Re: thanks Greg for the ID of Lacrosse 2 91-017A #21147

From: Greg Roberts (grr@telkomsa.net)
Date: Sun Aug 09 2009 - 15:15:57 UTC

  • Next message: Gerhard Groenewald: "apex"

    Afternoon Gerhard and other readers
    
    No problem in giving identifications but sometimes there may be a slight 
    delay - depends how close my nose is to the grindstone.
    
    First Ive heard about asteroids being classified -- dont believe it as there 
    is absolutely no reason to do such unless of course someone is worried we 
    might pick up a NEO on collision course with earth :-))).
    
    I tried solving your image with APEX but it was too noisy - one of your 
    previous images I did solve okay but I think the main problem is that the 
    field of view is too large in this case and the star trails too short -- ie 
    the image scale makes it difficult for APEX to find stars. What APEX does is 
    choose the 30 or 45 brightest stars from the star catalog used for the field 
    of view and try and find a match with 20 stars in the image. The problem is 
    "which are the brightest stars in the image" when they are all very much 
    alike so APEX just cant match enough stars to give a valid solution. Also 
    the magnitude scale in your image is not necessarily the same as the 
    magnitude system used in the star catalog for a  variety of reasons, so when 
    you have something like say 6000 stars from the catalog in the field of view 
    its easy to get the "30 or 45 brightest stars " wrong. In my early days with 
    APEX I used a catalog that gave something like 90000 stars in the field of 
    view from the catalog - the computer churned away for a LONG time as it 
    tried to match stars - in one or two cases it got it right but more often 
    than not it failed - one has to "sort of match" the field of view to the 
    star density in the catalog - too big a field of view makes this generally 
    difficult.
    
    Cheers
    Greg
    
    
    
    -------------------------------------------------------------------------
    Subscribe/Unsubscribe info, Frequently Asked Questions, SeeSat-L archive:  
    http://www.satobs.org/seesat/seesatindex.html
    



    This archive was generated by hypermail 2b29 : Sun Aug 09 2009 - 15:17:08 UTC