Checkerboard images

[Yourdog]

Hello, I am new to the forum and I've recently started taking photos of the sky, this is my first telescope and have been enjoying it a lot although I need a lot more practice and time hah. I have been searching through multiple forums trying to figure out what is causing a checkerboard pattern on my images. Recently I took a lot of photos of the Moon and when zooming in to the photos there is a checkerboard pattern with the entire photo. Users have recommended using PIPP before processing the images through AutoStakkert and Registax. I have attempted doing so, but with no luck (I still don't really know what I'm doing) and am looking for assistance. I have had successful shots of the Moon, Jupiter, and Saturn in the past and I don't know what has changed or what my next step to take is.

1. Camera: ZWO ASI120MC-S
2. AWB OneSky
3. FireCapture v2.6

These are the results from FireCapture settings:

FireCapture v2.6  Settings
------------------------------------
Camera=ZWO ASI120MC-S
Filter=L
Profile=Moon
Filename=Moon_231153.zip
Date=170619
Start=231153.926
Mid=231227.365
End=231300.804
Start(UT)=041153.926
Mid(UT)=041227.365
End(UT)=041300.804
Duration=66.878s
Date_format=ddMMyy
Time_format=HHmmss
LT=UT -6h
Frames captured=843
File type=JPG
Binning=no
ROI=1280x960
ROI(Offset)=0x0
FPS (avg.)=12
Shutter=1.000ms
Gain=7 (7%)
WRed=50 (off)
USBTraffic=100 (off)
FPS=100 (off)
SoftwareGain=10 (off)
Brightness=2 (off)
AutoGain=off
AutoHisto=75 (off)
Overclock=0 (off)
Gamma=50 (off)
AutoExposure=off
WBlue=50 (off)
HighSpeed=off
Histogramm(min)=0
Histogramm(max)=59
Histogramm=23%
Noise(avg.deviation)=n/a
Limit=none
Sensor temperature=31.2°C
 

Also attached is an image with the issue, and bonus of Jupiter and Saturn to show I've had decent results.

I do not have any light or dark frames, I just took pictures. Please help, and thanks in advance.

[knobby]

Nice images, looks it's somehow picked the wrong debayer pattern.

Should be GRBG for the 120 mc s

 

Edit - you also seem to be capturing in jpeg from the firecapture log, try an avi and then use autostakkert to debayer / align frames

Edited June 22, 2019 by knobby

[vlaiv]

Hi, welcome to SGL.

Image that you posted is a single frame of the target, right?

Checker board pattern is quite normal for OSC (one shot color) - cameras. These cameras don't provide color "out of the box" - they in fact have something called Bayer matrix. This means that there is 2x2 grid of pixels each recording different color - red, blue and two green - and it repeats over the whole image.

When you capture raw image from such camera - you end up with "mono data" with such grid - that is what is causing checker board pattern.

Way to get rid of it is to perform something called Debayering (or sometimes demosaic). Software should handle that for you.

[Yourdog]

2 hours ago, knobby said:

Nice images, looks it's somehow picked the wrong debayer pattern.

Should be GRBG for the 120 mc s

 

Edit - you also seem to be capturing in jpeg from the firecapture log, try an avi and then use autostakkert to debayer / align frames

How can you tell? I've also tried going through each one of the debayer options but it just changes the colors, but keeps the squares. I've also used AVI in the past but I seem to be having more success with pictures so I normally just stick with JPEG.. still a newbie so I don't really know what I'm doing wrong.

 

2 hours ago, vlaiv said:

Hi, welcome to SGL.

Image that you posted is a single frame of the target, right?

Checker board pattern is quite normal for OSC (one shot color) - cameras. These cameras don't provide color "out of the box" - they in fact have something called Bayer matrix. This means that there is 2x2 grid of pixels each recording different color - red, blue and two green - and it repeats over the whole image.

When you capture raw image from such camera - you end up with "mono data" with such grid - that is what is causing checker board pattern.

Way to get rid of it is to perform something called Debayering (or sometimes demosaic). Software should handle that for you.

That was just one photo out of over 800 but I'm capturing with JPEG setting with no limit until I feel I have enough photos captured. I am trying to figure out which programs to use and how to use them, I tried loading them in PIPP and AutoStakkert changing the debayer options but it doesn't seem to change anything besides color while keeping the squares.

[vlaiv]

36 minutes ago, Yourdog said:

That was just one photo out of over 800 but I'm capturing with JPEG setting with no limit until I feel I have enough photos captured. I am trying to figure out which programs to use and how to use them, I tried loading them in PIPP and AutoStakkert changing the debayer options but it doesn't seem to change anything besides color while keeping the squares. 

If you captured JPEG (which you should not do in any case) - than every other program just believes that image is color image. Problem is with capture application - it recorded raw output from the camera and converted it into JPEG without doing debayering.

Here are some pointers.

- Don't use jpeg or avi with compression. Most compression algorithms that deal with images and video are created to exploit "natural looking" images and are in fact changing the data (to make it more compressible), while keeping natural look. When working with astronomical images - you don't want your data to be changed at all, so avoid these.

- If you are going to shoot a video (for planetary that is best approach) - use SER format. If you are going to shoot stills - FITS is the best format for astronomy.

- Set your capture application to work with SER format - use 16bit if you can. Don't debayer yet - just record mono data. If your capture application does not offer these options - look up FireCapture or SharpCap applications - these are both very good for planetary recording.

- Shoot a set of darks (also in ser format - same exposure, just cover telescope). Also shoot set of flats / flat darks if you have a flat panel. You can load all these in PIPP to do calibration. Now you can select different workflow - either debayer in PIPP or leave bayer matrix intact and just do calibration and other small processing before you open your movie (again select SER as output) in AS!3.

It is best if you watch some tutorials (I'm almost certain you can find examples of how to work with SharpCap / PIPP/ AS!3 on youtube), or read some blog posts / tutorials on this topic and just try things out.

[Yourdog]

Thank you, I will experiment more. Do you think there is a way to save these JPEGs? I mostly use FireCapture and have tried SharpCap. I've seen SER and FITS in the program but since I didn't really know much I just stick with what I've always seen - JPEG. Just curious, why do these programs have JPEG as an option?

[vlaiv]

You can try following, but not sure what sort of quality are you going to get.

Take any image preview/manipulation program (like irfanview or similar), take one of your jpegs and save as monochrome png (convert to monochromatic and then save as monochromatic image in png format).

Import that into PIPP and do debayer - it should remove pattern (as much as possible) and you should be able to get something resembling a good image (how much depends on how much "damage" jpeg did to it).

You can even import sequence of png images and create a SER file and try stacking it with AS!3.

It might even be possible to import jpegs directly into PIPP - but it will consider those to be color, so not sure if debayer option will be available.

As for JPEG being available - because it is easy to add (there are ready made libraries to save in all common formats - like .bmp, .png, .tiff and jpeg for example) - and might be useful to someone under some circumstances

[Yourdog]

Thanks so much! I will try this out when I get home later and I'll post if I get good results.

[Yourdog]

On 22/06/2019 at 18:30, vlaiv said:

If you captured JPEG (which you should not do in any case) - than every other program just believes that image is color image. Problem is with capture application - it recorded raw output from the camera and converted it into JPEG without doing debayering.

Here are some pointers.

- Don't use jpeg or avi with compression. Most compression algorithms that deal with images and video are created to exploit "natural looking" images and are in fact changing the data (to make it more compressible), while keeping natural look. When working with astronomical images - you don't want your data to be changed at all, so avoid these.

- If you are going to shoot a video (for planetary that is best approach) - use SER format. If you are going to shoot stills - FITS is the best format for astronomy.

- Set your capture application to work with SER format - use 16bit if you can. Don't debayer yet - just record mono data. If your capture application does not offer these options - look up FireCapture or SharpCap applications - these are both very good for planetary recording.

- Shoot a set of darks (also in ser format - same exposure, just cover telescope). Also shoot set of flats / flat darks if you have a flat panel. You can load all these in PIPP to do calibration. Now you can select different workflow - either debayer in PIPP or leave bayer matrix intact and just do calibration and other small processing before you open your movie (again select SER as output) in AS!3.

It is best if you watch some tutorials (I'm almost certain you can find examples of how to work with SharpCap / PIPP/ AS!3 on youtube), or read some blog posts / tutorials on this topic and just try things out.

Ok, I tried again tonight with both SER for video and FITS for stills at 16-bit. I ran both of Jupiter and Saturn and they are coming out clear but I am still getting the squares when I use those formats. I know I am missing something, I just haven't figured it out yet. I didn't get any darks because I was too much in a hurry since it started to rain and I didn't think about it because I was focusing on getting all of my equipment in a safe location. I do not have a flat panel. I ran the captures through PIPP first, both with debayering in PIPP and without, but either way I was still getting the same results. If I don't debayer within PIPP,  the options within AS!3 just don't turn out good.. I will watch videos and read up on everything. Just giving status updates and maybe looking for some more pointers. I played around with irfanview and it was able to remove some of the checkerboard pattern but I am giving up on those pictures and will try again when the moon comes back around.