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

Traffic cameras remain the primary source data for surveillance activitiessuch as congestion and incident monitoring. To date, State agencies continue torely on manual effort to extract data from networked cameras due to limitationsof the current automatic vision systems including requirements for complexcamera calibration and inability to generate high resolution data. This studyimplements a three-stage video analytics framework for extractinghigh-resolution traffic data such vehicle counts, speed, and acceleration frominfrastructure-mounted CCTV cameras. The key components of the frameworkinclude object recognition, perspective transformation, and vehicle trajectoryreconstruction for traffic data collection. First, a state-of-the-art vehiclerecognition model is implemented to detect and classify vehicles. Next, tocorrect for camera distortion and reduce partial occlusion, an algorithminspired by two-point linear perspective is utilized to extracts the region ofinterest (ROI) automatically, while a 2D homography technique transforms theCCTV view to bird's-eye view (BEV). Cameras are calibrated with a two-layermatrix system to enable the extraction of speed and acceleration by convertingimage coordinates to real-world measurements. Individual vehicle trajectoriesare constructed and compared in BEV using two time-space-feature-based objecttrackers, namely Motpy and BYTETrack. The results of the current study showedabout +/- 4.5% error rate for directional traffic counts, less than 10% MSE forspeed bias between camera estimates in comparison to estimates from probe datasources. Extracting high-resolution data from traffic cameras has severalimplications, ranging from improvements in traffic management and identifydangerous driving behavior, high-risk areas for accidents, and other safetyconcerns, enabling proactive measures to reduce accidents and fatalities.

Application of 2D Homography for High Resolution Traffic Data Collection using CCTV Cameras