ImageIntegration issue - Green resulting image

[Luis Abraham]

Hi everyone! I'm new in the forum and also using PI .

Before using PI, I always used DeepSkyStacker to stack my pictures but I don't know what is wrong this time because I'm getting some star trails in the resulting image , so I decided to try PI's ImageIntegration and I realized the stacking was very good , I got pinpoint stars in all the image but the color was not good at all .
For any strange reason, I'm getting a green image as result:


I've been reading some tutorials and different threads in the forum but I don't really know what I'm doing wrong .

I'm working with .CR2 raw files from my Canon 450D, no darks, no flats and no bias.

I already set the recommended DSLR_RAW format settings in order to PI read my .CR2 files exactly as they have been created by the camera, but the result was the same .
I tried with different ImageIntegration parameter but at the end I always got a green image.
Also, I've just followed the Harry's video tutorials but I got the same green resulting image .

This is what I'm doing:
1. Got the .CR2 files from my camera.
2. Go to PI and use StarAlignment tool.
3. Finally I try to stack the resulting .fit files with ImageIntegration.


Please let me know if I'm missing a step or if I'm doing something wrong here.

Thanks!

[Enzo De Bernardini]

Hi Luis,

I also I have a DSLR and get a picture that at seems green. Don't worry, the color will be normal after some calibration procedures: BackgroundNeutralization, DBE and ColorCalibration (especially BN and CC)

Let us know how was it, and welcome to PI perfect-alignment 

Best regards,

Enzo.

[Cleon_Wells]

Luis, you need to use the BatchDebayer script on the _c.fit files after you Calibrate the your raw files or if  you don't Cal your files, then after you BatchFormatConvert your CR2 files to .fit files run the BatchDebayer script. Then you Register and Integrate the subs.
Cleon

[pfile]

right, the flow should be: calibrate -> debayer (batchDebayer script) -> register -> integrate.

if you uncouple the channels in the ScreenTransferFunction tool (press the 3 colored cubes above the "1" - it will become deselected) and then press "A", STF will compute an auto stretch that balances the channel intensities. that will get rid of the green/blue cast. if you want you can copy the STF to the HistogramTransformation by dragging the STF arrow down to where the HT processes's arrow is, and then drag the HT's triangle to the image. then you have to turn off the STF on the image (control-S). don't forget to reset the HistogramTransformation process before trying to use it again, as the R,G,B channels will still be loaded with the STF data.

[Cleon_Wells]

Luis, I took an image of my backyard and used this image to help me understand what the different conversion settings in the Format Explorer, DSLR_RAW ,  Raw format preference window does. Try opening a raw CR2 file with different Raw preference settings and use the STF  to see the results of these changes. Create a .fit file and BatchDebayer it to see the results. You could use the Histogram Transformation process directly , skipping STF.  In the HT window select an open file by clicking the Real Time preview icon (the eyeball) and learn how to use the adjustment settings in this window.
And for some light reading___NOT ---  LOL, check this info out on DSLR files.

http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/index.html

Cleon

[Cleon_Wells]

correction.

Luis,  In the HT window select an open file (( with the pull down View Selection function and then use the Real Time Preview function  to see the changes you make with the HT settings)) by clicking the Real Time Preview icon (the eyeball) and learn how to use the HT adjustment/settings in this window.

Cleon

[Luis Abraham]

Thank you very much for you inputs and advises!

I'm going to play with all those configs and tools during this weekend.

About raw calibration... I didn't calibrated my .CR2 files before Register and Integrate procedures because I think I need darks, flats and bias, am I right? is there any way to calibrate them without having darks, flats and bias?

Have a great weekend guys!

[Cleon_Wells]

Luis, You don't have to Cal the subs to format convert them,  use the BatchFormatConversion utility on  the CR2 files/raw bayer data, to convert the raw file to a .fit format and then Debayer the .fit files before you register and integrate them. Remember to check the settings in the DSLR_RAW Format preference window by double clicking on DSLR_RAW in the Format Explorer tab before you convert the raw bayer data.  The only time you will work with the raw bayer subs is during Calibration.  If you work, (reg and integrate) with the raw bayer data with out conversion you will get strange unusable results, I think this is one of the reasons some give up on PI. The bottom line is there is no EASY BUTTON in PI, you have to learn what the basic settings do and how they should be set to get a good registered  and integrated master. Check out the Tutorials and Processing Examples section of the forum for calibrating  your DSLR_RAW subs after you get a good integrated master.
Cleon

[pfile]

Thank you very much for you inputs and advises!

I'm going to play with all those configs and tools during this weekend.

About raw calibration... I didn't calibrated my .CR2 files before Register and Integrate procedures because I think I need darks, flats and bias, am I right? is there any way to calibrate them without having darks, flats and bias?

Have a great weekend guys!

yes, at the very least you need darks in order to calibrate your lights. if you want to let pixinsight scale the master dark (the "optimize" checkbox in the master dark section), then you need a master bias frame as well. and if you include flats you will need master bias and dark frames.

if you are not going to calibrate the CR2 files, you could set DSLR_RAW to debayer them directly and after doing BatchFormatConversion to .fit format you could then register those .fit files and stack them. i did post in another thread regarding the difference in histograms between doing this and setting DSLR_RAW to BayerRGB mode and then doing BatchDeBayer. the resultant histogram of the two .fits files are different (for my flats, at least), and at this point i dont know which one is more "correct".