Naslov
Dependent Component Analysis for Blind Restoration of Images Degraded by Turbulent Atmosphere

Autori
Du, Qian ; Kopriva, Ivica

Izvornik
Neurocomputing (0925-2312) 72 (2009), 10-12; 2682-2692

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
atmospheric turbulence ; image restoration ; independent component analysis ; dependent component analysis

Sažetak
Atmospheric turbulence is an inevitable problem in long-distance ground-based and space-based imaging. The optical effects of atmospheric turbulence arise from random inhomogeneities in the temperature distribution of the atmosphere. A consequence of these temperature inhomogeneities is non-stationary random distribution of the refraction index of the atmosphere [1]. Atmospheric turbulence can make distant objects being viewed through a sensor (e.g., a digital camera or video recorder) to appear blurred. Also, the time-varying nature of the turbulence can make the appearance of objects to wave in a slow quasi-periodic fashion. When a target is small and moving, its actual location becomes very difficult to estimate. This phenomenon greatly hinders accurate target detection, tracking, classification, and identification.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Računarstvo

POVEZANOST RADA