A suite of tools written in Pyraf, Astropy, Scipy, and Numpy to process individual QuickReduced images into single stacked images using a set of "best practices" for ODI data.
I am currently working on improving the aperture correction step of the odi_phot_process.py script. The aperture correction function lives in full_calibrate.py. The current issue is that the standard deviation of the correction is still too large, even with filtering by peak counts from a fwhm measurement. I am going to see if I can implement a sigma clipping step to improve the standard deviation without throwing out too many stars.
final images should ideally be North-up, East-left and be pixel aligned for ease of use. Trimming will also reduce the size of final image by a significant amount.
Trimming should try to return the maximum possible area while avoiding any cell gap artifacts at the edges.
Mosaic filters at WIYN have smaller fields of view than the broadband SDSS filters, and so don't need to use all of the information provided in the full format images. Can we specify just to use OTAs covered by the filter and things will work out? Will there be other issues with narrow band images -- no photometric calibration for Hα for example. Finding out by trying to process data for J. Salzer.
some fields would benefit from an illumination correction, but can't be properly flat-fielded because the nature of the data makes it difficult to construct a dark sky flat. We should provide a standard master dark sky flat to users to use in these cases, constructed from high quality data.
Use artificial star tests to produce tables of detection completeness at a range of instrumental magnitudes. These tables can then be used to generate completeness limits in color-magnitude space.
Just had a case where ota_sourcefind only found one source, which threw an error when trying to apply an np.where filter. Adjusting the detection threshold fixed this, but might be better to skip otas with so few bright sources. I will add the functionality to deal with these cases.
SQL queries are currently disabled on the SDSS website, so HTTP queries need to be made instead. These HTTP queries are returning different magnitude values than the earlier SQL queries, but identical RA and Dec values.
Line 37 in ota_sourcefind.py had been changed to the following to help with and issue with the 5x6 data.
w.wcs.ctype = ["RA---TPV", "DEC--TPV"]
It looks like this does not work for podi. All of the stars were returned with identical Ra and Dec values. I will try adding an if statement to only set the keywords to the TPV values if it is odi 5x6 data.
Currently we are using the QuickReduce header keyword to add the background to the final images, even though we are calculating the backgrounds for each OTA. We should use a more statistical approach to determine the final background level.