Using Stereo Vision and Fuzzy Systems for Detecting and Tracking People
Identificadores
URI: https://hdl.handle.net/10481/86424Metadata
Show full item recordEditorial
Springer
Materia
People Tracking Stereo Vision Fuzzy systems Particle Filtering Color Information
Date
2010Referencia bibliográfica
Paúl, R., Aguirre, E., García-Silvente, M., Muñoz-Salinas, R. (2010). Using Stereo Vision and Fuzzy Systems for Detecting and Tracking People. In: Hüllermeier, E., Kruse, R., Hoffmann, F. (eds) Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications. IPMU 2010. Communications in Computer and Information Science, vol 81. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-14058-7_60
Sponsorship
FCT Scolarship SFRH/BD/22359/2005; Spanish MCI Project TIN2007-66367; Andalusian Regional Government project P09- TIC-0481Abstract
This paper describes a system capable of detecting and tracking
various people using a new approach based on stereo vision and fuzzy
logic. First, in the people detection phase, two fuzzy systems are used to
assure that faces detected by the OpenCV face detector actually correspond
to people. Then, in the tracking phase, a set of hierarchical fuzzy
systems fuse depth and color information captured by a stereo camera
assigning different confidence levels to each of these information sources.
To carry out the tracking, several particles are generated while fuzzy systems
compute the possibility that some generated particle corresponds
to the new position of people. The system was tested and achieved interesting
results in several situations in the real world.