Title :
Automatic extraction of information for catenary scene analysis
Author :
Montreuil, Florent ; Kouadio, Regis ; Petitjean, Caroline ; Heutte, Laurent ; Delcourt, Vincent
Author_Institution :
LITIS, Univ. de Rouen, St. Etienne-du-Rouvray, France
Abstract :
This paper presents an original system for the automatic recognition of catenary elements. Based on a bottom-up approach, our analysis is composed of two stages: the first stage provides recognition hypothesis regarding previously segmented elements, with a precision rate of 91.7 %. The second stage allows to correct some hypothesis using a Markovian model which analyzes the catenary element sequence. It allows to improve precision up to 98.1%.
Keywords :
Markov processes; image retrieval; image sequences; object recognition; power overhead lines; railway electrification; Markovian model; automatic catenary element recognition; automatic information extraction; bottom-up approach; catenary element sequence analysis; catenary scene analysis; overhead wires; recognition hypothesis; Analytical models; Europe; Feature extraction; Hidden Markov models; Image segmentation; Signal processing; Wires;
Conference_Titel :
Signal Processing Conference, 2008 16th European
Conference_Location :
Lausanne
