Astrometry in browser

[zemogle]

Currently working on this in branch feature/astrometer. The key features are:

• [x] Show thumbnails for block
• [x] Blink/cycle through images
• [x] Switch between fullsize images by clicking thumbnail
• [x] Switch between images in sequence
• [x] Plot sources positions on image from backend positions (in x,y)
• [x] Add dragable aperture to each image to align with target
• [x] When dragging aperture, see zoomed section of the image
• [x] Report position of target aperture at required resolution/scale
• [x] Need a list of sources for each frame
• [ ] Need a list of moving objects (if any are found) for each frame

[talister]

Hi Edward, Sarah and I have been working on the bottom two points. We are putting the finishing touches to the code to add zeropoint calibration to frames which don't have them - the code to extract info out of the SExtractor FITS catalogs is done.

What we need is somewhere to store the info.. I have made a start on the feature/CatalogSources branch (it needs a link to SourceMeasurement table and a migration creating) but it seems pretty duplicative of the SourceMeasurement table and it's making me twitchy. Would be interested in your thoughts before I plough ahead.

[zemogle]

I can understand your twitchiness. I think CatalogSource is sufficiently different to SourceMeasurement that having 2 models is justified. We will get huge numbers of CatalogSources very quickly. However I don't think we need to keep these forever. We could do some housekeeping and remove all CatalogSources which belong to Frames > Blocks which we have reported?

Ps. Do you die a little bit inside when you have to spell "catalogue" that way?

[talister]

OK, I will press on with Catalog(ue)Sources then. It would be nice to remove the duplication of the astrometric and photometric catalogs and move those out of CatalogSources and SourceMeasurement and into Frame - not wholly sure the mass data migration is worth the pain though. Yes, I think an auto-vacoom will be needed and a good idea.

(I don't tend to notice the American spellings so much now and given that my naming decision of 6+ years ago for the AstroCatalogues directory is still causing grief (see recent Network Changelog), I will go with the path of least resistance...)

[zemogle]

How are we getting on with this? It would be good if we could pull in the parallel developments soon.

[zemogle]

Made lots of process on feature/astrometry this week. We now have

• Loading of candidates into analyser
• Blinking images per candidate
• Accept/reject by candidate
• Submission of accepted candidates
• Create (or update as appropriate) SourceMeasurement on submit
• Show accepted candidates in MPC style report
• Email MPC (temporary values for header and to: address at present)
• Unit and Functional tests

To do:

• [ ] Regularly update candidates with detections (most of the work for this has been done already)
• [ ] Allow user to reposition candidate marker and update RA/Dec