Dolynska I. Modelling and researches of modulation transfer function of space remote sensing systems

The relevance of the dissertation is determined by necessity to compute the real spatial resolution taking into account the influence of internal and external factors on the quality of space images. The mathematic modeling of modulation transfer function taking in to account the influences of main factors together with threshold modulation function was performed to determine the real spatial resolution. The results of resolution computation for satellite images of different imaging systems reveal that the effect of the optical transmission system, residual defocusing, diffraction, discrete sensor and atmosphere influence can in 1,2-1,5 times decrease the spatial resolution of satellite images compared to the same values given by manufacturers of appropriate imaging systems. A dominant effect on image quality has an atmosphere state, including its aerosol component. In this manuscript two methods which allow taking into account the influence of the atmosphere on satellite images are considered: based on standard models; using vegetation indices resistant to the atmosphere influences. The method allowing to remove the aerosol component from satellite image that does not require information about the state of the atmosphere proposed and implemented. The method is based on construction of a "dark" image channel, determining the transmittance of the atmosphere, eliminating artifacts and computing the restored image. Satellite images of such satellites remote sensing as QuickBird-2, Ikonos-2, WorldView-2, Landsat-8, Pleiades-1 was used for different atmosphere influence methods researches.