We have compiled a list of "best practices" for CCD and DSLR imaging with MaxIm DL.
If you are an experienced astroimager, but new to MaxIm DL, this is a good place to start. If you're new to astroimaging, we strongly recommend that you first review The Basics, and read the following sections, then circle back to this page.
For best results, follow the steps below in sequence.
Step 1: Finding Your Target
Use the Observatory Control window to locate your target. The planetarium features are streamlined to include only "in the field" operations, but the catalogs are fairly extensive.
If you GOTO an object in the catalog or in the planetarium display, your image can be automatically tagged by the Autosave Sequence feature. It will automatically name your files, add the object name to the FITS header, and will help the Stack command identify images of different objects.
Use the Auto Center feature of the Telescope Tab to compose your target.
The Zoom Tab is also very useful for composing your shots, especially if you are using a focal plane rotator.
You can overlay images from your camera on the Zoom tab, using the PinPoint Astrometry feature to determine the exact location of your image.
The All Sky Tab is useful for situational awareness; it will help you select targets that are well above the horizon for the entire duration of the image capture session.
The Status Tab is also helpful in determining twilight times, telescope altitude, moonrise, transit time, etc.
Step 2: Imaging
Your first task is to create the highest quality data possible. To do that, you want to collect as much light as possible while maintaining good quality. You will also need to collect calibration frames to clean up your raw data.
Operate your CCD camera at the lowest temperature possible, but do not run the cooler at 100%. Adjust the temperature setpoint to a temperature that will run the cooler no higher than 90% or so throughout your imaging session. The cooler temperature is set on the Setup Tab.
While operating the camera at the same sensor temperature as you use during imaging, collect Dark Frames as a minimum. Collect a large number, at least 10-20 frames with the same exposure time as your light frames, so that you can average out the noise. Calibration frames can be collected by setting the Frame Type selector on the Expose Tab. Load your calibration frames for processing into the Process menu Set Calibration command.
Collect as much light as you can. This will allow you to see fainter targets. Usually the best results are achieved by taking 5-10 minute individual exposures, and taking as many of them as you can.
If you cannot take long unguided exposures, then use an autoguider. If you do not have an autoguider available, then you may need to take 30 second to 60 second exposures to avoid trailing; however, you will not get as good results as for longer exposures. Also much more processing time will be required to stack the images.
When acquiring your images, use the Dither function (see Autosave Sequence) so that each captured subexposure is slightly offset from each other. That allows you to eliminate the hot pixels during stacking by using either SD Mask or Sigma Clip function (see the Stack command Combine Tab). This can make a huge difference to your image quality.
Step 3: Calibration
Calibration should always be done first, before any other processing.
For explanation of calibration functions please review the Image Calibration section.
The calibration functions can be set up using the Process menu Set Calibration command, or the simplified Calibration Wizard command.
Ideally you should acquire both dark and flat-field frames. You will also need either flat-darks or bias frames. Please review the Image Calibration section for more information.
Flat-field frames do require some effort and preparation. When you start out you may wish to skip this step; however as you go to fainter targets flat-fielding will become more important.
Once calibration is set up, MaxIm DL remembers the settings including your calibration frames. You can often reuse calibration frames for some time; however it is a good idea to refresh them monthly. Flat-fields may require more frequent updating, especially if you get dust spots on your optics.
You do not have to perform calibration on each image as a separate step; the Stack command can do it "on the fly". This saves time and effort.
The Set Calibration can also be configured to apply a Bad Pixel Map to your images. This is useful if your camera has bad pixels or columns that are still visible after dark frame calibration.
Step 4: Color Conversion (one-shot cameras only)
If you are using a "one-shot color" camera, such as a DSLR, then you need to use the Color menu Convert Color command to translate your raw Bayer-encoded images to color. We strongly recommend using RAW mode for DSLRs instead of JPEG mode, as the quality is much higher.
Convert your images from RAW to color after calibration, but before other processing.
The Convert Color command needs to be set up for your specific model of camera. Presets are available for many models, but since available camera models change frequently your camera may not be represented. You may need to use "Generic RGB", or a preset for another camera model with the same sensor.
Often Convert Color initially requires a bit of trial-and-error to set the X and Y Offset. When the offsets are set incorrectly you will get bizarre colors. It's usually helpful to take an image of something very recognizable the first time so you can set these correctly. Once they're set they will always work for your camera.
The first time you use Convert Color click the Reset Scaling button, and check Auto Background Level. If your colors aren't perfect use the Color menu Color Balance command to adjust it. Tip: Once you've got your color balance figured out, load another raw frame and go back to the Convert Color command. Click the CB button to copy the color balance settings over. Now whenever you use Convert Color your preferred color balance will be automatically applied.
Step 4: Color Filters (monochrome camera and color filters only)
If you are using color filters and a monochrome camera, the initial processing is exactly the same for every image. The color planes are put together during or after stacking.
The Stack command will put together the color images for you automatically.
If you have narrowband frames, the Stack command will output separate images for each; however they will be aligned with the other color planes. the Color menu Blend Narrowband Images command helps you put them together.
Some users like to perform separate processing on the Luminance frame when doing LRGB imaging; that is a matter of personal preference. All of the image processing commands in MaxIm DL work on color or monochrome images.
Step 5: Stacking
MaxIm DL's powerful stacking command, Process menu Stack, can do a lot of work automatically, including:
On-the-fly calibration and color conversion, so you don't need to use the Batch Process command and consume lots of disk space in the process
Automatic rejection of poor quality images, user adjustable and with manual override if desired.
Automatic image alignment, or manual if desired. Usually Auto Star Matching or Astrometric alignment are the most convenient and accurate for deep-sky imaging.
Automatic color filter stacking for RGB and LRGB imaging. You can ask the command to keep separate frames for each color band, as well as combine them all into one color image. Narrowband frames are also stacked and aligned, and can be conveniently merged into the basic RGB stack using the Blend Narrowband Images command.
Your high-quality stacked images represent a lot of processing, so this would be a very good time to save everything to disk. By default the FITS files generated by the Stack command are stored in IEEE Floating Point format, and it is highly recommended that you use this format. This is necessary due to the large increase in signal-to-noise ratio when you stack frames. You are likely to have images that go well beyond what the usual 16-bit integer data can represent.
Step 6: Color Balance
A slight tweak to the color balance may be necessary. Use the Color Balance command, which has two handy mouse-based color adjustment features. The first one allows you to click on a typical background area, to automatically set it to grey. The second one allows you to click an area or star that should be white; it will automatically adjust the overall color balance to correct any color cast.
Step 7: Remove Pedestal
You have a "pedestal" if the lowest pixel in your image is greater than zero. Some amount of pedestal is necessary, to avoid the darker parts of your image being clipped to zero. Also shades of grey very close to zero are not represented well by video monitors.
All that said, if there is a large pedestal on your image, you might want to cut this down to a smaller level; it makes the subsequent steps easier. The Remove Pedestal command is useful for this. Do not reduce the pedestal down to zero; you need some headroom at the bottom.
Step 8: Background Color and Gradients
If you are imaging in suburban conditions, you will often find slight color gradients in your image. Even in very dark locations you will see some color background due to natural sky glow.
Boost the Screen Stretch to emphasize faint background variations. The foreground objects will be saturated, but this doesn't actually affect the data. This step lets you see the flaws in the background more clearly.
Use the Filter menu Flatten Background command. There is an automatic version that may work well for images that have a lot of background area, but the more interactive command often gives better results.
Click on background areas all over the image, and then use the Preview function to see the effect. You can use Simple mode to remove gradients, or the Complex mode to remove vignetting and other effects.
Once the background is uniform, you may still have an overall color cast, especially in the presence of light pollution. First try to neutralize it with the Color Balance command.
You may have some color variation in the background, especially if you were imaging with multiple filters in the presence of changing light pollution. (The telescope motion during tracking can produce this effect.) Use the Filter menu Remove Background Color command to remove any residual color in the background, without affecting the foreground image.
Readjust the Screen Stretch for a pleasing appearance.
Step 9: Compressing the Dynamic Range
Your imaging camera has collected high bit-depth data from astronomical objects that have huge differences between bright and dark features. Unfortunately neither your video monitor nor your printer can adequately represent this. You need to compress the dynamic range, but you want to do it carefully. Some parts of the brightness range have little interesting detail; we want to de-emphasize those regions while emphasizing interesting structure in other regions - without making the image look artificial. This requires some artistry.
Before you start, set Screen Stretch to Max Val. The image will all but disappear. That is okay because we're about to compress the dynamic range, and we want to see the results properly.
A simple way to quickly compress the dynamic range for deep-sky objects is to use the Digital Development filter. In many cases this produces instant beautiful results. However you will have more control if you use the following procedure instead:
Use the Curves command to gently boost the faint areas while reducing the contrast of the bright areas (a mild convex curve). At first not much may happen; this is fine. You want to do this gradually so you'll have fine control.
Repeat the Curves command several times. Watch the histogram in the Screen Stretch window while you are doing this; if the black level of the histogram starts to creep way up, washing out the image, then pull it back using the Remove Pedestal command.
When you get close to your optimal image, stop and tweak the Curve for best results; or use the Levels command for a final tweak.
After compressing the dynamic range of the image, you want to make sure you have a background level that is low but not zero. Use the Remove Pedestal command to set the background to about 10% of the maximum level. For an 8-bit image this corresponds to a background level of about 25 ADUs.
After compressing the dynamic range, colors will often look washed out. Use the Adjust Saturation command to bring them back up a bit. Use a light touch! Compare with the original image to make sure you're not overdoing it.
Step 10: Sharpening
The brighter areas of your image will have very high signal-to-noise ratio, and will likely tolerate a certain amount of sharpening. We can do this selectively as follows:
Open the Filter menu Unsharp Mask command and click the >> button to enable Pixel Range Restriction.
Turn on View menu Information Window and browse around the image with the cursor. Identify the brightness level where the noise level becomes low enough to tolerate some sharpening.
Set the Pixel Range Restriction between this brightness level and 9999999 (greater than the brightest pixel). Set Reduce Radius to 1 and Feather Distance to 2 as a starting point.
Select Kernel - Low Pass or Kernel Low-Pass More.
Mask Weight should typically be set between 75 and 90.
Turn on Full Screen preview and you will see what effect this has. Tweak the values to get the level of enhancement you want.
You will need to pay particular attention to the Mask Weight and the lower Range value. As always, a light touch is recommended; avoid overdoing the enhancement, and watch out for excessive noise near the lower end of the range.
Step 11: Smoothing
The background areas of the image will be the noisiest. You may find smoothing these areas a little will improve the appearance of the image.
Use the Information Window to find the approximate background level in the image.
Open the Filter menu Kernel Filters command and set it to Gaussian Blur mode.
Use the >> button to turn on Pixel Range Restriction.
Set the Range Restriction between 0 and the background level you found above.
Adjust the Gaussian Blur Radius and tweak the Range Restriction to taste.
A small amount of blurring of the background will help suppress noise and make the image look cleaner. Again use a light touch. You want a smooth and low-noise background, but don't overdo it or the image will look artificial.
Sometimes if your image is too smooth and noise-free, it may not display well on computer monitors due to Mach Bands. These are an optical illusion caused by the human visual system; if you see a very small step in intensity, it appears greatly exaggerated. You can break up Mach Bands by deliberately adding a very small amount of noise to the image. Use the Add Noise command to add a very small amount of Gaussian noise to the image. Often just a couple of ADU will do the job.
Developing Your Workflow
The most important suggestion is to experiment! There are many more functions in MaxIm DL than have been mentioned above, and they can be applied with different settings and in different order. The above is only a starting point; trial-and-error works best.
Over time you will develop your own preferred workflow, but remember to be flexible. You will discover that certain types of images respond better to some techniques than others. You will also find differences when you use different imaging equipment; for example, wide-field images usually require different processing from high resolution, narrow field images.