The traditional incident detection is based on time axis or site axis to judge the events by vertical or transverse analysis, which losing the relativity of the monitoring points and the time continuity of every detectors. In this paper, wavelet and neural network are used to analyze the testing data by their own features from vertical or transverse. Wavelet Multi_resolution Analysis is used to searching the data mutational sites in time domain, and the monitoring points data at the same moment are dealt with neural network. Based on the relationship between Wavelet transform coefficient and parameter change of traffic stream, and the Pattern Recognition of neural network, the precision of the accident detection is improved greatly.
The Research of Freeway Incident Detection System Based on Wavelet and Neural Network
Second International Conference on Transportation Engineering ; 2009 ; Southwest Jiaotong University, Chengdu, China
2009-07-29
Conference paper
Electronic Resource
English
Research , Highways and roads , Information technology (IT) , Transportation management , Traffic accidents , Construction , Freight transportation , Neural networks , Water transportation , Air transportation , Rail transportation , Wavelet , Optimization , Public transportation , Traffic management
The Research of Freeway Incident Detection System Based on Wavelet and Neural Network
British Library Conference Proceedings | 2009
|Freeway Incident Detection Research Based on Wavelet Analysis
British Library Conference Proceedings | 2007
|Freeway incident detection using hybrid fuzzy neural network
Tema Archive | 2007
|