Asteroid Zoo Talk

Viewing tips (for novice seekers or new tips for the others)

  • meegja by meegja

    Here some tips to make viewing and detecting easier. Well, for me at least 😃 so maybe also for others. For some, some points will be obvious but maybe not for others:

    • Under Tools, check "Invert". This gives an negative view which actually makes detecting way easier.
    • Start with the Cycle view first (left from Play button). When an asteroid is there, a "rainbow" like moving stripe will be visible.
    • Hit Cycle again to stop it.
    • Do a "double check". The Cycle option is just showing so much, a lot of things are not visible in it. Do not use the Play button but instead move the slider with your mouse from left to right and back and repeat that in different speeds. This way the frames are showing much faster, giving more FPS and making any movement way more obvious.

    Posted

  • Chipuk by Chipuk

    Thanks meegja. I used invert and double check, but wasn't aware of cycle view - cheers

    Posted

  • dudrea by dudrea

    You also can use arrow keys to change frame. Easier and faster for me

    Posted

  • Emili_Sancha by Emili_Sancha

    ...and numbered keys to jump over dark frames. ex. view blinking frames 1 and 4, because 2 and 3 frames are only black...

    Posted

  • mrpsb by mrpsb

    Great tips, I just spotted one I'd have otherwise missed using the above!

    Posted

  • Devilstower by Devilstower

    The "scrubbing" using the slider rather than the Play button is a great tip. Play cycles too slowly to be useful.

    Posted

  • Timecruiser by Timecruiser

    Is it just me, or is the Cycle option totally useless? Because it darkens the total image, making the tiniest points invisible -- the very ones we need to see if there is an asteroid in the frame set. I'm new, but I have found several asteroids so far using the Invert option. Most were already known. But when I find one I go to the Cycle option and the asteroids don't show up there.

    Posted

  • Dr.Asteroid by Dr.Asteroid scientist, admin

    Thank you for working out these details - we each had a way of looking at it during development (I'm a invert and just let cycle guy), but by far it will be most useful if you tell us what works best. I'm working on a general guide that will be something that everyone can refence - and ideally add to.

    Posted

  • Emili_Sancha by Emili_Sancha

    Cycle color works fine but darkens a lot the images, maybe a brightness slider would help...or work in inverted image.

    "Play" works very slowly. we have a trend to use the mouse to run back and fordwar into images... maybe a speed slider would work fine.

    Posted

  • dudrea by dudrea

    What works best for me ? It's the function "play" with inverted view when images aren't not too much misaligned (not enough often alas !)
    I also need a "speed slider" and an optional grid would be nice not only to locate asteroid, but also to help browsing all the screen
    Never been able to see anything with cycle view.
    I stopped marking artifacts : too long, too annoying with the given tools and not sure that's useful..

    Posted

  • meegja by meegja

    I too did stop marking most obvious artifacts for now. I think that it's way better when you only have to click Artifacts once, then mark them in all frames and then Done. Or at least make the same system for Artifacts that Asteroids use.

    Posted

  • djsato by djsato admin

    Use keybindings! You can now use keys 1-4 to flip through the frames manually. This is especially useful for skipping any bad frames in the set. In addition, the space bar toggles the pause/play button.

    Some of you may have noticed that this functionality was around before, but it you needed to first click on the relevant area of the interface. Now you can use the keys directly.

    Posted

  • meegja by meegja

    @ miltonbosch

    1. About the the sets which are very noisy/grainy: color cycling is only usefull on those sets for the most obvious asteroids. But faint ones can be spotted: in invert/negative view they can be spotted when squinting your eyes and manual movement of the slider.

    2. I agree that marking artifacts is important. But the way it works now is just way to much work to mark all of them. Around 95% of the work is marking artifacts and the way it works now, I am developing a real case of RSI: 4 artifatcs in a frame times 4 frames is 16 markings in 1 set. That is 16*(click Artifact + click on artifact spot + click Done) + (4*select frame).
      So for now I stand on marking the artifacts that could be mistaken for an asteroid or alike.

    Posted

  • meegja by meegja in response to djsato's comment.

    Could be nice. But a lot of people can not use key-bindings in their browser: they have set their browser in such a way that if they start typing outside a text box like this, the word they are typing is searched on that page. I have that setting too (I use Firefox): when I am not in a text box and on a big page and looking for parts where "telescope" is mentioned, I start to type that and the browser is highlighting that word. Then with the F3 key I cycle through multiple mentions of the search.
    So when using the keys 1 - 4 here, it starts to search for 1, 2, 3 or 4 in the text instead of flip through the frames. I know the arrow keys do the same but then I first have to click a frame number and then the arrow keys can be used.

    Posted

  • stonepenny by stonepenny

    I use the slider first, for artifacts, but wonder if their crosses then obscure for the asteroids? Then I do play, and visually scan, to see if anything moves in relation to anything else. I do find inverse helpful, with play, too. Never got the hang of the colour bit... Are asteroids always completely linear, or do their paths wobble a bit, like gravity making them curve? Also, do they always travel at a constant speed? This shows with the distance travelled between each view.

    Posted

  • stonepenny by stonepenny

    agree about the artifacts... doing them like the asteroids would be much easier, and faster. (Same problem with ravens on Condor Watch!)

    Posted

  • stonepenny by stonepenny

    would be helpful to have the option of what you've clicked on each screen you could see on all - at present, changing screens (ie invert) looses your crosses.

    Posted

  • meegja by meegja in response to stonepenny's comment.

    As I understood it, the time between each frame is around 10 minutes. So if an asteroid or other planetary object is there, the speed is constant and the path is straight. Their could be a difference in speed or path but then the times between frames would have been months.

    Having said that ... there are always unlikely or very rare events. Maybe 2 asteroids at the critical point of colliding or very close encounter, maybe that would show as non straight lines and different speeds over a period of 30 minutes?
    Or a spacecraft of some kind with physical change of speed and path? You never know 😉

    So when you use Color Cycle and you see a kind of worm like straight movement cycling in a kind rainbow color and some distance in between the worm parts, then it's most likely a asteroid. If you see a color change though that is (almost) fixed at 1 point, then it is most like an artifact. Same goes for when you see a kind of random color movement.

    Posted

  • stonepenny by stonepenny

    meegja, thanks for the reply. So, all asteroids are expected to travel at the same speed/distance... I''ve clocked some that are over 4 photos, yet different distances between frames. Perhaps size of asteroid has an effect on speed, too...

    Posted

  • meegja by meegja in response to stonepenny's comment.

    Yes, within the sets we view, the distance between frames should be about equal. If you have a set where there is movement of something but the distance does differ, it's most likely some kind of artifact.

    Having said that: some sets of pictures are very grainy, they have a lot of "noise" in it. In those sets it's sometimes very hard to see clearly what happens. So in those sets it can look like the distances do differ: between frame 1 and 2 the distance can look bigger then between frame 3 and 4. As long as it is not "jumping" around, that too can be an asteroid.

    But no matter what: if you think you spot an asteroid, mark it as one. It could be a mistake but it also could be right. There is no wrong or right with this kind of observing, it's just acting and marking on what you see. I too have sets where I was really in doubt, so I marked it. You never know 😉
    If you got it "wrong", no harm is done. All these sets are viewed and marked by many others. And not only now but also in the future. Only when in a set the same object is marked by different people as an asteroid and in the future the same happens, then it becomes clear.
    And you never know: maybe you spot one that is very hard to see and is also marked by some others. And in the future the predicted path is confirmed, maybe your name (along with some others) will be shown with that asteroid 😉

    Posted

  • meegja by meegja

    For those interested, I made a tutorial video. See http://youtu.be/qtQ7w-d9BhU

    It's 38 minutes long and it covers most of the basics. It has a click-able chapter jumplist, best viewed in HD.

    Think it's useful for beginners but maybe more advanced users can benefit too. It mostly handles the use of it and different kind of artifacts. And of course the spotting of asteroids 😃

    First time I ever made a tutorial like this so don't expect a teacher-like tutorial ... and yes, I seem to say "Ehhhm ..." a lot 😉

    And no, my English is not that good, my native tongue is Dutch, so ...

    Posted

  • artdent42 by artdent42

    Many browsers allow you to hold down the CTRL key and then press the + key to magnify the image. I find that a useful way to examine a small area . Remember to use CTRL - to go back to the original size.

    Posted

  • Cpt._Pete by Cpt._Pete

    My simple tips are:

    1, If it moves in a pattern like the No4 ( Up,Down and left,right) then it`s an Artifact regardless of what it looks like.

    2,If you THINK its moving ,click Asteroid ,place the markers for all 4 picture BUT DONT hit done
    then cycle throu the images and see if it moves ,if it does click Done ,if not click Cancel

    Posted

  • stonepenny by stonepenny

    Cpt. Pete - I agree, I do these, too. Cycling through also helps to confirm if the placed marks are linear.
    I've also found sometimes it's easier to mark out of numerical order, so the 4th may be done before the others, etc., especially if marking starts on two, not one. I sometimes find the circle a pain, as it can cover where I want to mark. 😃

    Posted

  • theSkipper by theSkipper

    One tip that I've not seen mentioned: all astronomical objects are blurred by atmospheric turbulence, forming the "seeing disc". Stars and asteroids, which are virtually point sources at the distance we're viewing them, form images with soft edges. Anything with a hard edge is an artefact!

    The mathematical function describing the seeing disc is the same for bright and for faint stars, but there is an outer limit of detectability which is further from the centre for bright stars, causing their images to be larger. Anything smaller than the seeing disc (compare the object with a genuine star of the same brightness) is an artefact!

    Posted

  • gordon.colliergmail.com by gordon.colliergmail.com

    sorry if this is a simple question but how do you mark potential multiple objects on the same scene?

    Posted

  • dudrea by dudrea

    You can, successively mark more than one artifact (or asteroid)
    But may be I'm not understanding what you mean by "multiple object"

    Posted

  • meegja by meegja in response to gordon.collier@gmail.com's comment.

    If you mean potential multiple asteroids: just start marking the asteroid with the Asteroid button and mark it in each frame and click the Done button. If there is a 2nd asteroid simply click the Asteroid button again and mark it in all frames and click the Done button. If there is a 3th asteroid: again click the Asteroid button ... and so on 😃

    Posted

  • stonepenny by stonepenny

    I'm finding the crosses large, they keep obscuring the next position to be marked. To keep the photo centres clearer, I'm marking line artefacts at the edges of the frame, where possible. I've noted, although the Asteroid Zoo info says the asteroid paths should be linear, some of the known ones paths are not ruler straight! I've also seen some small dots that may be planets, not asteroids! I marked as asteroids, having no other option, and put them in favourites. 😃

    Posted

  • stonepenny by stonepenny

    Image AAZ0000sm4 - sometimes asteroids are dark - going left to right above 2 suns, just right of centre of pic.

    Posted

  • meegja by meegja in response to stonepenny's comment.

    No, they can not be dark. If they are, they would have been made out of something that is anti-light 😉

    Every solid body in our solar system that we can see with the naked eye or through a normal telescope, is only visible because it reflects the light of our star, the Sun.

    Therefore, if you see a black spot, it's an artifact. Or, like in the set you give ( http://talk.asteroidzoo.org/#/subjects/AAZ0000sm4 ), it's most likely caused by the noise/grain of that set, giving the illusion of movement.

    Posted

  • stonepenny by stonepenny

    finding again and again, sneaky asteroids peeking out from behind a star, including known ones!

    Posted

  • stonepenny by stonepenny

    Image AAZ0001wdg
    asteroid coming out below binary star on right - sneaky! (frames 3 n 4)

    Posted

  • stonepenny by stonepenny

    meegja - by dark, I meant, only part of the asteroid is lit by star light, the rest is dark. So, you mean, an asteroid will be lit up regardless? It can't be part light, part dark, like a crescent moon?

    Posted

  • meegja by meegja in response to stonepenny's comment.

    Stonepenny: asteroids are very close. The asteroid belt itself is approximately 251,000,000 miles from the sun as the closest star (Proxima Centauri) from the sun is approximately 4 light years (23,514,502,164,993 miles). So if you see an asteroid moving, it will move in front of a star.

    Ofcourse not all asteroids are in the asteroid belt but the same principle goes for those.

    And an asteroid is always lit up by the sun, the light coming from the stars is simply not strong enough to lit up an asteroid. And if it could, it would lit up the asteroid 100% since stars are all around but at staggering distances: from 4 light years to billions of light years distance.

    And an asteroid can be partly lit like a crescent moon but then the position of the asteroid is between the Earth and the Sun, like the planet Venus which has phases like the moon. But then also the asteroid must be very big to see it. Most asteroids in the position between the Earth and Sun (and there are not a lot of them) are simply too small to see the crescent effect.

    Posted

  • stonepenny by stonepenny

    Just a suggestion... some frame sets allow more movement than others, so I look out for potential asteroids moving more than the frame set, although this is not always the case!

    Posted

  • stonepenny by stonepenny

    when I think I've found an asteroid, I put it on the brightest frame, then use the colour cycle - if it changes colour, red/green/etc, then it should be an asteroid. Hope this helps 😃

    Posted

  • stonepenny by stonepenny

    just a suggestion... some of the frames are so blurred, you can't make anything out, so, if you put viewing into 'invert', you may be able to see all the frames clearly.

    Posted

  • OldKingSol by OldKingSol in response to stonepenny's comment.

    Some asteroids have dark material on parts of them, so as they move through space and spin, they reflect different amounts of light at different times. So yes, you will sometimes have asteroids whose apparent brightness appears to vary - though that still won't show up as a crescent, it will still look like a blob of light. These, I believe, are likely to have once been part of a planetary ring system (one of the moons of Jupiter displays this trait quite prominently, though I don't remember which exact one at the moment - one side is dark, the other is light). Just mho, oc. 😃

    Posted

  • toonic by toonic

    yeah full ack to OldKingSol

    Posted

  • stonepenny by stonepenny

    for those who get black/fuzzy frames, try the invert, sometimes it helps settle the frames, so you go back, and they're viewable.
    Also, I use the 'cycle' for colour confirmation - an asteroid, although not necessarily clear against the black, will turn colour as the cycle runs, so will be clear red, green, etc. 😃

    Posted

  • nuppard by nuppard

    Is classifying meant to work with an iPad using iOS 7 and Safari? The reason I ask is when I've selected the option and also marked it as an asteroid or artefact and then click done - it then tells me I need to mark the asteroid/artefact first as if my mark has not been recognised even though I can see the cross hair image on screen. If this software does not work with iPads, it would help if you make this and other limitations much more obvious before wasting peoples time.

    Posted

  • Andy_Arg by Andy_Arg in response to meegja's comment.

    ...move the slider with your mouse from left to right and back and repeat that in different speeds... VERY, VERY, USEFUL TIP!!!

    Posted

  • std.nrd by std.nrd

    I zoom in (in the browser) 200%. MUCH easier to see, especially if like most people you have a newer "retina" display or something similar

    Posted

  • meegja by meegja in response to std.nrd's comment.

    That zoom tip indeed is a good one!! I am on a 27" Iiyama LED screen and zooming in with the browser indeed makes it easier. Bit more change on false positives though cause with the grainy/noisy images sets, you see more pixel/noise movement. But great tip, thanx!

    Posted

  • calin_uio by calin_uio

    Here's a very useful VB Script I found on the forums to run the scrolling:

    set ws = CreateObject("WScript.Shell")
ws.SendKeys "%{TAB}" ' return to parent window
for i = 1 to 40
ws.SendKeys "1" 
wscript.sleep 80
ws.SendKeys "2"
wscript.sleep 80
ws.SendKeys "3"
wscript.sleep 80
ws.SendKeys "4"
wscript.sleep 80
next

    The images aren't all stabilized, I wish they run some software stabilization algorithm before putting the images online. Such a pre-filtering software would also be able to handle the #badset problematic. I hope I am wrong but don't think there is a single asteroid in any #badset image.

    The timestamp of the images would also be interesting, since the Catalina survey go way back to 1990, and I am curious how much the optics and image quality did increase over the years.

    Posted

  • dudrea by dudrea

    Very useful script !!! I always use it. but forgot to thanks the author... What I do now.

    I wrote another one useful when I want skip one (or more) bad frame.

    set ws = CreateObject("WScript.Shell")
    ws.SendKeys "%{TAB}" ' return to parent window
    Dim nb,K1,k2,K3,K4
    k1=1
    k2=1
    k3=1
    k4=1

    k1=UserInput("Enter zero for skip key 1 one otherwise")
    k2=UserInput("Enter zero for skip key 2")
    k3=UserInput("Enter zero for skip key 3")
    k4=UserInput("Enter zero for skip key 4")

    J=50
    while j > 0
    for i = 1 to j 'repeat the process
    if k1 = 1 THEN
    ws.SendKeys "1" 'press 1 to display frame 1
    wscript.sleep 80
    END IF
    if k2 = 1 THEN
    ws.SendKeys "2" 'press 2 to display frame 1
    wscript.sleep 80
    END IF
    if k3 = 1 THEN
    ws.SendKeys "3" 'press 3 to display frame 1
    wscript.sleep 80
    END IF
    if k4 = 1 THEN
    ws.SendKeys "4" 'press 4 to display frame 1
    wscript.sleep 80
    END IF

    next
    j=UserInput("Nb iterations sup ? : ")
    wend
    Function UserInput( myPrompt )
    ' This function prompts the user for some input.
    ' When the script runs in CSCRIPT.EXE, StdIn is used,
    ' otherwise the VBScript InputBox( ) function is used.
    ' myPrompt is the the text used to prompt the user for input.
    ' The function returns the input typed either on StdIn or in InputBox( ).
    ' Written by Rob van der Woude
    ' http://www.robvanderwoude.com
    ' Check if the script runs in CSCRIPT.EXE
    If UCase( Right( WScript.FullName, 12 ) ) = "\CSCRIPT.EXE" Then
    ' If so, use StdIn and StdOut
    WScript.StdOut.Write myPrompt & " "
    UserInput = WScript.StdIn.ReadLine
    Else
    ' If not, use InputBox( )
    UserInput = InputBox( myPrompt )
    End If
    End Function

    Posted

  • Meanjean4321 by Meanjean4321

    i found that when there is a light color background i tip my computer back a little so more are visible

    Posted

  • Gill_M by Gill_M

    Thanks for all the wonderful tips and the thoughtful sharing folks. It is great to get a more positive and constructive vibe as some conversations seem to be reflecting people's unhappiness with the lack of asteroids and the lack of feedback. I've been feeling this myself and I'm noticing that fewer people are logged on each time. Today I am the only person here and it feels lonely...all right I'm in New Zealand so it always feels lonely and I guess you folks are mostly asleep or working but still!

    Anyway my low tech tip for RSI reduction is to put up the four frames at once gizmo and do the artifact marking there which should save about three whole clicks 😃. I feel it is nice to use this gizmo now and then as it's pretty redundant except for those bad sets that jump around horribly.
    Even more low tech... I have drawn up a grid on a clear overhead projector sheet in cm squares which I use when the sets are bouncy. I put the frames on four up and then use a large star or two as a reference anchor and then check each square. Really laborious but if like me you dread missing things then it might suit you.

    Happy hunting all.

    Posted

  • Puppyhogg by Puppyhogg

    How to look up the map coordinates?

    Posted

  • sisifolibre by sisifolibre

    To mark artifacts more easy: use 4-up view

    anyway a keyboard shortcut would be appreciated to quickly mark all artifacts. For example: press A to enable tool "artifact", you press A to confirm the selection

    Posted

  • scottakendall by scottakendall

    I assume that the number is extremely low, but what is the percentage of asteroid discoveries per image set?

    Posted

  • artdent42 by artdent42

    That's a tough question to answer. It seems to depend upon what area of the sky was used for the sets we are given. Lately, I would guess it has been about 3 to 4 per 100. There was a time when the sets were from a prime area of the sky and it was more like 1 per 5. Twice I ran into 3 asteroids on the same set.

    Posted

  • MvGulik by MvGulik

    Personal Classifying habits:
    Note: While using high animation frame rate(s), and being a bit obsessed with trying to catch the hard / faint one's.

    Inverted view:
    Used 99% of the time.
    Human eyes are just way better in detecting faint brightness changes in the light(er)-range than they are in the dark-range.
    If those obscure cases are not your thing / target. Using normal view might actually be less distracting.

    Normal (not inverted):
    Only used when the inverted image's are actually worse than the normal view.
    rare, but it happens.

    Cycle view:
    Don't though it anymore.
    Its detection range is limited to major visible asteroids only, ... that jump out at high(er) frame rates anyway. Ergo: A waist of time.

    Manual flipping (mouse + image bar):
    Used to fine-check if a potential "maybe case" target is not showing any none asteroid like behaviors.
    Ergo: Consistent movement in only one direction with a fairly good linear character.
    (Falling back to 1..4 number keys in case of a set with nasty offset or brightness changes.)
    Note: The image-bar can also be manipulated with the left- and right-arrow keys. But it needs to have(be) focus(est) first. (Not useful if the classifying page is zoomed-in to the point that its having a horizontal scroll bar.)

    4-up view:
    Only used as quick loader + checker so see if both inverted- and normal-view are both missing the same frames.
    its rare a image actually fails to load, but it happens.
    (As the marking accurate in 4-up view is obviously less than with single view, I don't use it for marking stuff.)

    Play button:
    Not used at all ... of course.

    Posted