DocumentCode
3226128
Title
The Method of Ostu Used in the Video Detection of Vehicles
Author
Zhang, Huiling ; Lu, Baichuan ; Yu, Yong
Author_Institution
Sch. of Traffic Transp., Chongqing Jiaotong Univ., Chongqing
Volume
2
fYear
2008
fDate
20-22 Oct. 2008
Firstpage
547
Lastpage
551
Abstract
The paper researches the threshold segmentation arithmetic of the video vehicle measurement. It is the image´s preprocessing that using the method of least square fitting to adjust the primal image and dealing the image with the median filter. For the threshold selection, the article gives the theory of the state method, 2-mode method, iterative method and Ostu method and analyses them to make the vehicle separating with the background. Through the dealing results, the state method is the most simple method, but it is easy effected by the researcher; The 2-mode method can be used in certain image and the result approved that this method is not fit for the video vehicle image; The iterative method has similar result as the Ostu method, but it has complex programming and need long time to process. So the Ostu method is selected to processing the video vehicle image in this project. Through test, the Ostu method can make the parameters of the vehicles´ extracted effectively.
Keywords
iterative methods; least squares approximations; traffic engineering computing; video signal processing; 2-mode method; Ostu method; image preprocessing; iterative method; least square fitting; median filter; state method; threshold segmentation arithmetic; vehicle video detection; video vehicle measurement; Automation; Filtering; Image segmentation; Iterative methods; Nonlinear filters; Pixel; Smoothing methods; Transportation; Vehicle detection; Vehicles; Ostu; traffic engineering; vehicle´s video detection;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Computation Technology and Automation (ICICTA), 2008 International Conference on
Conference_Location
Hunan
Print_ISBN
978-0-7695-3357-5
Type
conf
DOI
10.1109/ICICTA.2008.219
Filename
4659823
Link To Document