Title :
A multimodal framework for vehicle and traffic flow analysis
Author :
Ploetner, Jeffrey ; Trivedi, Mohan M.
Author_Institution :
Comput. Vision & Robotics Res. Lab., California Univ., La Jolla, CA
Abstract :
This paper presents an overview of a novel multimodal system being developed at UC San Diego for vehicle detection and traffic flow analysis. A distributed multimodal array (DiMMA) framework is presented for sensory data acquisition, processing, analysis, fusion, and "active" control mechanisms needed to recognize objects, events, and activities which have multi-modal signatures. Current sensing modalities being researched include video, audio, seismic, magnetic, and passive infrared. Feature extraction and data fusion techniques are being investigated to improve robustness and study the advantages and disadvantages of each sensing modality. Preliminary results of this rapidly deployable system are discussed, along with possible future expansions, including laser range scanners, geophones, pneumatic road tubes, and traditional inductive loops
Keywords :
data acquisition; feature extraction; object recognition; sensor fusion; traffic control; vehicles; data fusion; distributed multimodal array; feature extraction; geophones; inductive loop; laser range scanner; multimodal signature; multimodal system; object recognition; pneumatic road tube; sensory data acquisition; traffic flow analysis; vehicle detection; Cameras; Detectors; Information security; Magnetic analysis; Monitoring; Road vehicles; Robustness; Sensor arrays; Traffic control; Vehicle detection;
Conference_Titel :
Intelligent Transportation Systems Conference, 2006. ITSC '06. IEEE
Conference_Location :
Toronto, Ont.
Print_ISBN :
1-4244-0093-7
Electronic_ISBN :
1-4244-0094-5
DOI :
10.1109/ITSC.2006.1707437
