Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Land Surveying and Geo-Informaticsen_US
dc.creatorLiu, Yachen-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/6522-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic University-
dc.rightsAll rights reserveden_US
dc.titleA comparative study of algorithms for vehicle detection with elevated video imagesen_US
dcterms.abstractWith the population growth and increasing of vehicles, traffic problem especially congestion problems become very serious. Vehicle detection from video sequence is the basic and important part of traffic surveillance and tracking. Detected vehicles can be further processed for vehicle tracking, calculate average vehicle speed, vehicle flowrate and occupancy. It is essential to detect moving vehicles stably and efficiently especially in complicated background environments. Researchers have proposed many vehicle detection methods. In this thesis, three vehicle detection algorithms including two simple methods frame differencing and approximated median filter and one complicated method Mixture of Gaussians were performed. An experiment including three test sequences with different traffic conditions and different illumination was designed to test performances of three algorithms. Based on the results, qualitative comparison with the performance of detecting slow moving vehicles, performance of vehicle completeness, remove noise and remove cast shadows were performed. Quantitative comparative analysis with recall and precision value was given based on experiment results in the first and the third test sequence. Results show that frame differencing algorithm is the fastest and simplest method among three algorithms. The speed of background updating is fast to eliminate noise such as moving leaves and traffic lanes. Approximated median filter algorithm is easy to implement too and holds a high recall and precision value. Mixture of Gaussians is the most complicated algorithm among three algorithms, but performance of Mixture of Gaussians method in this experiment is not so good because of noise and "holes" on vehicles.en_US
dcterms.extentviii, 53, xiv leaves : ill. (some col.), 1 col. map ; 30 cm.en_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2012en_US
dcterms.educationalLevelAll Masteren_US
dcterms.educationalLevelM.Sc.en_US
dcterms.LCSHTraffic flow -- Mathematical models.en_US
dcterms.LCSHImage analysis -- Mathematical models.en_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertationsen_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
b24752903.pdfFor All Users (off-campus access for PolyU Staff & Students only)4.01 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/6522