Once your images are calibrated you can begin the basic image processing steps to enhance your pictures.  Basic image processing involves stretching the image for the best display, combining images to improve detail, and using processing filters to enhance the image.

Stretching

There are two types of image stretching.  The first -- display stretching -- simply determines how the image is displayed.  The second, called permanent stretching, actually affects the data within the image.

Display Stretching

Display stretching is important because most CCD cameras are 16-bit systems, which means an image taken by the camera contains over 65,000 levels of brightness.  Even 12-bit cameras contain over 4000 brightness levels.  A computer monitor is an 8-bit system, displaying only 256 levels of brightness.  In order to bring more of the brightness levels from the CCD image into the range of the monitor, you stretch the image.

MaxIm DL has a sophisticated (but easy-to-use) display stretching function.  This is the Histogram function.  Histograms are easy to understand and are vital to image processing in many programs, including MaxIm DL and Photoshop.

A histogram is simply a graph showing the number of pixels in an image with a given value.  Take for example the simplified image below. 

Above:  A simplified astronomical image, containing various levels of brightness.

This picture contains 100 pixels in a 10 x 10 square.  If we assign values to each shade of grey in the image, say, black = 0, dark grey = 1, medium grey = 2, light grey = 3, and white = 4, then we can construct a graph showing how many pixels there are for each value.

Above:  A histogram for the simplified image shown above.  The horizontal axis represents pixels values from dark (0) to light (4).  The vertical axis indicates the number of pixels of a given value.

In our image, 25 pixels are black and have a value of 0.  52 pixels are dark grey and have a value of 1, so we see a spike in the graph at this value indicating that there are more dark grey pixels than any other individual shade.  This is a typical shape for the histogram of an astronomical object.  There are then only 12 medium grey, 8 light grey, and three white pixels.  This is very similar to a real astronomical image in that much of the image is dark but only some areas are true black (usually the result of noise and dead pixels).  Only a small fraction of the image is in the mid-tone range, and very little is pure white.

The biggest differences between this simplified histogram and that of a real CCD image are the number of pixels (typically millions, rather than a hundred) and the levels of brightness (as many as 65,535 instead of the 5 shown above).  Using the histogram you can define which pixels are displayed as black and which are displayed as white.  Using permanent stretching (described in the next section) you can define how the values in between black and white are displayed.

Now for a real example.  Begin with a calibrated image and open the Screen Stretch window (if it is not already) by clicking on the Screen Stretch button on the toolbar (shown on the left below).

Above:  Screen Stretch button on the Toolbar.

MaxIm makes a guess at a histogram setting for the image.

Above:  Default display of the Whirlpool Galaxy image.

Above:  The histogram for the image above.

How to Read the Histogram

The red triangular slider indicates the black level in the image.  All the pixels with values lower than this point (left of the slider) will be displayed as pure black.  Everything to the right of the green slider (all pixels with higher values) will be shown as white.  The remaining 254 shades of grey will be composed of the pixels with values between the two sliders.

The Minimum and Maximum numbers indicate the pixel values of the black and white points.  All pixels with values less than 132 in the above example will be displayed as black.  All pixels with values above 436 are white.

In the above image of the Whirlpool Galaxy, it might be desirable to enhance the detail visible in the cores of the two galaxies.  As displayed above the cores appear overexposed.  By dragging the green slider in the histogram to the right, higher and higher pixel values are included in the image.

Above:  New Screen Stretch settings.

Above:  The image after adjusting the Screen Stretch.

While more of the core detail is visible, the faint parts of the image are perhaps a bit dim.  In order to make the background as dark as possible without losing any detail in the faintest portions of the image, drag the red slider to the very left end of the histogram curve (but not all the way left in the histogram).

Above:  New Screen Stretch and image.

Again, the display stretching only affects how the image is seen on the monitor.  The ideal solution would be to bring both the bright details into view while keeping the fainter detail bright.  This can be done with permanent stretching.

Next, Permanent Stretching...

Next Page