ETD system

Electronic theses and dissertations repository

 

Tesi etd-07022018-144035


Thesis type
Tesi di laurea magistrale
Author
FIORETTI, FEDERICA
URN
etd-07022018-144035
Title
A single camera UKF algorithm for the estimation of train trajectories and mapping of relevant obstacles close to railway lines
Struttura
INGEGNERIA DELL'INFORMAZIONE
Corso di studi
INGEGNERIA ROBOTICA E DELL'AUTOMAZIONE
Commissione
relatore Prof. Avizzano, Carlo Alberto
relatore Ing. Ruffaldi, Emanuele
controrelatore Prof. Pollini, Lorenzo
Parole chiave
  • projective geometry
  • unscented kalman filter
  • structure from motion
Data inizio appello
19/07/2018;
Consultabilità
secretata d'ufficio
Riassunto analitico
This thesis presents a Structure from Motion pipeline specifically designed with the aim of supporting the monitoring operations of the railway infrastructure using a monocular camera mounted on the train's tractor. The visual instrument adopts a line-based approach to assess the camera poses together with a manifold Unscented Kalman Filter, which operates on the set of robustly matched features. Consequently, the train and the detected objects can be localized in metric dimensions. Results for a simulated scenario and for real images are reported.
File