[PINPOINT-994]  Too many false solutions
Type Change Request
Priority High
Severity Major
Component Astrometric Engine
Fixed In Version [6.0b] 6.0b
Versions Affected [5.1d] 5.1d
Severity Closed
Resolution Complete
Reported By Bob Denny
Resources Bob Denny
Start Date 8/27/2013

Description 
Over the years the conditions have changed to include bigger sensors and faster optics, resulting in wider FOVs. This can allow PinPoint to be more picky about matching. There have been reported a small number of false solutions. At a minimum increase the base MIN_MATCH_STARS from 8 to a higher number. In addition, probably increase the output of the heuristic that auto-chooses Plate.MinMatchStars based on the number of available catalog and available image stars.






This will require LOTS of regression tests!! 

Comments
2/12/2014 4:15:07 PM   Bob Denny 

 

  SVN Comment
 

 

  Author
  rbdenny
 

 

  Repository
  svn+ssh://rbdenny@a2_svn_dc3/home/rbdenny/svn/astro/pinpoint
 

 

  SVN Revision
  122
 

 

  Affected files
  /trunk/engine/PinPoint.rc (Modified)
/trunk/engine/PinPoint.vcproj (Modified)
/trunk/engine/Plate.cpp (Modified)
/trunk/setup/PinPointSetup.vdproj (Modified)

 

 

  Check-in comment
  --- PinPoint 6.0b ---
Fix typo in rough plate scale tests during iteration. Upper limit was too tight. Now it really is "twice" corresponding to "half" on the low end. Engine 6.0.5.
GEM:994
2/12/2014 3:32:45 PM   Bob Denny 
It was a simple error. The The high side of the test should have been "twice" corresponding to the low side "half". It was 1.5, and causing failures on streaked images.
2/12/2014 3:15:14 PM   Bob Denny 
Reopened today due to this Comm Center thread by Stan McQueen. I suspect I know what the issue is... still too tight limits during iteration. If this is the case it will be relatively safe and simple to fix. Cross our fingers...
10/23/2013 8:27:26 AM   Bob Denny 

 

  SVN Comment
 

 

  Author
  rbdenny
 

 

  Repository
  svn+ssh://rbdenny@a2_svn_dc3/home/rbdenny/svn/astro/pinpoint
 

 

  SVN Revision
  115
 

 

  Affected files
  /trunk/engine/PinPoint.rc (Modified)
/trunk/setup/PinPointSetup.vdproj (Modified)

 

 

  Check-in comment
  Beta 3 - Fixed plate scale test in just-previous commit GEM:994
GEM:859
9/24/2013 4:04:01 PM   Bob Denny 

    
        

            SVN Comment
        

        

            Author
            rbdenny
        

        

            Repository
            svn+ssh://rbdenny@a2_svn_dc3/home/rbdenny/svn/astro/pinpoint
        

        

            SVN Revision
            104
        

        

            Affected files
            /trunk/HTML Help/PinPoint.mdb (Modified)

            /trunk/HTML Help/doc/PinPointRELNOTES.htm (Modified)

            /trunk/Regression Test/FalseSolution/FalseMatchSample June 2013.fit.stars (Modified)

            /trunk/Regression Test/Test1/File.History (Modified)

            /trunk/Regression Test/Test1/test1-PSF-RemHotPix.vbs (Modified)

            /trunk/Regression Test/Test1/test1-RemHotPix.vbs (Modified)

            /trunk/engine/Plate.cpp (Modified)

            /trunk/setup/PinPointSetup.vdproj (Modified)

            		
        

        

            Check-in comment
            Final touches on false solution resistance and docs. GEM:859
9/15/2013 9:51:41 AM   Bob Denny 
Reduce absolute minimum back to 8, instead increase fraction from 0.2 to 0.25. Doing regression testing and it is looking good. This is the ticket where I will accumulate the regression testing time.
9/15/2013 9:00:08 AM   Bob Denny 
Increased absolute minimum to 10.