This study examines cyclists’ preferences regarding infrastructural and environmental attributes, including lane type, pavement, lane colour, and the presence of greenery, water, and buildings. To do so, we conduct an image-based stated route-choice experiment in which respondents evaluate alternatives that vary in cycling environment, travel time, and the number of traffic lights. For the survey design, we used over six thousand real-world street-level images of the cycling environment, providing a more realistic choice context for respondents than could be achieved with a traditional stated-choice experiment based on tabular data with text and numbers. We analyse these route-choice data using two complementary modelling approaches: traditional discrete choice models with interpretable environmental attributes derived from GIS data, and a recently proposed computer vision-enriched discrete choice model that integrates image-based information directly into the discrete choice modelling framework. Results show that cycling environments are important factors in explaining cyclist route choice behaviour, with cyclists preferring green areas and separated cycling lanes over urban areas and mixed-traffic roads. In addition, we use the developed models to map where the cycling network in Rotterdam, our case study, aligns —or conflicts— with cyclists’ preferences. Finally, we demonstrate how the developed model can provide planners with ex-ante evidence on the prospective effectiveness of proposed infrastructure upgrades.

Perceived traffic safety influences cycling behavior, yet its value relative to travel time and other route attributes remains unquantified. Building on the stated choice experiment and street-level image dataset of Terra et al. (2025), we extract safety perception scores from cycling-perspective images using a computer vision model and estimate mixed logit models to test whether perceived safety affects route choice after controlling for visual street-level features. Cyclists are willing to accept 64 additional seconds of travel time for a one-unit increase in perceived safety (scale: very unsafe to very safe). Safety preferences vary across demographic groups: older cyclists, recreational cyclists, and those with positive cycling attitudes place more weight on safety, while commuters prioritize speed. These willingness-to-pay estimates enable planners to quantify perceived safety improvements for cost-benefit analyses and to score existing cycling networks for targeted infrastructure upgrades. (Replication code: https://github.com/koito19960406/cycling_safety_perception).