowing what a realistic detection rate should be. Probably even the project isn't sure yet, though; this is raw science, so of course there are always surprises in the data...

Maybe one reason is the same as explained in your Planetary Society Blog in relation to Deep Impact's exoplanet search?

I sent an email to EPOCh's Principal Investigator, Drake Deming, at Goddard Space Flight Center, to ask him how the search would work. It turns out that he plans to turn Deep Impact's flaw to his benefit....
That's what I call making lemonade from lemons. The camera blur spreads the point-source light from stars out over several pixels on the camera's CCD. Deming explains here how that helps.... "to get high signal to noise, we have to collect lots of photons from the star. That's where the defocus helps. Each pixel of the CCD has a limited capacity to collect photons before it saturates. With a defocused image, we have about 75 pixels collecting light for us, so we can collect lots of photons in each exposure without saturating, and that gives us the high signal-to-noise ratio that we need."

I think the point sources are not actually points primarily because of two reasons:
1) telescope point spread function
2) atmospheric turbulence averaged out over the exposure duration

Yes, variable stars and KBOs are indistinguishable by image size for the reasons Emily gives. But we can distinguish them after all the clicks are compiled because a KBO will produce a series of detections aligned along its orbital path, while variable stars have no such pattern.

Regarding detection rates, we expect there to be a KBO (on any orbit, whether accessible to New Horizons or not) in perhaps every 50th thumbnail, given their known density in the sky. So most of these point sources are going to be variable stars, but at the rate you guys are clicking through them, you've probably nailed quite a few KBOs too.

John

Timely response, John, thanks!

Okay, so I just did image 3000 with 622 flags for a detection rate of 20.7% (asteroids included, and I'd guess that those represent about 10% of my total, so my KBO rate would be around 18%). Looks like the project is expecting genuine KBO finds in 2% of the images, so my rate is 9 times greater than expected.

Mentioning all this as feedback to the project, is all. This is hella fun, John!

I'm finally gonna be able to see cross-eye stereo pairs after going cross-eyed staring at these images.

I had 5 objects which were perfectly round in one image. Clicked on them all and was sent to an error screen telling me that perhaps I was being over-zealous for selecting 5 objects in one image. But I KNOW they were right!

I'm certain that I am seeing the same star field multiple times per session. I try to limit myself to 50 - 75 screens per session so that I don't get a massive headache.

It would be nice if we could enlarge the image on the screen. I have a lot of screen real estate that is not being used. For us folks with older eyes, it would be a real blessing if we had the option to make the image 25-50% bigger.

Can you use keyboard commands (like the ones that make browser text bigger)? For example ctrl + or ctrl and mousewheel (on my keyboard at least).

Thanks - the CTRL+mousewheel worked this time. It didn't work the first several days I tried it. Perhaps updating my browser and clearing the cache reset something that was screwing things up.

You must have seen the same image I did with 5 objects. I also got the note about selecting too much, but talked to Pamela and she said that those are legit selections, but that they are most likely variable stars. I don't think the variable star database will compare with Kepler, but it might still be interesting to someone.

Is this what you saw?
Click to view attachment

Up to 2500 views now and this is easily the most artistic (and one of the least scientifically useful) images I have seen.
Click to view attachment

I don't know a lot about art but I know what I like.
Click to view attachment

That's a beaut cents! ...& Just think of the marketing value -postcards, coffee mugs, caps etc.

I just joined -found what look like two near field asteroids and ~24 variables/KBOs in 96 images. Let's see if my "find" rate stays the same for the next 96...

Up to now I haven't felt the need to break my viewing rhythm to use the click-on list available for comments on image quality, but for some reason a lot of horrible images (and not very artistic to my eye) have cropped up and I've been clicking on the "Simply terrible image" comment a lot. I gave up after about twenty unreadable images in a row. Tried again later and still about half the images are useless. I saw nothing like this in my first 2500!

Yuk!
Click to view attachment

We may be getting the leftovers no one else wanted (here's a good place to quit and loggout), or they may have been deliberately presenting the better images first in order to get the best return soonest, or both, or something else. I've been noticing it too, but I still get the occasional good image.

Just today, I had to quit when I got some kind of terrible SQL violation error. I'm heading back to see if I can get back on.

EDIT at 45 past the hour: Okay, I can't even get logged on. Anyone have any idea what this is about:

Warning: mysql_connect(): Unknown MySQL server host 'zoobuilderdatabases.cvqgcgieedcl.us-east-1.rds.amazonaws.com' (2) in /home/zoobuilder/public_html/icehunters/admin/database_functions/mysql_functions.php on line 5 Unable to select database icehunters. Please verify the name is correct in admin/zoo-config.php

(Software engineer here) That, there, is a failure of the Amazon AWS service Ice Hunters is using as their server backend. Or a misconfiguration to same.

AWS is basically a way to rent time on central computers. Unfortunately, sometimes, they go down. You also have to get the names exactly right.

Okay, thanks maschnitz.

Time to find something else to do tonight.

Just got through about 150 more mostly problem free images and then hit another patch of unreadable ones. Gave up again after about twenty. I hope my marking those as "Simply terrible" helps weed them out. If I knew for a fact that it did, I might hang in there longer and mark more "terrible" ones.

I just started paying attention to the counters at the lower left of the screen. The count changed by 5 to 8 after each time I clicked "done" so I guess that was about how many of us were counting at the time. The count change then dropped to 2 or 3 just before I quit. I guess others were dropping out as well.

When you first start counting, the count really jumps around. It takes a string of 50 or so image views before the count settles down to a fairly constant change.
Edit: Or not. Sometimes the count flips back and forth between trend lines a few hundred counts apart.
Click to view attachment

A near miss!
Click to view attachment

Close to you...
Click to view attachment
By the way, it's been clear sailing with nary a bad image on my current ice hunting expedition.

Two asteroids!
Click to view attachment

Overlapping KBO/VARs?
Click to view attachment

This is getting ridiculous! All this in one sitting?!
Click to view attachment

I find I'm marking about 1.5 objects per frame. From my own background in ML and data labeling (plus the note on the page to overlabel) I'm guessing this is just fine. It hugely reduces the space requiring followup, and that's probably all that's wanted/needed.

I have a bit more trouble with the labeling of image quality. I realize those are of only secondary value, but I do like to do things right.

--Greg

I've only been rating image quality if I get a streak of real garbage pics on the theory that it might be a bad batch. The occasional lemon...well, that just happens.

Here's today's best shot for me: asteroid & big honkin' KBO/variable star. John, question: Is there an expected detection rate for variables? Would like to know as a vector check for our respective discovery rates.

I'll bet that when you first start, they give you a bunch of pix that had likely KBO/Variables in them. That both lets you come up to speed and lets them get an idea of how good you are at it. Somewhere around 150 images, my find rate for KBOs dropped drastically. (I stopped at 300--my hand hurts.)

When I click the "My Icehuntings" link, I see that nearly everything I've marked has also been marked by 8 to 16 people, so I'll guess I'm on the right track.

The rules I developed for myself are more or less as follows:

1) Don't click really little ones.
2) Don't click if it has even a single black pixel completely inside it.
3) Don't bother with asteroids unless they're very obvious.
4) Just skip bad pages--don't try to say why. (Did I mention my hand hurts?) :-)

I think they ought to give us keyboard shortcuts for all the buttons; minimizing mouse movement really helps reduce RSI.

--Greg

I agree with all your rules save #3, Greg; I think that the faint ones are by far the most likely new discoveries.

Like many, I find this a really exciting project. I'm interested in how it was set up, and your surmises suggest that we may not have been fully informed about what we would be 'fed' and what would be done with our responses. If true, that's unfortunate. Are there things we don't know about how our input is being used? This is a technique employed by psychologists but surely not necessary here. So, NH team (and I know you are our friends), is Greg right?

As you can see in my last post I just went through a really productive stretch, and that was as I neared 4000 images viewed. I've been through dry stretches and productive stretches and stretches of unreadable trash, but I have no reason to think it's anything but chance.

John Spencer said, "...we probably will add artificial objects in the 2011 images when they are posted- it's important to add artificial objects to the data to test what fraction of objects of a given brightness we can actually find."

So there may be artificial objects in some of the images, but I doubt that the images are fed in any particular order to any particular person.

Why do you say that? It's a very sensible thing to do. When people are new to the task, it makes sense to give them plenty of well-understood images. I'd expect to have to discard the first few dozen from each person anyway. People are great at this sort of task, but a) they take time to master it and b] they benefit from a little help.

Once someone is up to speed, then it makes sense to turn them loose on the less-well-understood stuff. But almost all human data labeling protocols I know of have an enrollment period. There's nothing wrong with that.

--Greg

I think these are really bad subtractions. What I want is a simple "bad image" button to indicate that I gave up on it. I might have marked one really obvious one, even on a bad image, but even then it should be taken with a grain of salt.

I also want keystrokes. E.g. press the spacebar for "done marking" and (maybe) press "X" for "bad image." Anything to minimize use of the mouse.

Given nprev's comment that bright asteroids will already be known, I've quit marking them entirely. They take too much time, and they're not what we're there for.

--Greg

Nothing at all, if it's signposted so that participants are fully briefed about how it's working and when they're entering a different phase.

Tic Tac Toe
Click to view attachment

I'm baffled that anyone would care--other than curiosity about the process. If you look back through the record of how well your votes correlate with others, you'll find some with a dozen or more prior votes and others with no one but you. Clearly there has to be some algorithm that decides when to give you a pic to help validate and when to give you one that no one has seen before. Since part of the draw here is potentially getting credit as the discoverer of an object, it's essential that everyone gets a fair chance at new images, but that's about the only legitimate concern I can think anyone could have about any aspect of the process. Certainly the idea that participants have a "right" to know the selection algorithm makes no sense; the researchers need to be able to change that algorithm depending on how its working. For test purposes, they'd probably want to try out a new algorithm on a subset of the users. I see no reason why they should announce any of this, unless it was something especially cool that they thought people would be interested in.

--Greg

Greg, I would still mark the bright asteroids. Remember, "bright" is a relative term here; Alan said a few months back that the limiting magnitude of these images is 28 (!). Probably the brightest thing we see here is around 14th magnitude or so, which is the same approximate brightness as Pluto seen from Earth.

There's no fancy algorithm for who gets to see which images, other than to ensure that all images are seen by a sufficient number of people. However I think there may have been some prioritization of the order in which the images were presented after the public release, with some of the better subtractions being put up first, which might explain why some people are hitting a "bad patch" after initial smooth sailing.

We are still working through the 2004/2005 preliminary search data, because (as always happens, even when you plan for it...) the 2011 data are taking longer to process than we'd anticipated. So there are still no artificial objects in the data.

And yes, please click on the faint objects too- our chosen target is likely to be faint, unless we're really lucky. But almost everything in these images, asteroids as well as KBOs, is likely to be new, as nprev said. The limiting magnitude in the data currently posted is about 25.5, and most things you see (apart from those "artistic" super-saturated stars) are probably fainter than mag 18 or something like that.

I'm surprised myself at how many variable stars are showing up, but I have no idea how many we should expect- dammit, Jim, I'm a planetary scientist, not a real astronomer... It's possible that we are learning something new about variables, though- there probably haven't been many surveys that go this deep in the Milky Way.

The keyboard shortcut idea is a good one- I'll pass it along and see if there's any chance of implementing that.

Thanks again for everyone's help with this!

John

Thanks for the thorough response, John. "Dammit, Jim"...

Ditto from me. Much appreciated.

Likewise. I'll wait until they keyboard shortcut (if it comes) before doing more; I probably overdid it yesterday (I did 600 some) and I was in pain most of today. (I should know better, but I got into it.) :-)

--Greg

I heard the Subaru telescope was damaged by a coolant leak.

Does this impact KBO search?

Yup, we were on the telescope that night- the first night of a two-night run. Fortunately the problem happened at the end of that first night, so we got one night of great data, but of course we lost the second night. Our drive back down the mountain in the morning was delayed by about an hour as the telescope operator tried to figure out what the heck had gone wrong- he was getting all sorts of error messages that he'd never seen before...

That was our last Subaru run for the year, so no further impact on our program from the anomaly. But I hope it's fixed soon, for the sake of all the other observers.

John

Looks like it was ethylene glycol & water - good news is that it is non-corrosive, and if the water's pure, non-conductive as well. By the color, it looks like the same stuff GM recommends for their radiators.

http://www.universetoday.com/87245/subaru-...eaking-coolant/

Thanks for the update.

Hope they can get coolant mess sorted out and back to observing. Saw a picture of the mirror with the coolant on it, gosh there is a quite a bit of fluid.

Is the coolant for the electronics or the drive mechanism for the scope? (perhaps if it was for the electronics it would have less particulates entrained in it?)

The "reviewed"/"for follow-up" counters are showing now. The "for follow-up" rate is almost 30%! I'm really looking forward to the "confirmed new KBOs" counter...

I'm surprised that finding objects is so easy in these images. I'm assuming there were fewer background stars in 2004/2005 than in 2009/2010? Will the background in the 2011 data be similar to 2004/2005 or less cluttered?

So far it looks like a quarter of the objects found have been reviewed and about 1 in 3 of those reviewed have been marked for follow up.
(I'm assuming that asteroids are included in number of objects found but not in objects marked for follow up and fudged my estimate accordingly.)
Click to view attachment

The one thing I can say about background star density in the 2011 data is that it is more uniform- in 2004/2005 we were moving in and out of the dust lanes near the galactic equator, while now we're (slowly) moving away from the equator, through unobscured star fields. We're also using longer exposures, so we're picking up fainter stars. Nevertheless, most of the 2011 images show a fair amount of "clean" sky between the stars when seeing is good, so I think the KBOs won't have many places to hide.

John

Thanks for the preview. I'm enjoying going through the current images picking out "objects". I wonder how many will survive the great purge that awaits when comparisons with the 2011 data begin.

Has anyone taken a look at their own contributions to see how many have been marked for followup? Something like 90% of mine seem to be marked for followup (not counting the 100 I did just this evening). That seems awfully high, given that they haven't even reviewed most of the objects found yet.

Or maybe it does make sense; even if all of us are 90% accurate, but when we're wrong, it's random, then you'd expect that if 30 of us looked at the same plate, 27 of us might label the same object, and three would label random garbage. So we'd all see great personal results, even though 75% of the objects we identified weren't suitable for followup.

It'd be cool to see more stats.

--Greg