DocumentCode :
2532804
Title :
Pedestrian classification in automotive radar systems
Author :
Heuel, Steffen ; Rohling, Hermann
Author_Institution :
Dept. of Telecommun., Hamburg Univ. of Technol., Hamburg, Germany
fYear :
2012
fDate :
23-25 May 2012
Firstpage :
39
Lastpage :
44
Abstract :
Urban traffic is dangerous for all car drivers and especially for pedestrians. There are about 5000 fatalities on German streets every year, which is absolutely too much. Human beings have strong limitations in the ability to measure precisely the distance and the speed difference between cars, which is the reason for several accidents. Therefore, the European Union has called all car manufactureres to intensify their research activities in protecting vulnerable road users. An automotive radar sensor measures the target range and radial velocity very accurately and with high resolution even in multiple target situations and there is tremendous progress in 24 GHz radar sensor development. The objective of this paper is to protect vulnerable road users by a target recognition scheme in any urban scenario. One main objective is the difference between lateral moving vehicles and pedestrians in terms of feature extraction and classification. Therefore, a pedestrian detection procedure is additionally integrated into the 24 GHz automotive radar sensor.
Keywords :
distance measurement; feature extraction; image classification; radar target recognition; road vehicle radar; velocity measurement; automotive radar systems; car drivers; distance measurement; feature extraction; frequency 24 GHz; lateral moving vehicles; pedestrian classification; pedestrian detection; radar sensor development; radial velocity measurement; speed difference; target range; target recognition scheme; urban traffic; vulnerable road users; Doppler effect; Doppler radar; Equations; Radar measurements; Vehicles; Velocity measurement; 24 GHz; Automotive; Doppler Spectrum; Lateral Velocity; Pedestrian; Radar; Range Profile; Recognition;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Radar Symposium (IRS), 2012 13th International
Conference_Location :
Warsaw
ISSN :
2155-5754
Print_ISBN :
978-1-4577-1838-0
Type :
conf
DOI :
10.1109/IRS.2012.6233285
Filename :
6233285
Link To Document :
بازگشت