“Eyes on the Street”

Estimating Natural Surveillance Along Amsterdam’s City Streets Using Street-Level Imagery

Book Chapter (2023)
Author(s)

T. van Asten (Dutch Ministry of the Interior and Kingdom Relations, The Hague)

V. Milias (TU Delft - Human-Centred Artificial Intelligence)

A. Bozzon (TU Delft - Human-Centred Artificial Intelligence)

A. Psyllidis (TU Delft - Internet of Things)

Research Group
Human-Centred Artificial Intelligence
Copyright
© 2023 T. van Asten, V. Milias, A. Bozzon, A. Psyllidis
DOI related publication
https://doi.org/10.1007/978-3-031-31746-0_12
More Info
expand_more
Publication Year
2023
Language
English
Copyright
© 2023 T. van Asten, V. Milias, A. Bozzon, A. Psyllidis
Research Group
Human-Centred Artificial Intelligence
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. @en
Pages (from-to)
215-229
ISBN (print)
978-3-031-31745-3
ISBN (electronic)
978-3-031-31746-0
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Neighborhood safety and its perception are important determinants of citizens’ health and well-being. Contemporary urban design guidelines often advocate urban forms that encourage natural surveillance or “eyes on the street” to promote community safety. However, assessing a neighborhood’s level of natural surveillance is challenging due to its subjective nature and a lack of relevant data. We propose a method for measuring natural surveillance at scale by employing a combination of street-level imagery and computer vision techniques. We detect windows on building facades and calculate sightlines from the street level and surrounding buildings across forty neighborhoods in Amsterdam, the Netherlands. By correlating our measurements with the city’s Safety Index, we also validate how our method can be used as an estimator of neighborhood safety. We show how perceived safety varies with window level and building distance from the street, and we find a non-linear relationship between natural surveillance and (perceived) safety.

Files

978_3_031_31746_0_12.pdf
(pdf | 0.843 Mb)
- Embargo expired in 02-12-2023
License info not available