IMAGE RESTORATION

Image restoration techniques are normally used to increase the definition of a CCD image. Optical aberrations, seeing, and tracking efficiency affect the images obtained with a CCD detector reducing its sharpness. The blurred image of a star, planet or galaxy can be significantly improved by deconvolving its Point Spread Function (PSF) in such a way that the end result is a sharper and more detailed image.

Several algorithms can be applied to the original image with impressive results. The best are the so-called interactive techniques. The PSF of the image has to be determined before using any image restoration algorithm. This usually consists in isolating a non saturated star in the image to be treated and using this information as its PSF. The software works in an iterative way calculating several approximations of the deconcolved image. Best examples of these algorithms are Maximum Entropy Deconvolution (MEM), Lucy-Richardson Deconvolution (LR) and Van-Cittert Deconvolution (VC). Direct algorithms can also be used with good results, such as the Wiener algorithm. There are however several drawbacks associated with the application of these algorithms (deconvolved images are usually noisy and they can not be used for photometry).

Image restoration techniques can improve the apparent sharpness of a CCD image by two to three times, meaning that medium size telescopes will perform like big telescopes.

M1- deconvolved images
M51- deconvolved images
M51- deconvolved images

Click on images for full-sized view.
Image restoration algorithms (M51). 1- Convolved image (original), 2- Maximum entropy deconvolution, 3- Richardson-Lucy , 4- van Cittert.