Restoration of Degraded Video Through Turbulent Atmosphere

The optical effects of atmospheric turbulence arise from random inhomogeneities in the temperature distribution of the atmosphere. As the consequence of these temperature inhomogeneities, the index of refraction distribution of the atmosphere is random and evolves with time. For distant objects being viewed through a sensor, atmospheric turbulence can make those still objects appear blurred and the time-varying nature of this turbulence can make the objects appear to waver in a slow quasi-periodic fashion. This phenomenon greatly hinders accurate target detection, tracking, classification, and identification. Current post image processing approach to mitigating this problem assumes space- and time-invariant point spread function (PSF) of the degradation procedure for simplicity and mathematical tractability. Aimed at this limitation, this paper presents a bind source separation based algorithm that can accommodate the space- and time- variant nature of atmospheric turbulence. The preliminary result demonstrates its effectiveness and simplicity in the restoration of videos degraded by atmospheric turbulence.