10.1.2 — Railway Intelligence — maturity: live
Encroachment Detection
Using repeat-pass satellite optical and SAR imagery to identify unauthorised structures, vegetation, and human activity encroaching on railway rights-of-way.
Satellite-derived radar and optical imagery catches illegal structures, encroaching settlements, and trackside obstructions before they become derailments — giving rail operators persistent, politically neutral eyes on every kilometre of corridor.
Railway authorities in most nations maintain legal corridors — typically 15 to 30 metres either side of the centreline — within which construction, cultivation, and permanent habitation are prohibited. In practice, enforcement is patchy. Ground patrols cover thousands of kilometres infrequently, and by the time an illegal structure is reported it is often occupied and politically difficult to remove. The encroachment backlog on major national rail networks in South Asia and Sub-Saharan Africa runs into tens of thousands of incidents, each one a latent derailment or collision risk.
Satellite constellations resolve this by providing a persistent, objective record of the right-of-way at regular intervals. Change-detection algorithms — applied to co-registered optical or SAR image pairs — flag new objects above a configurable area threshold (typically 10–25 m²) within 24 to 72 hours of acquisition. SAR provides cloud-independent coverage; optical adds the spectral richness needed to distinguish a concrete foundation from a tarpaulin shelter. Together they allow enforcement teams to be dispatched before concrete is poured and political facts are created on the ground.
The operational outcome is a shift from reactive litigation to proactive prevention. Authorities receive a geo-tagged alert with before-and-after imagery attached, evidence that is admissible in most national land-tribunal systems. Recidivism drops when communities understand that the corridor is continuously monitored. Insurance premiums and accident-liability exposure both fall. Nations that have piloted commercial feeds for this task quickly discover that the data is delivered on the vendor's schedule, in the vendor's format, and disappears when the contract lapses — taking the historical baseline with it.
Frequently asked
What satellite technologies are most effective for detecting rail corridor encroachments?
Synthetic Aperture Radar (SAR) is the workhorse: it penetrates cloud, works day and night, and produces coherent-change-detection products that highlight even subtle surface disturbances. Sub-metre optical imagery from constellations such as Planet SkySat or BlackSky provides the visual confirmation an analyst or court needs. For a sovereign programme, combining a small SAR constellation with a tasked optical capability delivers both all-weather detection and high-resolution evidence.
How quickly can a satellite-based system detect a new encroachment after it appears?
Detection latency depends on constellation size and revisit rate. A commercial SAR pair (e.g. ICEYE + Capella) might deliver a 6–12 h revisit over priority corridors. A sovereign eight-satellite SAR constellation in sun-synchronous LEO can approach 3–4 h. Alert generation — from downlink to flagged geometry in a GIS dashboard — typically adds another 30–90 minutes depending on ground-segment automation.
Can satellite imagery be used as legal evidence in rail right-of-way disputes?
In a growing number of jurisdictions, yes — provided the imagery is accompanied by ISO 19115-compliant metadata, a calibrated accuracy statement, and a certified chain of custody. India's ISRO-led BHUVAN-Rail pilot demonstrated a 63% reduction in litigation time when satellite-derived encroachment maps were admitted alongside cadastral records. Nations should legislate admissibility standards early so that evidence gathered by their own sovereign system has full legal weight.
Why should a government own this capability rather than subscribe to a commercial encroachment monitoring service?
Commercial services are available today from Planet, ICEYE, Capella, and others, but access can be curtailed by export controls, pricing negotiations, or vendor business decisions. A rail network is critical national infrastructure; uninterrupted, politically independent surveillance is a security requirement, not a procurement preference. A sovereign constellation also lets the government retask assets instantly for emergency response, border surveillance, or disaster assessment — multiplying the return on investment across agencies.
What ground-segment infrastructure is needed to turn raw satellite data into actionable encroachment alerts?
You need at minimum: a national direct-reception ground station (S/X-band for LEO), a SAR processing cluster (radiometric calibration, terrain-correction, coregistration), an automated change-detection pipeline (coherent change detection or deep-learning object recognition), a vector database of authorised right-of-way geometry, and a GIS dashboard with alert routing to railway inspectors. Cloud-hosted processing is feasible early on, but a sovereign compute layer avoids dependence on foreign cloud providers for sensitive infrastructure data.
How does this capability interact with on-ground CCTV or trackside sensor networks?
They are complementary, not competitive. CCTV and IoT sensors give continuous, real-time point surveillance but cover only instrumented locations and degrade with power outages or vandalism. Satellite SAR provides autonomous, periodic wide-area coverage of every metre of track with no ground hardware in the field. The optimal architecture fuses both: satellites detect and locate anomalies at network scale; CCTV and patrol teams provide real-time eyes at flagged locations.
What accuracy can we realistically expect from change-detection algorithms today?
State-of-practice coherent change detection on high-resolution SAR (0.5–3 m) achieves ~88–94% detection probability for footprint changes ≥4 m² under test conditions, with false-positive rates of 10–20% in complex peri-urban environments. Deep-learning classifiers trained on domain-specific data can push false positives below 8%, but they require substantial labelled training datasets from the target corridors — another reason a sovereign programme that accumulates its own historical imagery archive holds a long-term advantage.
Which orbits and satellite sizes are appropriate for a sovereign encroachment-detection programme?
Sun-synchronous LEO (520–580 km altitude) is standard for SAR and high-resolution optical — it maximises surface coverage, keeps revisit intervals short, and the orbit is well-understood from an operations standpoint. Microsatellite buses in the 100–500 kg class (e.g. based on ICEYE or similar heritage) are the cost-effective choice; a constellation of 8–12 such satellites can be procured and launched for less than the cost of a single traditional large SAR satellite while delivering superior revisit. Nanosatellites (CubeSats) are too resolution-limited for court-admissible encroachment evidence at present.