Cycling desire line analysis

This is the source code for the scientific paper Computational desire line analysis of cyclists on the Dybbølsbro intersection in Copenhagen by S. Breum, B. Kostic, and M. Szell. The code applies DBSCAN and Dynamic Time Warping to analyze patterns in cyclist behaviour with a data set of 11,553 cyclist trajectories, which had been extracted via a custom-trained YOLO from a high-resolution 1h video from June 2021 of the [https://www.openstreetmap.org/#map=19/55.66524/12.55843](Dybbølsbro intersection).

Paper (preprint): https://arxiv.org/abs/2211.01301

Setting up code environment

The code environment has been successfully tested on the following operating system: Ubuntu 20.04.3 LTS (GNU/Linux 4.4.0-19041-Microsoft x86_64)

To set up the code environment, run:

conda env create -f=environment.yml
conda activate desirelines
conda install -c conda-forge ipywidgets
pip install --user ipykernel
python -m ipykernel install --user --name=desirelines

Download and unpack data folder from https://zenodo.org/record/7288616

Then, run Jupyter Notebook with kernel desirelines (Kernel > Change Kernel > desirelines)