Backyard Voyager
RegiStax is a very powerful program used for aligning and stacking images- either stills or frames of an AVI. Developed by Cor Berrevoets, the program is available free of chrage for download from here http://registax.astronomy.net/
I have had several people ask me about using RegiStax. So, I decided to put together a little tutorial. The idea here,is to provide a basic understanding of how the program works so you can get the ball rolling and process your images. Of course, the best way to learn is to experiment yourself and see how the controls work.
I hope this helps a little.
The most important thing to remember about using Registax, or any other image processing program, is that it can only work with what it is given. Use the highest quality images or AVI that you can. If the seeing is poor, you are better off just putting the camera away and enjoying the night. You can get some image detail enhancement but for high resolution imaging, you need a steady sky. Also, the more images the better as the Signal to Noise ratio improves with the square root of the number of images.
Finally, focus... take your time. Use a Hartmann mask to help. Critical focus is essential if you want a great final result. Once again, if the seeing is poor, you will have a hard time achieving critical focus. I like to focus on a star. I start by locating a fairly bright star close to the object I want to image. Get the star centered and decrease the image size to 320x240,and crank up the frames per second to 20 or 30 (this is no good for capturing so remember to set it back down when you are finished focusing), next crank up the gain and then increase the shutter speed to dim the star. What all this does is give you a star that isn't saturating your CCD, and an almost live feed so you can see the results of adjusting the focus easily. If you are using a Hartmann mask, bring the star images together until they merge (you may need to decrease the shutter speed to compensate for the reduced light grasp with the mask in place). If not, bring the star to as small a point as you can.
Reset the camera image size and frame rate settings and use auto white balance and auto exposure to start and let the camera make the first approximation for settings. Then, reset to manual exposure and adjust the gain to between 50-80% (if possible) and the sutter speed as fast as you can and still have a good image. Use a frame capture rate of 5-10fps and an image size of 640x480. I have successfully used 320x240 and rescaled the image with RegiStax. This has the advantage of allowing faster frame rates and shutter speeds, which in theory, should help freeze the seeing. Grab about 1-2 minutes worth of video. At 5fps, this would be around 600 frames. I like to shoot around 1000 frames, so I tend to use 10fps. For objects like the Moon, Mars, or Saturn, you can go longer. Jupiter rotates too fast to go much beyond 2 minutes.
Loading
The first step in processing an image is to load it into the program. To do this, click the "Select" tab and navigate to the file you want to load. If you want to load an AVI, click and open it. If you want to work on a series of still images, highlight all the image files at once and click load.

This loads the image files or AVI and activates the "Align" tab. You can accept the defaults for the tabs located on the right side of the screen.
Once the image files or AVI have been loaded, you must select an alignment feature. To do this pick the size of the alignment box from the menu and then left click on the feature in the image. A white box will appear. A high contrast portion of the image works best. The smaller the box, the faster the processing will go. For planetary images, the box should be large enough to cover a good size bit of the planet. Also, you can change the quality estimator too. The most accurate is probably the "gradient" mode. Now, make sure the "automatic" checkbox is selected as well as the "colour" checkbox if you are processing colour images and press the "Align" button. The program will now start to align the images centered on your alignment feature. Don't worry about the "Flat/Dark" tab for now, we'll come back to that later. You can also leave the other settings on their defaults for now.

Once you press the "Align" button, the program takes over and will align, sort the images by quality, and stack them. This can take some time too- especially if you have a slow computer. So just sit back and relax. One thing about RegiStax, the various features of the program will only become functional at certain stages. You basically start with the left tab and move to the right. Now, you can do all this yourself by selecting the "Manual" checkbox in the "alignment" section rather than "automatic", but I like to let the program run through its paces once and see what comes out.
The next window to open is the "Optimize" tab. Here, the program goes thru the images and sorts them based on quality and aligns them. This usually takes several minutes as the program will run several iterations to achieve optimal alignment.
After the "Optimize" tab comes the "Stack" tab. At this stage, the program is stacking the images to form the final composite image. This tab has several features we should note. Primarily, the "Histogram stretch" checkbox. When checked, the final image will be stretched and brightened.
The "Stack Graph" and "Stack List" tabs on the right give you access to the images that will be stacked. You can select individual images from the "List" or set the cutoff point from the "graph". Selecting the best images from the list will give you the best output.
The Optimize and Stack windows will flash by as the program runs. But you will not be able to change any of the settings until it has finished unless you are running in "Manual" mode.
Processing an image
Finally, we get to the most interesting Tab-"Wavelet". Here we can bring out the hidden detail in the image. Once the "Wavelet" window opens you will see a selection of sliders. These are used to sharpen the image. They are arranged on the left side and run from #1 which adjusts the finest details to #6 which adjusts the coarsest.

Slide the #6 slider about half way and see what it does to the image. You will notice that only a small portion of the image changes. This is a "preview" area.There are tabs on the right for Brightness/Contrast, gamma, histogram adjustment, and RGBshift. The RGBShift feature is a great tool. It is used to correct for atmospheric induced red/blue shifts. You can use these to adjust the image or save the image and adjust these values in a program like PhotoShop. Once you have adjusted the image to your liking, click the "Do ALL" button. This applies the changes to the entire image.

The next window is the "Final" tab. This window allows you to adjust the hue, saturation and rotation of the image. Again, all these functions can be changed in PhotoShop.
Now, there are two things you can do here... save the image or go back and reprocess. You may be able to get more detail, or a smoother image if you try this. Go back to the "Optimize" tab and select "Resample". I like to use the "Mitchell" setting and a factor of 2 or 3.

When you do this, the image size is increased. Now, go back to the "Alignment" tab and reselect the alignment feature and click the "align" button again. This will run the whole process over again. Because of the resampling, the processing will be a lot slower this time.
Once finished, you will be able to use the fine adjustment sliders on the "Wavelets" page without increasing the noise.

Again, make the adjustments you like and this time, resize the image to 25% or 50% and save. At this stage, I usually save the large image and use Photoshop to resize. Generally, I like to set the resolution to 300 and the pixel size to 450. I also might run a soft Gaussian blur or "Noise" filter to soften the image a bit. That is a personal choice of course. Below is the final result.

Flats and Darks are images that are taken to correct for inconsistancies in the image train and for the noise from the CCD camera.
A Dark image is made by blocking the incoming light (say by putting the lens cap on) and shooting a few images or AVI under the same conditions as were used for the imaging session. A Flat field image is aquired by taking a picture (or AVI) of a uniformly lit, white field using the same optical set up that the images were aquired under. You want the flat to be exposed at about 50% of pixel saturation for your camera. This can be done by using a light box that fits over the end of your telescope, or by shooting the twilight sky, or by shining a light on a white sheet or T-shirt drapped over the telescope. The flat field is divided into the images to remove defects like dust and vignetting. Dark images are subtracted to remove hot pixels. RegiStax can make a master dark or flat field image for you. Simply open the flat images (several stills or an AVI) and click on the "Flat/Dark" tab. Select "create flat" to make the master flat or "create dark" to make the master dark. Once created, save the images. You load these images here as well. Once loaded, the images will be applied during processing. Just click the checkbox and RegiStax will do the rest. If you apply flats and darks and you find your image comes out dark... play with the gamma and image/contrast sliders on the Wavelet page to see if you can bring up the image.
One final thing about flats and darks. It is important... no, it is imperative, that the optical train be exactly the same for the darks and flats as it was for the imaging session, so that the defects are in the same orientation on the CCD all the time. That means that you can not rotate the camera or remove it before shooting the darks or flats.
Now, using flats and darks is very important when you process still images. However, when you are stacking an AVI, the image drifts around the field a bit and thus, the defects and blemishes that are present tend to get averaged out of the image. So Flats and Darks are not as big a deal when processing AVIs.
The other tabs we haven't touched on yet are the "Project File" tabs. These allow you to save a project and come back to it at a latter time. This is great and allows you to finish processing an image when it is convienent for you.
The other neat thing is on the wavelets tab. This is the "Save Scheme" and "Load Scheme". These allow you to save a set of manipulations and then apply them at the click of a button.
I hope that you found this short tutorial helpful. We have only scratched the surface here, but it should be enough information for you to put together some nice images. The best teacher is experience. As you use the program you will get a better feel for what the individual functions can do. We are all indebted to Cor for providing such a fine program to the astronomical community.
If you have any coments, please feel free to email me.
Cheers and good luck!
Tim