English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Accurate travel time estimation is paramount for providing transit users withreliable schedules and dependable real-time information. This paper is thefirst to utilize roadside urban imagery for direct transit travel timeprediction. We propose and evaluate an end-to-end framework integratingtraditional transit data sources with a roadside camera for automated roadsideimage data acquisition, labeling, and model training to predict transit traveltimes across a segment of interest. First, we show how the GTFS real-time datacan be utilized as an efficient activation mechanism for a roadside camera unitmonitoring a segment of interest. Second, AVL data is utilized to generateground truth labels for the acquired images based on the observed transittravel time percentiles across the camera-monitored segment during the time ofimage acquisition. Finally, the generated labeled image dataset is used totrain and thoroughly evaluate a Vision Transformer (ViT) model to predict adiscrete transit travel time range (band). The results illustrate that the ViTmodel is able to learn image features and contents that best help it deduce theexpected travel time range with an average validation accuracy ranging between80%-85%. We assess the interpretability of the ViT model's predictions andshowcase how this discrete travel time band prediction can subsequently improvecontinuous transit travel time estimation. The workflow and results presentedin this study provide an end-to-end, scalable, automated, and highly efficientapproach for integrating traditional transit data sources and roadside imageryto improve the estimation of transit travel duration. This work alsodemonstrates the value of incorporating real-time information fromcomputer-vision sources, which are becoming increasingly accessible and canhave major implications for improving operations and passenger real-timeinformation.

Computer Vision for Transit Travel Time Prediction: An End-to-End Framework Using Roadside Urban Imagery