Gray spaces — also called urban voids, leftover spaces or spaces of uncertainty — are neglected, underutilized or invisible areas of the urban fabric. They present unexploited opportunities for development, but pose risks (blight, insecurity, devaluation) if not strategically identified.
This tool implements the detection workflow proposed in the ASCAAD 2024 paper “Decoding Gray Urban Spaces”:
- Zone of study — you define the boundary of the urban area to examine.
- Spatial model — the street network (road-centre lines) and building footprints are retrieved from OpenStreetMap and converted into a segment map.
- Space syntax analysis — angular segment analysis computes, for every street segment:
- Choice (through-movement potential, normalized as NACH = log(CH+1)/log(TD+3)),
- Integration (to-movement potential, normalized as NAIN = NC1.2/(TD+2)),
at a pedestrian radius (400 m) and a vehicular radius (1200 m), matching the paper's case study of Al Darb El Ahmar, Cairo. The engine reproduces depthmapX's Tulip analysis (T1024) exactly: polylines are exploded into straight segments as in depthmapX's map conversion; turn costs are discretized into 1024 angular bins; metric radii are measured centre-to-centre along the path; node count, total depth, integration (NC²/TD) and choice follow segmtulip.cpp semantics, including bidirectional pair counting for choice. The implementation is validated by an automated test suite against hand-computed ground truths. An optional isovist-based visibility grid approximates the paper's Visual Graph Analysis at a human scale.
- Detection — segments in the low tail of both choice and integration (and optionally visibility) are flagged and clustered into gray-space candidates.
- Decoding matrix — each candidate is described through spatial characteristics (size, contextual location, physical condition, accessibility) and its surrounding context, producing intervention guidance per the paper's Table 2. Field validation remains essential: in the pilot study, space syntax alone confirmed ~77% of gray spaces; the rest emerged from contextual factors observed on site.
Citation
Maged, M., Ismail, A., Elsayed, E., & Mohareb, N. (2024). Decoding Gray Urban Spaces: A space syntax approach to detecting and identifying underutilized urban areas. Proceedings of ASCAAD 2024, 411–425.
depthmapX development team (2020). depthmapX (Version 0.8.0) [Computer software]. github.com/SpaceGroupUCL/depthmapX
Notes & limits
- Analysis runs entirely in your browser; no data leaves your device except the OpenStreetMap query.
- Results depend on OpenStreetMap completeness. In dense historic fabrics, informal alleys and interstitial spaces may be unmapped — exactly where gray spaces hide. Load your own surveyed line map (GeoJSON) for best fidelity.
- Space syntax does not capture land use, aesthetics or socio-economics; the decoding matrix and field visits bridge that gap.