Image Processing Reference
In-Depth Information
Chapter 1
Image Stabilization
1.1 Introduction
Image stabilization is the technique of improving image quality by actively remov-
ing the apparent motion of an object induced by vibration, tracking errors, and dif-
ferential refraction in the atmosphere. It is the apparent motion of the object be-
cause usually the object itself is quite stable, yet in the imaging system the object
appears to be moving. The result of using image stabilization is an image that is
sharper and has higher contrast and resolution.
In astronomical imaging, image stabilization is an important tool for removing
the effects of the earth's atmosphere. Light from a distant star or other heavenly ob-
ject travels through the vacuum of space and arrives at a space telescope having un-
dergone very little in the way of physical changes. However, when that same light
is propagated through a turbulent medium such as earth's atmosphere and reaches
an earth-based telescope, the arriving light may have little more than its frequency
in common with its transmitted form (Tyson 2000)! For centuries, astronomers
have observed this effect as the blurry images seen through their telescopes; how-
ever, only recently have scientists begun to understand and develop techniques to
compensate for this effect (Babcock 1953).
The distorting effect of a turbulent medium on light passing through it im-
poses a limit on the performance of optical systems. As a result, there is signifi-
cant demand for atmospheric compensation systems for use in astronomical,
communications, and military systems. New applications are also being devel-
oped for industrial and medical purposes (Love 1999; Restaino and Teare 2000;
Wittrock 2003).
Image stabilization is arguably one of the most important advances in the sci-
ence of imaging. The techniques and instrumentation used to correct for image mo-
tion are based on understanding the physical effects that degrade image quality. To
this end, two important components of image motion can be identified as beam
wander and image dancing, defined by Lawrence (2004) as angular deviation of the
beam from the line-of-sight path, possibly causing the beam to miss the receiver;
and fluctuations in the beam angle of arrival, causing the focus to move (“dance”)
in the image plane, respectively.
It is convenient at this point to associate beam wander with the element of a
system transmitting a beam of light and image dancing with the element receiving
1
