STATISTICALLY CORRECTED OCEAN THERMOGRAPHY

A corrected thermal imaging method for a wind-roughened water surface is proposed. This method can be applied even if the imaging system is placed at low altitudes, say on a ship or on land, as well as on an airborne platform. Such a technique is desirable from the following points of view: the surveillance of bioresources (fish), environmental assessments of a seaside industrial zone (hot waste water), and temporal complement and spatial interpolation of satellite observations of thermal images. The method is based on the analysis of optical characteristics of a model of a wind-roughened water surface, namely, the Gaussian-Joint North Sea Wave Project (JONSWAP) model, where the surface displacement obeys Gaussian distribution while its spectrum is specified by the JONSWAP wave spectral model. We present the basic temperature-correction formula and the algorithm for correction. The correction takes into account thermal emission of the water's surface, radiation of the sky reflected on the water's surface, and both absorption and emission by the atmosphere along the light path. This formula can be used for temperature correction of an infrared image of sky and random water surface. The experimental results that we obtained are encouraging.