Evaluation of Near-Miss Crashes Using a Video-Based Tool [Summary]

Kondyli, Alexandra; Schrock,Steven D.; Tousif, Fahmid; University of Kansas

Advanced Search

Select up to three search categories and corresponding keywords using the fields to the right. Refer to the Help section for more detailed instructions.

Search our Collections & Repository

Advanced Search
Custom Query

All these words:

For very narrow results

This exact word or phrase:

When looking for a specific result

Any of these words:

Best used for discovery & interchangable words

None of these words:

Recommended to be used in conjunction with other fields

Language:

Dates

Publication Date Range:

to

Document Data

Title:

Document Type:

Library

Collection:

Series:

People

Author:

Clear All

Query Builder

Query box

Clear All

For additional assistance using the Custom Query please check out our Help Page

i

Evaluation of Near-Miss Crashes Using a Video-Based Tool [Summary]

2023-11-01
By Kondyli, Alexandra ; Schrock,Steven D. ; Tousif, Fahmid

[PDF-351.06 KB]

English

Details Related Documents You May Also Like

Details:

Creators:

Kondyli, Alexandra ; Schrock,Steven D. ; Tousif, Fahmid

Kondyli, Alexandra ; Schrock,Steven D. ; Tousif, Fahmid Less -
Corporate Creators:

University of Kansas
Corporate Contributors:

Kansas Department of Transportation. Bureau of Research
Subject/TRT Terms:

[+]

Accuracy Evaluation Machine Vision Mathematical Prediction Near Crashes Signalized Intersections Software
Publication/ Report Number:

K-TRAN: KU-21-4
Resource Type:

Brief
Geographical Coverage:

Kansas ; United States

Kansas ; United States Less -
Edition:

Technical Summary
Corporate Publisher:

Kansas Department of Transportation. Bureau of Research
Abstract:

This research investigated the prediction accuracy of a video-based tool developed by Transoft Solutions for predicting near-miss crashes at signalized intersections. The research team selected two signalized intersections in Overland Park, Kansas, and collected two weeks of video data from both locations. Only weekday data were collected for the two weeks in February and March 2021. The data were provided to Transoft Solutions, and analyzed results were accessed from the vendor website. Approximately 10% of the data were sampled for manual validation, which included drawing vehicle trajectories and conflict spots on top of the computer screen and measuring time manually in milliseconds. Both post-encroachment time (PET) and time-to-collision (TTC) data were validated based on three conflict categories (critical conflicts, minor conflicts, and potential conflicts) and three weather and traffic conditions (rainy peak condition, sunny peak condition, and sunny off-peak condition). Four performance measures (mean absolute deviation - MAD, root mean squared error - RMSE, mean absolute percentage error - MAPE, and root mean squared log error - RMSLE) were selected, and an ANOVA with Tukey’s post hoc test was carried out for each analysis. Both the PET and the TTC data showed that sunny weather had better predictability than rainy weather. Statistical analysis revealed a significant difference between means from observed and predicted values for the PET data. Overall, the video-based tool by Transoft Solutions demonstrated moderate predictability and overestimated the conflict measures for both the PET and the TTC data.
Format:

PDF
Collection(s):

US Transportation Collection
Main Document Checksum:

[+]

urn:sha256:c9f1e87a07f217a44753414bc720d04791782bbe8978d5f184e0e53f3959d8ab
Download URL:

https://rosap.ntl.bts.gov/view/dot/72548/dot_72548_DS1.pdf
File Type:

No Additional Files

More +

Evaluation of Near-Miss Crashes Using a Video-Based Tool [Summary]

Details:

Supporting Files

Related Documents

You May Also Like