aceacedey / kolkata_data_pmc_paper_trafficcountestimationusingcrowdsourcedtrajectory

The repository includes detailed data descriptions and the codes utilised in this article titled Traffic count estimation using crowd-sourced trajectory data in the absence of dedicated infrastructure, published in Pervasive and Mobile Computing.

Kolkata_Data_PMC_paper_TrafficCountEstimationUsingCrowdSourcedTrajectory

This release contains the raw data that are used in the paper titled Traffic count estimation using crowd-sourced trajectory data in the absence of dedicated infrastructure.

The raw trajectory data interpreted camera data with signal timing information, traffic counts and queue lengths data were collected at a single-lane signalised road intersection at the Webel-More in Kolkata. Validation datasets are manually collected using a video camera installed at the observer's location. A ground-truth dataset of vehicle counts, departure headways, and signal timing is manually constructed using time-stamped video camera footage and visual inspection. A smartphone-equipped vehicle with the GPSlogger Android application is used to collect the set of vehicle trajectory data.