Changes between Version 9 and Version 10 of MOPS_IASC20130319_tests130422

Timestamp:

Jul 8, 2013, 10:43:26 AM (13 years ago)

Author:

Mark Huber

Comment:

--

MOPS_IASC20130319_tests130422

@@ -227 +227 @@
  * m,m (really same e,e)=1 -- can use for unconv recovery test as well
  * e,m mix=0
- * f,f and vf,vf =~83
+ * f,f and vf,vf =~83 -- out of ~500 objects is ~15% lost to just depth
  * c+f=0
- * f+e/m mix=~20 -- really means missed from f but uncertain so cannot use for faint test, will be part of edge test but because faint no good
+ * f+e/m mix=~20 -- really means missed from f but uncertain so cannot use for faint test, will be part of edge test but because faint it is not good, an additional ~5% loss
  * odd/?/offset mix=3
- * total=109 
+ * total=109 -- this means 109 objects would be lost with the 3/4 detections required from a total of ~500 objects, or ~20% total
  * some types could be double counted..
  * if f+m/e then likely still f missed with other pair -- just excludes for missed/found pair (see pos or neg only below)