27th Congress of International Council of the Aeronautical Sciences, 19 - 24 September 2010, Nice, France
Paper ICAS2010-7.1ST1

CORRESPONDENCE AND CLUSTERING TECHNIQUES OF A STEREO VISION SYSTEM FOR THE DETECTION OF OBSTACLES AROUND AIRCRAFT IN AERODROME AREAS

J. Gauci, D. Zammit-Mangion
Cranfield Univ., UK

Keywords: correspondence, clustering, stereo vision, obstacle detection

To reduce collisions between aircraft and obstacles in aerodromes, an onboard stereovision-based obstacle detection system has been designed. This paper focuses on two computational blocks of this system: correspondence and clustering. The implementation of these blocks is discussed and the results of tests using synthetic and real images are presented.

view full paper