Reporting of peak values for color and extracted FITS images

[ghawkins]

I'm looking at a raw FIT image from a color camera, and then looking at the P12 extraction for the same image (done with ASTAP).  When I zoom into the raw image sufficiently and mouse over the pixels, am I looking at the RGGB pattern?  If so, I would assume the peak value for the P12 pixel in the raw image should be the same as the peak value in the extracted image?  

[karenacollins]

If you are looking at the raw image before debayering in AIJ, then, yes, you would be looking at the bayer pattern. I don't understand what is meant by the P12 pixel though.

Karen

[ghawkins]

ASTAP identifies the RGGB pixels as P11, P12, P21, and P22 respectively.

So I'm trying to match the first G pixel in the initial RAW image to its extraction (P12) in the image split into color channels by ASTAP (that's one red, two green, and one blue images).

Best regards,

Gary

[karenacollins]

The mapping can be complicated depending on the image orientation and fits header keywords. However, you should be able to mouse around in a set of 4 pixels and match them to the same set of 4 pixels in ASTAP (after figuring out image rotation differences). Does the latest AIJ v5 daily build's debayering separation into RGB stacks not match those of ASTAP? Be sure and use the debayering option that is specified as being for photometry.

Karen

[ghawkins]

Hi Karen,

I'm looking into whether the AIJ and ASTAP images are the same after color speration.  Does the image split function work for a stack of images, or is it just applicable to single images?

Best regards,

Gary

[karenacollins]

Image or Stack.

[ghawkins]

I'm assuming the green extraction consists of the average of the two green pixels? In which case, if green is the stronger of the color channels, the reported peak values will not agree since in the RAW image it will be the greater of P12 or P21, and in the extracted image the green peak will be (P12 + P21) / 2.

[karenacollins]

See this post:
http://astroimagej.170.s1.nabble.com/Extracting-individual-color-pixels-from-DLSR-or-other-Bayer-pattern-color-images-to-run-photometry-ir-td1857.html