Ticket #293 (closed design issue: suspended)

Opened 12 years ago

Last modified 7 years ago

Persisting photometric flags

Reported by: jbecla Owned by: jbecla
Priority: normal Milestone:
Component: dbSchema Keywords:
Cc: Blocked By:
Blocking: Project: LSST
Version Number:
How to repeat:

not applicable


How should the photometric flags be implemented? Photometric flags we might need based on common queries include: saturated, binned1, binned2, binned4, blended, nodeblend, child, edge, bright, nopetro, too_large (in queries dbQuery023?, dbQuery028?, dbQuery030?, dbQuery033?, dbQuery041?, dbQuery042?). Should they be kept per source? Per object? If per object, do they apply to stacked images?

Change History

comment:1 Changed 12 years ago by TimAxelrod

  • Status changed from new to assigned

comment:2 Changed 12 years ago by jbecla

  • Component changed from unknown to dbSchema

comment:3 Changed 11 years ago by jbecla

Discussed at DataAccWG telecon 11/7/08. Flags related to mask planes in masked image are needed for DC3a, flags related to extended objects (eg child, nopetro) are needed for DC3b. Tentatively planning to implement using functions + no-name set of bits in schema.

  • Step 1: decide which flags needed in schema (Tim).
  • Step 2: try implementing functions (K-T or Jacek).
  • Step 3: update schema in EA (Jacek)

comment:4 Changed 11 years ago by jbecla

  • Milestone changed from DC3 Design to DC3a Apps DB Schema

comment:5 Changed 11 years ago by Tim Axelrod

  • Owner changed from TimAxelrod to rhl

They certainly have to be on a Source rather than an Object basis.

I agree that for DC3a it may be enough to have a one-to-one correspondence between photometric flags and mask planes, with the convention that if any pixel in a source footprint has a particular mask plane set, the source has that photometric flag set.

I think the mask plane definitions have to come from Robert, so I am reassigning this one to him.

comment:6 Changed 11 years ago by rhl

  • Owner changed from rhl to jbecla

These are per-Source bits; so e.g. saturated means that this Source has at least one saturated pixel in its Footprint.

Some of the others are not so closely related to the mask planes, e.g. nopetro which is related to a failure in measuring Petrosian quantities. Although I don't quite know when this particular bit will be added, I am confident that there will be bits set in DC3a that need to be saved.

comment:7 Changed 10 years ago by jbecla

  • Milestone changed from DC3a Apps DB Schema to DC3b Apps DB Schema

comment:8 Changed 8 years ago by jbecla

  • Status changed from assigned to closed
  • Resolution set to suspended

comment:9 Changed 7 years ago by robyn

  • Milestone DC3b Apps DB Schema deleted

Milestone DC3b Apps DB Schema deleted

Note: See TracTickets for help on using tickets.