TRB Pubsindex
Text Size:
x

« Prev Next »

Return to Results | New Search

Bookmark and Share

Title:

Vision-Based Analysis of Cyclists' Speed

Accession Number:

01368045

Record Type:

Component

Abstract:

The behavioral studies of road users are very important in the design, planning, analysis, and evaluation of highway and roads. Road users speed is one of the main factors that require knowledge in such behavioral studies. Automated video analysis technique allows researchers to collect cyclist data less expensively than the rest of other methods with a relatively very small probability of error. The purpose of this study is to investigate the potential of measuring a cyclists speed from automated video analysis using computer vision techniques. The cyclist tracking technique presented, estimates cyclist location data based on subsequent video frames that have been extracted. Photo coordinates are translated to roadway coordinates using a homography matrix. Results showed that the cyclists speeds range between 5.09 km/h and 41.07 km/h. The cyclists mean speed estimation was found 15.84 km/h for all observation. The manually average crossing cyclist speed (ground-truth speed) was found 18.29 km/h, while the result using the computer analysis for the same number of cyclist that used same path was found 15.93 km/h Conditions and characteristics of the roadway, light conditions, and traffic conditions were investigated in this study as factors potentially affecting cyclist speed. The effect of longitudinal grade on cyclist speed was clearly demonstrated. Cyclist speed decreased by 33.43% on uphill. Also, the average cyclist speed was found to be 10.97 km/h under dark conditions which is less than the average cyclist speed at day time by 30.74%. The correlation coefficient between measured and calculated cyclist speed was obtained and found to be 0.87.

Supplemental Notes:

This paper was sponsored by TRB committee ANF20 Bicycle Transportation

Monograph Accession #:

01362476

Report/Paper Numbers:

12-3441

Language:

English

Corporate Authors:

Transportation Research Board

500 Fifth Street, NW
Washington, DC 20001 United States

Authors:

Kassim, Ali
Pascoe, Leigh
Ismail, Karim
Abd El Halim, Abd El Halim

Pagination:

17p

Publication Date:

2012

Conference:

Transportation Research Board 91st Annual Meeting

Location: Washington DC, United States
Date: 2012-1-22 to 2012-1-26
Sponsors: Transportation Research Board

Media Type:

Digital/other

Features:

Figures; Photos; References; Tables

Uncontrolled Terms:

Road conditions

Subject Areas:

Highways; Pedestrians and Bicyclists; I20: Design and Planning of Transport Infrastructure

Source Data:

Transportation Research Board Annual Meeting 2012 Paper #12-3441

Files:

TRIS, TRB

Created Date:

Feb 8 2012 5:16PM