Classification of laserscanner measurements at intersection scenarios with automatic parameter optimization

Author

Wender, Stefan ; Schoenherr, Michael ; Kaempchen, Nico ; Dietmayer, Klaus

Author_Institution

Dept. of Measure., Control & Microtechnol., Ulm Univ., Germany

fYear

2005

fDate

6-8 June 2005

Firstpage

94

Lastpage

99

Abstract

Object classification at intersection scenarios is necessary in order to provide a general environment description. Objects are observed using a multilayer laserscanner. Significant features for object classification are identified and their extraction is described. Classification is performed using well-known techniques of statistical learning. Classification results of several neural networks are described and compared with classification performance of support vector machines.

Keywords

automated highways; feature extraction; image classification; learning (artificial intelligence); neural nets; object detection; optical scanners; road safety; automatic parameter optimization; feature extraction; intersection scenario; laserscanner measurement; neural network; object classification; road safety; statistical learning; support vector machine; Accidents; Distance measurement; Feature extraction; Laser modes; Multi-layer neural network; Neural networks; Statistical learning; Support vector machine classification; Support vector machines; Vehicle safety;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Vehicles Symposium, 2005. Proceedings. IEEE

Print_ISBN

0-7803-8961-1

Type

conf

DOI

10.1109/IVS.2005.1505084

Filename

1505084