Hutech Astronomical Products
Equipment for the Serious Amateur Astronomer


 

Home


 

Why Floating Point Processing?

 

 

Internally, StellaImage uses 32-bit floating point (real) data format for each component (color) of image data, no matter what the original input format of the image. This preserves the integrity of the data throughout the processing sequence used on the data. This is of particular importance when iterative advanced processing techniques such as Lucy-Richardson deconvolution needs to be applied to images.

To avoid degrading data when saving processed images, StellaImage supports output (and input) in 32-bit floating point FITS format. This allows further processing to be done on saved images without worrying about image degradation caused by saving to files with lower dynamic range.

In contrast, popular programs such as Photoshop are not designed to handle wide dynamic range astronomical data. Most operations involve only 8-bit integer data (per color) though some operations may be allowed on 16-bit integer images. These programs are suitable only for simple processing of astronomical images, or for applying the final touchup to an image meant for web display or printing.  


Copyright 2000-2001 by Hutech Corporation and ScienceCenter.Net
For technical difficulties with this site email: hutech-web@sciencecenter.net