POST PROCESSING

This is the best bit in some ways as your image will start to come alive here. There are numerous processes that you can carry out to bring the best out of your captured data but there are some ‘standard’ processes that apply to most images and although we all have our own preferred methods, the following is what I do followed by a worked example of the most basic operations.

Having stacked the individual images in either Registax or MaximDL (complete with a 'dark frame subtraction' if appropriate) and saved the resultant image as a 16 bit TIFF file (.TIF), you can import the image into PhotoShop (but most graphics manipulation software has similar functionality). Never save your work in progress as a .JPG (JPEG) file as this file format has 'lossy' compression which means that every time you open and then save the file after working on it you will add to the losses in data. Always work on the file in a non-lossy format like TIFF when still in 16 bit mode or .BMP if you have had to drop down to 8 bit mode. By all means, save a copy in .JPG format for email or internet use when you have finished processing but not before!

Background brightness

In the UK it is rare (but not impossible!) to find a dark sky site so many images will display a red/orange/brown tint to the background sky caused by light pollution. I like my images to show a reasonably dark background sky so this is my start point.

Using the ’levels’ command, you can set the black level using the 'black pipette' to get rid of much of the light pollution. To do this, choose a small area of the background where there are no stars or nebulosity. The image will immediately look more vibrant as the main object will now be contrasted against a dark background. (see worked example). In fairness, this is a rather 'down and dirty' method and I explain a much better method in my book 'Making Every Photon Count', however as quick way of revealing what you have captured is can be quite useful.

Brightening the Image

Still using 'levels', the image can be ‘stretched’ (enhanced in brightness) by using the middle slider with the RGB Channel selected until the light pollution just re-appears and then backing off the slider a small amount. (see worked example)

This will enhance the object enormously but may result in a colour imbalance with perhaps too much or too little of one colour becoming apparent. Selecting the required channel (Red, Green or Blue) and, again using the centre slider to reduce or increase the level of the chosen colour will go a long way towards redressing the balance.

Tonal Range

The next procedure is to select the ‘curves’ function to adjust the general balance and tonal range to your taste. This is a very powerful tool and needs to be used with care as it is very easy to over-cook the adjustments. If your software does not have a ‘curves’ function, do not despair as there are other means of achieving the same result using just levels and saturation functions.

Colour Balance

Sometimes it is necessary to make fine adjustments to the colour balance and this function will allow these small adjustments to be carried out by increasing the intensity of one colour at the expense of another, the 'pairings' being as follows:-

Cyan Red

Magenta Green

Yellow Blue

Increasing the Red intensity will be at the expense of the Cyan and so on for the other pairs.

Saturation

Once the brightness and colour balance have been adjusted to your taste, it can be beneficial to increase the colour intensity. The function used to adjust colour intensity is called 'saturation' and again, using the available sliders in the saturation command dialogue box, the intensity of individual colours can be increased or decreased

Finishing touches

Smoothing

Sometimes, at this stage, I import the image into a programme called 'Neat Image' to get rid of any 'grain' in the image caused by sensor noise and/or sky conditions. This program will smooth out background sections of the image without destroying the detail of the main object. I then blend the Neat Image version with the original until I get just the right amount of 'smoothing'. To do this, copy the smoothed image and then paste it on top of the original image giving you a new layer. Now adjust the 'opacity' until there is just a small amount of the 'grain' of the original showing through the new smooth layer. Finally adjust the opacity either up or down until you get the exact amount of smoothing you require. If the exposure values and sky transparency are good then this procedure is not normally required.

To a more limited extent, a similar adjustment can be made by using the ‘Gaussian Blur’ function available in most graphics manipulation programmes.

Sharpening

As the name implies, this is the opposite of smoothing and should be used with caution as its use can increase the intensity of sensor noise and make stars look most unnatural!

Cropping

Although the image should have been well framed early on in the imaging process, the final image may well include peripheral objects that spoil the overall appearance of the image. Careful cropping of the image can remove these and ensure that the main object is centred nicely in the frame. Be sure to apply the same percentage crop to both the height and width to maintain the correct aspect ratio. (see worked example)

Re-size and file format

Depending on the final use of the image, you may wish to re-size it accordingly - for example, I always re-size my images to 600 pixels width and convert them to low compression JPEGs for inclusion on my website BUT, before you do this, keep a full size version as a 16 bit TIFF file or 8 bit.BMP file as a minimum on your computer in case you want to re-visit the image for more processing in the future - this is what rainy evenings were designed for ......

Click here for a worked example

Copyright Steve Richards 2008