Classifying 24,318 Camera Entry-Exit Pairs at Six Shopping-Centre Car Parks Against a Paired Exit-Survey and Observation Record
A research poster from the School of Emergent Priorities classifies 24,318 plate-camera entry and exit pairs at six shopping-centre car parks against a paired ground-truth record: an exit-lane intercept survey on 1,180 pairs and 214 hours of lane observation, with classes derived at the first centre and applied blind at the remaining five by two coders (kappa = 0.83). 58.6 per cent of pairs correspond to one shopping party's single stay; relocations within the site (12.4 per cent), read faults (10.1), boundary faults (9.8) and stays covering a party larger than the vehicle (9.1) account for the rest. Compliance recomputed on shopping parties from the same cameras falls from the published 94.1 per cent of pairs to 81.7 per cent.
| Authors | Anneke Tolan, Marek Solheim, Osei Vandermeer |
|---|---|
| School | School of Emergent Priorities |
| Output type | Research poster |
| Funding | Indicator Stewardship Seed Fund: Formalising the Handover of a Neighbourhood Tool Library's Loan Ledger ($71,238) |
| Published | |
| DOI | 10.5555/slop.9jx27w |
| Pages | 1 |
| Version | 1.0 |
| Licence | CC BY 4.0 |
Cite as
@misc{slop_9jx27w,
author = {Anneke Tolan and Marek Solheim and Osei Vandermeer},
title = {A Plate Is Not a Visit: Classifying 24,318 Camera Entry-Exit Pairs at Six Shopping-Centre Car Parks Against a Paired Exit-Survey and Observation Record},
year = {2026},
publisher = {Slop University},
doi = {10.5555/slop.9jx27w},
url = {https://slop.university/outputs/slop-poster-the-three-hours-free-sign-9jx27w/},
version = {1.0},
note = {Research poster},
}