The Accuracy of Vehicle Speeds, Decelerations, and Brake Onset Times Calculated from Onboard Dash Cameras

<div class="section abstract"><div class="htmlview paragraph">Videos from cameras onboard a moving vehicle are increasingly available to collision reconstructionists. The goal of this study was to evaluate the accuracy of speeds, decelerations, and brake onset times calculated from onboard dash cameras (“dashcams”) using a match-moving technique. We equipped a single test vehicle with 5 commercially available dashcams, a 5th wheel, and a brake pedal switch to synchronize the cameras and 5th wheel. The 5th wheel data served as the reference for the vehicle kinematics. We conducted 9 tests involving a constant-speed approach (mean ± standard deviation = 57.6 ± 2.0 km/h) followed by hard braking (0.989 g ± 0.021 g). For each camera and brake test, we extracted the video and calculated the camera’s position in each frame using SynthEyes, a 3D motion tracking and video analysis program. Scale and location for the analyses were based on a 3D laser scan of the test site. From each camera’s position data, we calculated its speed before braking and its average deceleration during braking, and we then compared these values to the reference speed and average deceleration, respectively, to estimate a bias and uncertainty for each camera and kinematic parameter. Across the 5 cameras tested here, speed estimates varied from an underestimate of 0.45 ± 0.05 km/h (bias ± uncertainty) to an overestimate of 0.80 ± 0.16 km/h, and average deceleration estimates varied from an underestimate of 0.012 ± 0.013 g to an overestimate of 0.049 ± 0.012 g. The video-based brake onset times lagged the actual brake onset times for four of the five cameras, varying from 14 ± 41 ms before brake onset to 94 ± 29 ms after brake onset. These data show that, for cameras of the quality tested here, dashcam video can be used to estimate vehicle speed, average deceleration, and the time of brake application with relatively small biases and uncertainties that vary between cameras.</div></div>

Authors

Publication Details

Journal
SAE technical papers on CD-ROM/SAE technical paper series
Published
2025-03-31
DOI
https://doi.org/10.4271/2025-01-8691
Citations
1
Primary Topic
Vehicle Dynamics and Control Systems
Type
article
Field-Weighted Citation Impact
0.68
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

The Accuracy of Vehicle Speeds, Decelerations, and Brake Onset Times Calculated from Onboard Dash Cameras

MB Ahrens, Gunter P. Siegmund, Cole Young, Thomas R. Flynn
1 citations
SAE technical papers on CD-ROM/SAE technical paper series
Vehicle Dynamics and Control Systems
0.68
article

The Accuracy of Vehicle Speeds, Decelerations, and Brake Onset Times Calculated from Onboard Dash Cameras

MB Ahrens, Gunter P. Siegmund, Cole Young, Thomas R. Flynn
article en
1 citations

Abstract

<div class="section abstract"><div class="htmlview paragraph">Videos from cameras onboard a moving vehicle are increasingly available to collision reconstructionists. The goal of this study was to evaluate the accuracy of speeds, decelerations, and brake onset times calculated from onboard dash cameras (“dashcams”) using a match-moving technique. We equipped a single test vehicle with 5 commercially available dashcams, a 5th wheel, and a brake pedal switch to synchronize the cameras and 5th wheel. The 5th wheel data served as the reference for the vehicle kinematics. We conducted 9 tests involving a constant-speed approach (mean ± standard deviation = 57.6 ± 2.0 km/h) followed by hard braking (0.989 g ± 0.021 g). For each camera and brake test, we extracted the video and calculated the camera’s position in each frame using SynthEyes, a 3D motion tracking and video analysis program. Scale and location for the analyses were based on a 3D laser scan of the test site. From each camera’s position data, we calculated its speed before braking and its average deceleration during braking, and we then compared these values to the reference speed and average deceleration, respectively, to estimate a bias and uncertainty for each camera and kinematic parameter. Across the 5 cameras tested here, speed estimates varied from an underestimate of 0.45 ± 0.05 km/h (bias ± uncertainty) to an overestimate of 0.80 ± 0.16 km/h, and average deceleration estimates varied from an underestimate of 0.012 ± 0.013 g to an overestimate of 0.049 ± 0.012 g. The video-based brake onset times lagged the actual brake onset times for four of the five cameras, varying from 14 ± 41 ms before brake onset to 94 ± 29 ms after brake onset. These data show that, for cameras of the quality tested here, dashcam video can be used to estimate vehicle speed, average deceleration, and the time of brake application with relatively small biases and uncertainties that vary between cameras.</div></div>

SAE technical papers on CD-ROM/SAE technical paper seriesVol. 1
Openalex Percentile: Top 32%
Vehicle Dynamics and Control Systems
0.68
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.