Title :
Recognition and 3D-reconstruction of objects from images using a priori information
Author :
Zakharov, A.A. ; Zhiznyakov, A.L.
Author_Institution :
Murom Inst. (branch), Vladimir State Univ. named after Alexander G. & Nikolay G. Stoletovs, Vladimir, Russia
Abstract :
The probabilistic approach of three-dimensional reconstruction of the visual environment of urban scenes from satellite and aerial images is presented. The mathematical model of the reconstructed objects is presented. Contour images of models of reconstructed objects are shown. Conditional probability of occurrence of recognizable signs and reconstructed objects are used in the model. Hough transform is used for feature extraction. The approach aim is to find maximum a posteriori probability of the synthesized model. A maximum a posteriori probability is using Monte Carlo Markov chain scheme. Possible transitions between the models for the iterative search are presented in the paper.
Keywords :
Markov processes; Monte Carlo methods; feature extraction; image reconstruction; maximum likelihood estimation; object recognition; probability; 3D-reconstruction; Hough transform; Monte Carlo Markov chain scheme; conditional probability approach; feature extraction; iterative search; mathematical model; maximum a posteriori probability; object recognition; three-dimensional reconstruction; urban scenes; Image recognition;
Conference_Titel :
Microwave & Telecommunication Technology (CriMiCo), 2014 24th International Crimean Conference
Conference_Location :
Sevastopol
Print_ISBN :
978-966-335-412-5
DOI :
10.1109/CRMICO.2014.6959436
