Citation

Bonneson JA, Fitts JW. Transp. Res. Rec. 1995; 1477: 31-40.

Copyright

(Copyright © 1995, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Sources of bias and variability in the measurement of time-based traffic events by using video recording-and-playback systems are described. Bias can affect accuracy, whereas the lack of precision can increase the minimum sample size. Equations are described that can be used to estimate the adjustment needed to remove bias from the recorded data and to estimate the standard deviation of the measurement process. The equations are sensitive to the method of extracting the individual traffic event times from the videotape. These methods include manual extraction, which uses a frame-by-frame analysis, and automated extraction, which uses a video imaging system to analyze the tape during playback at normal speed. The manual method is found to yield less bias and lower variability than the automated method.

Language: en

Keywords

Data acquisition; Image analysis; Video cameras; Feature extraction; Highway traffic control; Estimation; Statistical methods; Imaging systems; Measurements; Videocassette recorders