Title :
Prediction of the Position of Pedestrian Crossing Road Section Based on Kalman Predictor
Author :
Ma, Guo-Sheng ; Yao, Jiao ; Yang, Xiao-Guang
Author_Institution :
Sch. of Transp. Eng., Tongji Univ., Shanghai, China
Abstract :
In order to effectively guarantee the pedestrian safety, coordinate relations between the image pixels and actual position of road is founded based on zebra crossing feature. A method which detects pedestrian and predicts the position of pedestrian in real-time for use is presented based on Kalman predictor. By this method, pedestrian-vehicle collision point can be prejudged and pedestrian information is provided for driver. Finally detecting and predicting method experiments were performed and the method has been proved to be correct.
Keywords :
Kalman filters; driver information systems; image processing; road accidents; road safety; road traffic; road vehicles; Kalman predictor; driver information system; image pixel; pedestrian crossing road section; pedestrian safety; pedestrian-vehicle collision point; position prediction; traffic accident; zebra crossing feature; Calibration; Cameras; Entropy; Genetic mutations; Information theory; Kalman filters; Pixel; Road accidents; Road safety; Road transportation; Kalman predictor; interesting area; zebra crossing;
Conference_Titel :
Measuring Technology and Mechatronics Automation, 2009. ICMTMA '09. International Conference on
Conference_Location :
Zhangjiajie, Hunan
Print_ISBN :
978-0-7695-3583-8
DOI :
10.1109/ICMTMA.2009.255
