Study on automated incident detection algorithms based on Multi-SVM Classifier

Author

Zhili, Cai ; Guiyan, Jiang ; Qiushi, Ding

Author_Institution

Shandong Jiaotong Univ., Jinan

fYear

2008

fDate

2-4 July 2008

Firstpage

1358

Lastpage

1362

Abstract

Aimed at the problem that many AID algorithms have lower detection rate and higher false alarming rate, a kind of AID algorithm based on multi-SVM classifier is proposed. Also, the framework and flow of this algorithm were designed, and eigenvector constituted by data of the same traffic parameter in some continuous periods was proposed, then the validity and portability were analyzed by simulation data. The results showed that the detection performance of the new algorithm is superior to the comparing algorithms.

Keywords

automated highways; eigenvalues and eigenfunctions; pattern classification; road traffic; support vector machines; eigenvector; freeway; intelligent transportation; multi-SVM classifier; traffic incident detection; Air transportation; Airports; Algorithm design and analysis; Analytical models; Detection algorithms; Educational institutions; Intelligent transportation systems; Support vector machine classification; Support vector machines; Traffic control; Freeway; Intelligent Transportation; SVM Classifier; Traffic Incident Detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and Decision Conference, 2008. CCDC 2008. Chinese

Conference_Location

Yantai, Shandong

Print_ISBN

978-1-4244-1733-9

Electronic_ISBN

978-1-4244-1734-6

Type

conf

DOI

10.1109/CCDC.2008.4597539

Filename

4597539