Full feature catalog

Theme A

Mapping & visualization

How the operational picture is drawn and driven.

• Interactive Leaflet map with switchable basemaps: standard, satellite, hybrid, terrain and relief.
• Geometry tools — circle, square, rectangle and measurement with live dimensions, plus map rotation.
• Time control — ±24 h slider, auto-play, UTC date and time, NOW button, real-time clock.
• Show/hide AOIs from the draw toolbar or the tools panel, with the state persisted server-side.
• Global view button to recenter on the default extent, and viewport persistence across reload, login and workspace switch.
• Coverage heatmap showing pass and coverage density directly on the map.
• Sun and terminator overlay — day/night boundary and solar angle.
• Swath corridors — sensor ground footprints drawn along the track.
• KPI bar — selected satellites, active AOIs, next opportunity, average score.
• Contextual detail panel per AOI, with tabs for details, opportunities and history.

Theme B

Satellite catalog & orbitography

501 satellites, 29 operators, real NORAD identifiers.

• Comprehensive catalog — 501 satellites across 29 operators, sourced from CelesTrak with real NORAD IDs.
• SGP4 propagation for real-time positions and orbit tracks.
• Automatic TLE updates once a day from CelesTrak; satellites it does not list (or all of them when it is unreachable) come from Space-Track in a single request, with the installation’s own account entered in Admin › Settings. Re-entered satellites are flagged.
• Filters by resolution (≤30 cm to >1.5 m), by operator, and by sensor type — optical (161), multispectral (127), SAR (85), AIS/maritime (70), RF/SIGINT (51) and hyperspectral (7), plus an IR/thermal filter across them.
• Pass computation — elevation, azimuth, duration and maximum elevation, computed reliably without external dependencies.
• Orbit worker — heavy propagation moved off the UI thread so rendering stays smooth.
• 3D Sim — a real 3D scene, not a spinning globe for show: live SGP4 orbit traces for up to 20 satellites at once, AOI footprints that light up on pass entry, driven by the same simulation clock as the 2D map.
• Operations view — a 2D satellite-watch screen for the selected fleet, one satellite in focus, showing only what is computed from public orbital elements and the Sun: position and country overflown, altitude, velocity, beta angle, node local time, eclipses with countdown, and ground-station contacts (AOS/LOS, maximum elevation, sky plot, 24-hour pass table) for 26 public reference stations — KSAT, SSC, ESA ESTRACK, NASA NEN, space agencies — plus the workspace’s own. Ground-track and stations/reception maps with the live link drawn during a contact; its own clock runs live, replays up to 1800× or jumps event to event.
• Operator comparison and quality scoring across the constellation.
• Real-time maritime AIS feed for vessel tracking alongside the satellite picture.

Theme C

Areas of interest

Everything about defining and managing operational areas.

• Creation — draw a polygon, rectangle or point, import GeoJSON, KML/KMZ or XLSX, or create manually.
• Full CRUD — edit, delete, rename, buffer, tag, prioritize, set status and request type.
• AOI Center — a centralized management and monitoring panel for the whole portfolio.
• Map interactions — hover for a mini toolbar, click for the detail panel, double-click to zoom, right-click for a quick order, with permanent interactive labels.
• Per-AOI history and GeoIdentify reverse-geocoding for automatic place naming.
• Undo and redo over a snapshot stack, with server-side save and reload of the AOI list.
• Export to GeoJSON, individually or in bulk.

Theme D

Opportunities & pass prediction

Turning orbital mechanics into a ranked shortlist.

• Per-AOI opportunities — passes of the selected satellites over seven days.
• Opportunity scoring combining maximum elevation, cloud cover, sun elevation and off-nadir angle.
• Opportunity cards grouped by day, sortable by score, by time or by satellite, with a cloud filter.
• Next Pass prediction — pick a point, get the next ten passes over seven days.
• Global opportunity windows reachable from the header.
• XLSX export of the opportunity set.
• Acquisition analytics and a conflict engine for competing windows.

Theme E

Tasking & orders

From wizard to delivered scene.

• Order wizard — multi-step, with a real-time cost estimator.
• Task planner in three views: timeline, kanban board and map.
• Tasking optimizer — scores candidate bands across every AOI and satellite and proposes a minimum-coverage plan, with weather conflicts flagged inline.
• Visual Mission Planner — drag and drop against a live budget.
• Global Gantt timeline across every AOI and satellite.
• Direct operator tasking — Planet, Capella, Maxar, Spire, Airbus OneAtlas and Satellogic.
• Weather-aware scheduling using Open-Meteo forecasts.
• Order placement and delivery — submission, download of deliverables, monitoring hook.
• Status tracking — provider order polling through the eight-step lifecycle.
• My Missions — a team-scoped mission library (Tentative through Archived lifecycle) with a 4-step feasibility wizard (real pass-finding and revisit-statistics engine, not a mockup) and one-click 3D replay of any saved mission, onboard-camera inset included.

Theme F

Providers & imagery search

Ten connectors behind one search box.

• Provider connectors — Airbus (Pléiades, SPOT), Maxar (WorldView), UP42, Copernicus/Sentinel, Planet Labs, Capella (SAR), Umbra (SAR, 0.16 m), ICEYE (SAR), BlackSky (optical, 0.35 m) and Satellogic.
• Unified multi-provider search returning STAC Feature 1.0.
• Archive browser across provider catalogs.
• Multi-source fusion — bundles combining several sources, a STAC basket and a My Bundles tab with refresh and cancel.
• Encrypted provider credentials (AES-256) per team, with a connection test.
• Provider health cache so a degraded connector is visible rather than mysterious.

Theme G

Image processing

What happens to the scene after it lands.

• Visual pipeline builder — assemble a DAG of processing operations in the browser.
• Processing interface for job submission, tracking and results.
• Asynchronous job execution proxied from PHP to the Python FastAPI backend.
• 30+ native Python operations — spectral indices, reprojection, subset, despeckle, mosaic, statistics, band extraction, contrast enhancement, image arithmetic, change detection, cloud mask, morphology, band math, pansharpening, K-means classification, watershed / Felzenszwalb / SLIC segmentation, SAR calibration, SAR speckle filter, connected components, orthorectification, atmospheric correction (DOS), supervised classification, index thresholding, COG and JP2 conversion.
• Orfeo Toolbox and ESA SNAP driven through their CLI and GPT interfaces.
• Processing layers rendered on the map, with a supervised classifier in the interface.
• Cloud-native processing — work directly from a COG URL.
• Super-resolution as a pipeline step.
• Roadmap: SAR Range-Doppler terrain correction.

Theme H

Artificial intelligence & ML

Inference that produces geometry, on your own hardware — object detection, change analysis and segmentation, not just processing operations.

• Object detection — ONNX models run with a sliding window, or classic blob detection. 15 classes out of the box (plane, ship, storage tank, harbor, helicopter, large/small vehicle, roundabout, bridge, sports fields and more). Output: heatmap raster plus GeoJSON, guaranteed in WGS84, with a confidence-threshold slider and per-class counts.
• Change detection — Change Vector Analysis L2 with Otsu thresholding, polygonized into a mask and GeoJSON polygons.
• Semantic segmentation — ONNX models or spectral K-means, producing a labelled raster and a GeoJSON legend.
• Detection display — results rendered as GeoJSON layers on the map.
• Natural-language query — plain English translated into search parameters, using a free hosted model or a paid one.
• Multimodal AI Vision — an LLM describes and answers questions about the scene.
• AI change monitor — change scored against a stored visual baseline.
• Model management — SatView ships with no detection engine baked in; your administrator plugs in an open-source model, or one under its own specific license, uploaded and managed through the Model Manager. SatView never bundles, sells or redistributes third-party AI models itself.
• Inference server — serves ML GeoJSON, lists available models and auto-indexes detections.

Theme I

Autonomous monitoring & alerts

The loop that runs without you.

• Monitoring rules evaluated automatically by a 15-minute cron.
• Rule types — cadence (periodic re-acquisition), retask_cloud (retry above a cloud threshold), retask_quality (retry below a quality floor), and ai_change (AI-detected change), and fire_detected (NASA FIRMS cross-check).
• Pass alerts when a satellite is about to pass over an AOI.
• Notifications — in-app, email and webhook, hardened against SSRF and spam.

Theme J

Content search & intel reports

Making findings findable — across every detection, every job, every acquisition.

• Content-based search over detected and indexed objects, with map and list display, filterable to dark-vessel and recurring-entity candidates.
• Auto-indexing — every detection job feeds the content index without a manual step.
• Intel reports — structured intelligence report generation.
• Professional PDF export — classification banner, cover page and tables.

Theme K

Collaboration & sharing

Because analysis is rarely a solo activity.

• Real-time collaboration over WebSocket, in shared workrooms.
• Collaboration panel — threaded comments with @mentions, team activity and reviews.
• Share links for reports and AOIs, including recipients without an account.
• Workspaces — personal and team, with strict data isolation.

Theme L

Security, classification & imagery restrictions

The controls that make the rest deployable.

• Embargoed zones — a maintained list of countries, each tagged with the regime it derives from (EAR, EMBARGO, ITAR), restricting imagery access by user and by recipient — share links included — with optional per-team rules and CSV export.
• Sensitivity classification — a seven-level ladder drawn from NATO, US and EU practice (UNCLASSIFIED, CUI, RESTRICTED, CONFIDENTIAL, SECRET, TOP SECRET, TS//SCI), with optional caveats (NOFORN, SI, TK) and an explicit REL TO releasability list, applied per workspace and shown as a persistent banner.
• Audit log covering every sensitive action.
• Rate limiting per user and per team.
• Encrypted credentials (AES-256) and secrets loaded from environment files — no hardcoded keys anywhere.
• Access rules enforced per role and per team.

Theme M

Administration & users

Running the platform day to day.

• Authentication — login, session heartbeat, single-session exclusion, bcrypt passwords, subscription expiry with grace period.
• Four role tiers — admin (global, every user and team), manager (scoped to the teams they lead: can add team members, can't touch other teams or create other managers), standard (team member, no management rights), and a write-restricted demo account.
• Admin panel — users, teams, satellites, settings, providers, access rules, audit, system health and embargoed zones.
• User and team management, including avatar upload.
• Configurable settings — AI provider, provider API keys, Python backend URL and secret, notifications.
• Billing and quotas — subscriptions, team quotas and usage events.
• Pipeline marketplace for sharing processing chains.

Theme N

API, SDK & interoperability

Driving SatView from outside.

• Public REST API v1 — aois, satellites, orders, search, monitoring, jobs, keys, openapi.
• STAC 1.0 endpoint consumable from QGIS, pystac and any STAC client.
• Python SDK — SatViewClient with resource modules for AOIs, satellites, orders, monitoring, jobs and keys.
• Developer portal shipped with the application: API docs, cURL examples and SDK usage.
• API keys managed per user.

Theme O

User experience

The details that decide whether people actually use it.

• Command palette — Ctrl + K reaches every command.
• Guided tour — an interactive walkthrough for new users.
• PWA and mobile — service worker with a versioned cache, offline support, install manifest, bottom navigation and a responsive mobile layout.
• Time Series Viewer — swipe comparison between acquisitions.
• InSAR viewer for surface deformation.
• In-app notifications and badges for active AOIs and pending items.
• Address and place search in the header.

Theme P

Intelligence & correlation

Cross-referencing detections, orbital data and tasking history that SatView already collects — plus a fused SO2/CH4 + wind atmospheric layer and a growing library of public data overlays.

• AIS correlation & dark-vessel fleet — ship detections matched against real AIS broadcast positions (3 km, ±30 min); a track seen at least twice with no AIS match at all is fused from AIS correlation and recurring-entity tracking into a dark-vessel fleet candidate.
• Recurring-entity tracking — vehicle, ship and aircraft detections clustered by position (150 m) across repeated passes.
• TLE maneuver watch — mean-motion and inclination jumps between TLE updates logged as maneuver candidates.
• Orbital conjunctions (SOCRATES) — CelesTrak close-approach screening against the full catalog, filtered to tracked satellites.
• Reverse tasking — which satellites, across the full 500+ catalog, pass over a given AOI next.
• Archive-vs-new check — searches the last 30 days of archive before a fresh tasking order goes out.
• Atmospheric fusion — Sentinel-5P TROPOMI SO2 or CH4 (methane) fused with NOAA GFS 10 m wind vectors on one map layer, with a live color-scale legend.
• Public data layers — CAMS air quality (NO2), ESA WorldCover land cover, live GDACS disaster alerts, GDELT geocoded news events and VIIRS night-lights change, toggled from the same Intelligence Layers panel.
• Decomposed confidence — detection score, AIS-correlation tier and recurrence count shown separately, never collapsed into one invented percentage.

Theme Q

Explosive Safety

Five quantity-distance and risk modes, all drawn to scale on real satellite imagery of the site rather than an abstract CAD exercise — see the dedicated Explosive Safety page for the full walkthrough.

• Risk — a real probabilistic QRA on DDESB TP-14's own methodology (Pf = Pe × Pf|e × Ep): event probability, overpressure/impulse, primary & secondary fragment debris, building collapse, plus a DESR 6055.09 Hazardous Fragment Distance reference circle for low-angle direct debris that TP-14 itself doesn't cover.
• Thermal — the separate burning-case scenario TP-14 treats on its own: fireball radius, radiative temperature and a fatality probability per event, never mixed into Risk's annualized criteria.
• Design — the full IATG 02.20 Hazard Division 1.1 matrix: every building is both a Potential Explosion Site and an Exposed Site toward every other, resolved through the real UN SaferGuard PES×ES table, with AN/BA and ATD reference types (27 CFR 555) checked automatically where they apply.
• Drawing — non-standard buildings outside Design's construction profiles: a saved per-wall k-factor template, plus France's own concentric Z1-Z5 flat-zone model as a genuinely separate point type.
• Localization — a fast four-tier k-factor envelope (VBD/IBD/PBD-APD/IMD) for a quick first pass, alongside the full matrix tools.
• Branded PDF reports — every mode exports a map snapshot plus full input/result tables and an explicit scope & limitations section.
• Blast exposure & protection — for a threat you do not control: Kingery-Bulmash overpressure zones with indoor fatality probabilities (CIA 1998) and an ESA WorldCover built-up check, UFC 4-010-01 levels of protection for your buildings, domino scenarios on a struck site (HD 1.1 added, EPA vapor cloud explosion and BLEVE compared), and your candidate shelter sites ranked by exposure — see the walkthrough.

Theme R

Launch Site Explorer

Launch-azimuth feasibility from exact spherical trigonometry, plus a structural safety screen along the resolved corridor — see the dedicated Launch Site Explorer page for the full walkthrough.

• Feasibility — cos(i) = cos(lat) · sin(Az) solved for both azimuth roots within the site's permitted corridor; LEO against a target inclination, SSO derived from altitude via first-order J2 nodal precession, or GEO's near-equatorial case, with a plane-change delta-v penalty reported when nothing in the corridor reaches it directly.
• Shipping lanes & air corridors — the resolved ray tested against global Major/Middle/Minor shipping-lane center-lines and live OpenAIP CTR/TMA/ATZ/Airway airspace, a structural "does this corridor cross a known route" check, not a live snapshot.
• Built-up land — the ray sampled every 10 km against ESA WorldCover's own published land-cover classification.
• Port & anchorage proximity — a self-hosted PostGIS store (NGA World Port Index, Global Fishing Watch anchorages, Marine Regions territorial waters) resolves the nearest port or anchorage, closing the gap a shipping-lane dataset alone leaves at port-level approach distances.
• Recent port & traffic activity — IMF PortWatch port calls/import/export (keyless, public) and Global Fishing Watch 4Wings vessel-presence density (all vessel types, your own account), both enriching the static screen with recent history.
• Custom launcher library — LEO/SSO/GTO payload figures for built-in and user-added launchers, referenced against every site's delta-v penalty.
• Branded PDF reports — a map snapshot plus every site's full feasibility and safety-screen detail, listing every shipping lane or nearby port individually, not just the worst case.
• Launch Site Study — several candidate sites compared on the same mission, vehicle and rules: ascent performance by trajectory optimisation (3-D point mass, J2–J4, US 1976 atmosphere; gravity turn, optimised guidance or optimal control by SQP in the POST formulation; direct or parking-orbit multi-burn missions with finite burns and upper-stage deorbit; upper-air winds and q·α; Monte-Carlo dispersions; validated within 0.11 % on the published multiple-stage ascent benchmark — see Launch Performance), drop zones of every stage with distance to land, logistics, weather availability, population, seismic hazard and your own environmental layers, ending in an illustrated ~20-page PDF study with a traceable, provisional recommendation (equivalence tolerances, Pareto dominance, sensitivity) — no hidden weighted score.

Theme S

Natural Hazards & Climate

Earthquake, tsunami and volcano exposure for every candidate site from published hazard maps and public historical catalogues, ranked side by side, plus month-by-month site climatology — see the dedicated Natural Hazards & Climate page for the full walkthrough.

• Earthquake — GSHAP 475-year peak ground acceleration and hazard class, every USGS catalogue event since 1900 within up to 300 km with the intensity it produced at the site (USGS ShakeMap for the strongest, otherwise Allen, Wald & Worden 2012 with ±1σ), distance rings and labelled epicentres on the map.
• Tsunami — site elevation and distance to the sea from Copernicus GLO-90, compared with the highest reliable wave in NOAA’s Global Historical Tsunami Database within 25/50/100 km (4 m margin for the elevation model’s accuracy) and the farthest inland reach recorded; doubtful and non-tsunami observations listed but never used as the reference.
• Volcano — volcanoes within 10, 30 and 100 km rings (Brown et al. 2015; Biass, Jenkins et al. 2024) from NOAA’s database compiled from the Smithsonian Global Volcanism Program, each with its last eruption, significant eruptions, maximum VEI and recorded deaths by cause.
• Comparison — every site ranked by its worst hazard level, then the next, from published thresholds only — no invented weighted score, and missing data shown as unknown, never as low.
• Site climatology — ERA5 reanalysis reduced to a monthly table: wind and gusts, monthly precipitation total and wet days, cloud cover, wave height and period for coastal sites, an annual row and the extreme months.
• Reports — branded PDF exports for both windows: map, comparison table, full per-site detail, every source and every limitation.

Theme T

Launch Performance

Maximum payload of a launcher from any site by trajectory optimisation, validated on published benchmarks — see the dedicated Launch Performance page for the full walkthrough.

• Vehicle model — 3-D point mass on a rotating Earth, J2–J4 gravity, US 1976 atmosphere, Mach-dependent drag, strap-on boosters with tabulated thrust, fairing jettison on heat flux, max-Q / max-g throttling, ignition delays, burn limits and last-stage reserve; circular, elliptical and GTO targets.
• Three guidance modes — gravity turn, optimised pitch and yaw offsets, and direct optimal control by SQP in the discrete-parameter formulation of NASA's POST, the payload maximised under the injection constraints; the better flown result is kept and the row says which.
• Multi-burn missions — direct ascent or parking orbit then restart (coast to the equatorial crossing, to perigee or for a set time), circularisation, ignition and coast limits, optional upper-stage deorbit taken off the payload; optimal control with the later burns flown as finite burns.
• Winds and loads — each site's upper-air wind on 12 pressure levels (annual, monthly or 95th-percentile), incidence, q·α reported and limited, jettisoned bodies falling in the same wind.
• Dispersions and safety — Monte-Carlo success rate, orbit statistics and impact ellipses per jettisoned body; expected casualties against the FAA 14 CFR 450.101 reference; drops at sea enforced by steering the trajectory.
• Validation — published multiple-stage ascent benchmark (Delta III class) within 0.11 %, Ariane 5G and Ariane 5 ECB, Europa III Δv budget, recomputed live in the app.