In both cases the pursuted objective is to remove defects and others dominances visible in images, parasitics effects induced by the photographic system and to increase all featuress that can improve the overall quality of the document.

Pre-processing or calibration

The first step of our digital processing is the pre-processing or calibration, an essential step for photometric applications as well as for all pictures taken in conditions of low light : planetaries and deep sky objects. 

This step is not mandatory if the subject brightness and contrast allow you to take snapshots, if the quality of images does not suffer of electronic noise or if the subject move quickly. Usually only the high resolution photography of the Moon satisfies these criteria.

Dark frames of a CCD detector respectively cooled at -10, -20, -30 and -40°C. Thermal noise is reduced by half for every 5°C decrease in temperature. These frames demonstrate this effect dramatically.

The calibration requires to take two frames in addition to the image frame of the subject in order to substract all errors imputable to the electronic and in a lesser extent to the driving system. These images consist of : 

- a dark frame recorded in complete darkness during the same duration and as close in time to the image frame as possible. Because, even maintained in a cold atmosphere the chip has a thermal response that generates parasitics photons. This dark frame records also both bias noise caused by the systematic noise in the camera electronics, and thermal noise, described previously. 

- a flat-field frame (FFF) got in photographying a uniform light source using the same optical system (same scope, camera orientation, focus, filter, projection) as was used for the image frame. This FFF being difficult to record, many amateurs content with picturing a twilight sky or, better, an uniformly lit diffusing screen or even a white canvas uniformly illuminated tied in front of the scope. This frame allows to record the vignetting, shadows carry out by dusts on surfaces near the detector (filter, etc) and others variables like differences of pixels sensitivity to light that modify the quantum efficiency of the CCD detector.

Once substracted (divided) from raws images frames, the result is what we call calibrated frames corrected for all irregularities and noises recorded in images. This step can however affects the pictorial quality of the frames because the calibration frames contains random noise of their own. So for photometric applications or if you search for the highest-quality frames, the dark and flat-field frames used for calibrations can be the means of several dark and FFF too. Once averaged, these frames will be substracted from raw images to get better calibrated frames. This process should not be skipped because it improves dramatically your images quality. 

To buy : Anti-Blooming Filter Software, by Kazuyuki Tanaka

M45 Chris Vaughn, AA6G B33, the Horsehead Okano Kunihiko M57 Michel Peyro

Post-processing or image processing

Once you took possession of your precious calibrated frame(s) you can go to the second step consisting in the image processing strickly speaking. This consists to take profit of digital functions which actions are identical to the ones we use in a darkroom,  like the unsharp masking or the gamma correction to enhance high spatial frequencies to improve features in both dim and bright areas of the picture. Once enhanced the brightest objects could be process with a Lucy-Richardson or VanCittert algorithm, while the Maximum Entropy and Convolution functions will be very useful to enhance features on dim objects that display a low signal-to-noise level. At last, Wiener algorithm gives very good results on all DSO in increasing the image definition. Another trick is restoring the image to improve the sharpness of the image frame.

At last, but this is mainly used in planetary imaging, you can extract the object from the background from all images. Now you can recenter the subject, accurately register references points (well identified points) so their data can be combined pixel for pixel to create either a composite image (juxtaposition of high resolution images too big to hold on a single frame) or to create a new image, more detailed, resulting of the stacking of all individual frames (compositage). You can also animate your individual frames or make an astrometry reduction.

If in deep sky astrophotography we usually content with stacking a few individual frames (between 3 and a maximum of about 100 RGB images exposed for a long time for the bravierst), in high resolution planetary photography, good results require often to stack a great number of frames, sometimes exceeding 1000 individual frames as we will explain in the next page about video techniques. We will also discuss on the subject in others chapters, mainly devoted to the signal-to-noise ratio and the picturing of Mars during oppositions (this last in French).

RGB versus LRGB When amateurs speak of LRGB image, theoretically they speak of a processing more sophisticated than simply adding four monochromes frames, 1B/W + 1R + 1G + 1B. Theoretically, to increase the signal-to-noise ratio, reduce the turbulence and others artifacts, RGB images should be the combination of several dozen monochromes images. The number is not very important, and in some cases even one RGB is enough. But usually most amateurs prefer to stack several RGB together to reduce the effects of the turbulence in using only one frame (1R+1G+1B). Then this RGB image is combinated with the luminance image. This last gives the contrast to the RGB composite, whithout wich the resulting image looks fine of course but lack of depth; it is not "crisper of details". LRGB image of Mars taken on August 23, 2003 by Jacques-André Regnier with a Celestron Nexstar 5 (127mm f/58) equipped with a  Powermate 5x and Philips Vesta Pro webcam. This image results of the combination of 800 RGB and 800 B/W individual frames. The luminance frame should be the combination of a few dozen to hundreds individual B/W frames. This is particulary important when picturing highly featured surfaces like planets (Mars, Jupiter, Saturn and in a lesser extent the crescent of Venus and the one of Mercury). The final image result then in the combination of all these individual frames.

The software provided with one's CCD camera should allow you to create the calibrated image from the raw image with the dark and flat-field frames. The image processing requesting a wide maneuver latitude and quite a lot of experience, I suggest you to get first a free or shareware software like Astrostack or IRIS for example and once you are used to work with it, go with a more performing tool like Picture Windows Pro, Photoshop, MaxImDL or even MIRA. As I explain in pages reviewing image processing software, these products are expensives but they have the merit to be powerful, relatively easy to use, they are compatibles with many images formats and are very complete.

Closing with a good advice. If you do not master IP techniques, a simple way to enter into the subject without having to read some books sometime austere and theoretical, is to ask advice to an advanced amateur - links are numerous on this website - who will give you steps to follow to process an image in a few keystrockes. From there, from meetings with friends or remotely, you 'd have learnt what are hidden behind sub-menus functions and settings of some filtering. Once this basis won by practice you could tackle to more technical matters. Now, it's your turn !

Next chapter

Page 1 - 2 - 3 - 4 -