20.500.12556/DKUM-16825 MODELIRANJE SLIK SAR Z AVTO-BINOMSKIM MODELOM MODELING SAR IMAGES WITH AUTO-BINOMIAL MODEL Doktorska disertacija predstavlja odpravljanje pegastega šuma z modeliranjem in izločanjem informacij iz posnetkov SAR (ang. Synthetic Aperture Radar - SAR), ki je izvedeno z Bayesovim sklepanjem. Novost, ki jo predstavljam v doktorski disertaciji je uporaba avto-binomskega modela pri Bayesovem sklepanju prvega reda, kjer ga za apriorno verjetnost uporabimo pri modeliranju posnetka. Verjetje v Bayesovem sklepanju prvega reda modelira pegasti šum, ki ga opišemo z gama porazdelitvijo. Odpravljanje pegastega šuma je izvedeno s cenilko največje verjetnosti MAP (ang. maximum a posteriori - MAP), ki jo je analitično zelo težko rešiti, zato z avto-binomskim modelom uporabimo aproksimacijo z diferencialom. Parametri avto-binomskega modela se določijo z Bayesovim sklepanjem drugega reda. Robovi v posnetku se določijo z algoritmom rasti regij. Glede na koeficient variacije so ločene homogene od heterogenih regije in adaptivno prilagajanje okolice avto binomskega modela. Eksperimentalni rezultati so pokazali, da predstavljena metoda zelo dobro modelira teksture in ima dobre lastnosti odpravljanja pegastega šuma in zelo dobro modelira teksture v realnih posnetkih SAR. This dissertation presents a model based despeckling of Synthetic Aperture Radar (SAR) images using Bayesian analysis. The use of first order of Bayesian inference with auto-binomial model is used for SAR image despeckling. The novelty of dissertation is an auto-binominal model, which models a prior probability density function (pdf). Speckle models the likelihood pdf, represented by Gamma distribution. An analytical solution of a maximum a posteriori (MAP) estimate using an auto-binominal prior is difficult to compute. Therefore, the approximate solution of MAP estimate is proposed using the differential. The best auto-binomial model is determined by the second order Bayesian inference. The edges in the SAR images are detected using region borders, which have statistically different properties. Coefficient of variation is used to distinguish between homogeneous and heterogeneous regions. The experimental results show that the proposed method is one of the best algorithms for the textural features extraction and its capability of speckle removing. The proposed despeckling method is comparable with state-of-the-art methods and well models textures presented in real SAR images. Avto-binomski model izločevanje informacij maksimum a posteriori (MAP) cenilka Synthetic aperture radar (SAR) ohranjevanje tekstur. Auto-binomial model information extraction maximum a posteriori (MAP) estimation Synthetic aperture radar (SAR) texture preservation. true true false [M. Hebar] Slovenski jezik Angleški jezik Doktorska disertacija 2010-11-24 13:24:41 2011-01-24 08:36:06 2022-04-13 08:56:32 2023-12-21 03:19:22 2010 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.93:621.396.963 254665472 URN:SI:UM:DK:XUBSJKV0 DR_Hebar_Marko_2010.pdf DR_Hebar_Marko_2010.pdf 1 B3A6251FA43C42916E0B51F8777F69BA 8eb234a02811dee0bf1e5190c6a1ccb4e183893817268f6d8ea66a28347a3a5b bd50f798-7c04-11eb-bb7a-00155d0001ca 20.500.12556/dkum/f3170827-47dc-400b-990b-360f632c0aa7 https://dk.um.si/Dokument.php?lang=slv&id=19264 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0