Asteroid Zoo Talk

Known object message doesn't show green circle

  • jflaming by jflaming

    I sometimes get a "these images contain a known asteroid" message, but the supposed green circle showing where they are doesn't appear.

    Posted

  • chrisy by chrisy

    I am getting the same! Not sure if it was just me or not. Pleased it's not my eyesight! 😃

    Posted

  • jfinch by jfinch

    Multiple times - same thing. Something is wrong with the software.

    Posted

  • escholzia by escholzia

    Yep - very frustrating.

    Posted

  • scibuff by scibuff

    I'm thinking that the problem could be that the message saying that there is a known asteroid is NOT correct - so there is no green circle to draw on the image

    Yesterday, I went through ~200 sets and about 5% of them had asteroids. Just now I got three of out 5 which suppose to have a known asteroid in them, but I don't see anything. Feel free to judge for yourself:

    http://talk.asteroidzoo.org/#/subjects/AAZ0000d7c

    http://talk.asteroidzoo.org/#/subjects/AAZ00003ct

    http://talk.asteroidzoo.org/#/subjects/AAZ0001026

    update: http://talk.asteroidzoo.org/#/subjects/AAZ000110s

    update: http://talk.asteroidzoo.org/#/subjects/AAZ0001met

    update: http://talk.asteroidzoo.org/#/subjects/AAZ0000fma

    update: http://talk.asteroidzoo.org/#/subjects/AAZ0001cfh

    update: http://talk.asteroidzoo.org/#/subjects/AAZ0001b5t

    So the idea is that the service which checks whether or not there is a known asteroid in the image set returns an incorrect answer (sometime). Whenever the answer is incorrect, the message about a known asteroid is displayed, but the green circle drawing tool cannot retrieve any data as to where to draw the circle so it either fails or doesn't even attempt to draw it.

    Update: now I've missed one on purpose and a green circle was shown, which support the hypothesis above ...

    http://talk.asteroidzoo.org/#/subjects/AAZ0002c9r

    I'm going start tagging all those without green circle as #incorrect_result

    Posted

  • sisifolibre by sisifolibre

    same problem, i am making a collection about this:

    http://talk.asteroidzoo.org/#/collections/CAZS000097

    Posted

  • dudrea by dudrea

    Same for me ! Thank's for your comments. I began to believe I was blind or stupid !

    Posted

  • xscode by xscode

    I tag with #knownasteroid #nogreencircle

    Posted

  • timread by timread

    Ah, should have read this thread before I posted the same question!

    Posted

  • sraden by sraden

    Hi, I'm a Zooniverse developer. Thanks for bringing this to our attention and sorry for the confusion. So the asteroid metadata on known objects distinguishes between known objects that would be very difficult for the user to visually detect and ones which might have been detected in this task. Lets call them "good knowns" for this purpose. The message "This subject contains at least one known asteroid (circled in green)" was supposed to display for good knowns, but you've detected a bug in that the message was display for any known object, but we only drew the circles for "good knowns" This from the information in scibuff's post above. Thanks! We are working on deploying a fix today for this.

    Posted

  • Staceva by Staceva

    Good to hear, I was beginning to think I was just terrible at spotting them!

    Posted

  • sraden by sraden

    A fix was deployed.

    Posted

  • sisifolibre by sisifolibre in response to sraden's comment.

    I am a user of Spain and sorry if I have not understood what you say or if I can not explain well, my English is not quite good.

    Firsth thank you very much for your work, second place....
    I understand that these are images were the programs have detected asteroids although they are very difficult to see with the human eye. I want to see where they are too!!
    It's possible that the program uses any mark or similar to tell us?.
    I have some dubts about if it may be good for the work we do here or not, but my curiosity is too much strong

    Posted

  • blyerkit by blyerkit

    I'm not sure if this might be slightly off-topic, but this feature has led me to being very cautious about declaring possible false positives. On a number of occasions I have declared nothing despite thinking that I might have something, but also that my brain might be just constructing motion from the flicker. Often when I have done this, I have received the message described above, with no green cirsle. Could this mean I am being too cautious?

    Posted