Can some of the more experienced CCD imagers let me know what their general image work flow looks like? Also how they save their files etc.
A few things have become obvious when handling lots of exposures using CCDs. One is the number of files and the number of new files created to handle the aligned files etc etc. The need to keep a log of what you took, how long, what temp, what colour, what binning was used.
It seems it goes something like this:
0. Create a library of dakrs for both 1 x 1 binning and 2 x 2 binning at the temperatures you will be imaging at. Take 3-5 darks and median combine the results to create a master dark and label it carefully and all darks under a master file of "Darks 1 x 1 binning" or "Darks 2x 2 binning".
Take a flat by exposing the telescope at dawn or dusk skies away from the sun at a uniformly lit part of the sky for long enough for the average pixel value to be arund 30,000. Use the clear filter.
1.Take exposures and make sure you have autosave on and the file you want the images to be saved carefully labelled with object name, colour of the image, exposure time, binning, temperature. Right here there is a problem as CCDSoft only records the colour and no other data if you do a colour run. Does add the temp, exposure length or binning (or can it be made to do that?). So I have started to label the files from a colour run like this: M20,L30 1x1 R15 G7.5 B15 2x2 -20C. each file gets that label and CCDSOft will put the colour at the end of each file as well.
2.Align the entire folder of images.
Again here is another problem. The common practice of luminance at 1 x1 binning and colour at 2x2 binning makes it so CCDSoft sometimes will align 1 x 1 and 2x2 and other times it will not (I think if they have been processed in any way it does not seem to want to do it). Now that means I have to resize the 1 x 1 images which seems to be another unnecessary step.
3. Dark subtract each image and then process them for background and brighten/contrast using CCDSoft. If you took a flat then divide the resulting image by the flat at this point.
Then save to each LRGB separate file for later combining of multiple individual colour shots. I also have been doing repair for cold and hot pixels, Hard to notice the difference but perhaps that shows up later.
4. Combine the several sub-exposures for each individual LRGB or Halpha and save.
5. Open these resulting combined images for LRGB and do a colour combine using CCDSoft and play with the colour balance so it looks OK. I have found in my limited expereience so far that CCDSoft does not do such a great job at this. Now it could be my images for RGB are not at the ideal exposure length but I have been advised for an SBig 2000XM it is R 1 G.5 B1
as a ratio and Luminance from seeing lots of images seems to be 2 - 4 times the length of the longest colour image.
Perhaps Photoshop CS2 would be better for the colour combine as it has layer technology. Comments here?
6. Now you transfer the image as a Tiff file to you image processing software for colour balance, noise reduction, curves etc etc.
Does that sounds about right?
The labelling and creation of new files to save processed files at different steps needs to be standardised for me so it reduces confusion and time in processing. Not remembering 100% what the exposure lengths were after the fact means you could subtract the wrong dark.
Also is there any freeware for an imaging log to record this for referral to afterwards when processing the images. I see sometimes people reprocess images rom several years earlier so that becomes an even more vital step to do.
Greg.
A few things have become obvious when handling lots of exposures using CCDs. One is the number of files and the number of new files created to handle the aligned files etc etc. The need to keep a log of what you took, how long, what temp, what colour, what binning was used.
It seems it goes something like this:
0. Create a library of dakrs for both 1 x 1 binning and 2 x 2 binning at the temperatures you will be imaging at. Take 3-5 darks and median combine the results to create a master dark and label it carefully and all darks under a master file of "Darks 1 x 1 binning" or "Darks 2x 2 binning".
Take a flat by exposing the telescope at dawn or dusk skies away from the sun at a uniformly lit part of the sky for long enough for the average pixel value to be arund 30,000. Use the clear filter.
1.Take exposures and make sure you have autosave on and the file you want the images to be saved carefully labelled with object name, colour of the image, exposure time, binning, temperature. Right here there is a problem as CCDSoft only records the colour and no other data if you do a colour run. Does add the temp, exposure length or binning (or can it be made to do that?). So I have started to label the files from a colour run like this: M20,L30 1x1 R15 G7.5 B15 2x2 -20C. each file gets that label and CCDSOft will put the colour at the end of each file as well.
2.Align the entire folder of images.
Again here is another problem. The common practice of luminance at 1 x1 binning and colour at 2x2 binning makes it so CCDSoft sometimes will align 1 x 1 and 2x2 and other times it will not (I think if they have been processed in any way it does not seem to want to do it). Now that means I have to resize the 1 x 1 images which seems to be another unnecessary step.
3. Dark subtract each image and then process them for background and brighten/contrast using CCDSoft. If you took a flat then divide the resulting image by the flat at this point.
Then save to each LRGB separate file for later combining of multiple individual colour shots. I also have been doing repair for cold and hot pixels, Hard to notice the difference but perhaps that shows up later.
4. Combine the several sub-exposures for each individual LRGB or Halpha and save.
5. Open these resulting combined images for LRGB and do a colour combine using CCDSoft and play with the colour balance so it looks OK. I have found in my limited expereience so far that CCDSoft does not do such a great job at this. Now it could be my images for RGB are not at the ideal exposure length but I have been advised for an SBig 2000XM it is R 1 G.5 B1
as a ratio and Luminance from seeing lots of images seems to be 2 - 4 times the length of the longest colour image.
Perhaps Photoshop CS2 would be better for the colour combine as it has layer technology. Comments here?
6. Now you transfer the image as a Tiff file to you image processing software for colour balance, noise reduction, curves etc etc.
Does that sounds about right?
The labelling and creation of new files to save processed files at different steps needs to be standardised for me so it reduces confusion and time in processing. Not remembering 100% what the exposure lengths were after the fact means you could subtract the wrong dark.
Also is there any freeware for an imaging log to record this for referral to afterwards when processing the images. I see sometimes people reprocess images rom several years earlier so that becomes an even more vital step to do.
Greg.
