Title :
Fusion scheme for pedestrian safety in Urban roads
Author :
Garcia, F. ; Olmeda, D. ; de la Escalera, A. ; Armingol, J.M.
Author_Institution :
Intell. Syst. Lab., Univ. Carlos III de Madrid, Leganes, Spain
Abstract :
Trustworthy sensors are the key points regarding road safety applications. The lack of reliable sensors able to fulfil the entire requirements for these situations makes fusion schemes mandatory. A fusion scheme that uses two sensors: far infrared camera vision and a laser scanner to perform pedestrian detection is presented. Both sensors have different field of view thus different detection zones are created according to the sensors available for detection in each zone. The proposed algorithm performs hybrid fusion by combining low and high level information. Experimental results show that combining both sensors detection is improved.
Keywords :
cameras; computer vision; infrared detectors; object detection; optical scanners; road safety; sensor fusion; traffic engineering computing; detection zones; far infrared camera vision; fusion scheme; high level information; hybrid fusion; laser scanner; pedestrian detection; pedestrian safety; reliable sensors; road safety applications; sensors detection; trustworthy sensors; urban roads; Cameras; Classification algorithms; Finite impulse response filter; Head; Laser fusion; Roads; Sensors; Data association; hybrid systems; real-world applications;
Conference_Titel :
Information Fusion (FUSION), 2010 13th Conference on
Conference_Location :
Edinburgh
Print_ISBN :
978-0-9824438-1-1
DOI :
10.1109/ICIF.2010.5712038