01
AI detection & inference
Multi-model weighted-box-fusion ensemble — open-vocabulary, grounding, thermal, C-UAS, pose and behaviour — with motion gating and a global tile budget.
The core loop
Detect, alert, decide, dispatch, operate — the entire loop, on hardware you own.
↻ Closed loop · operator retains acknowledge, launch and abort at every stage
Capability inventory · 11 domains
Every domain shipped, not roadmapped.
Detection, sensors, drones, mapping, recording and access control are one system with one audit log — because in an accreditation review, five vendors is five arguments.
02
Camera management & ONVIF
One-click discovery, credentials encrypted at rest, hot-add a camera in about 60 seconds with no restart, per-camera GPS and day/night handling.
03
Sensor registry
Fence, radar, seismic, thermal and acoustic devices over MQTT or signed webhook, on a unified alert path with fail-closed security.
04
Alerts & real-time
Three-level grading, priority queue, auto-pin of alerting feeds, graded browser buzzer and 50 ms WebSocket batching.
05
Drone fleet C2
Four-drone command centre with manual and sector-aware auto-dispatch, a smart state machine, battery and loiter policy, live telemetry.
06
Maps & geospatial
Multi-layer offline basemap, marker clustering, three-state camera markers, configurable boundary overlay and GPS pickers.
07
Recording & playback
Per-camera segmented recording, hierarchical browser, HTTP-Range streaming, variable speed and retention pruning.
08
Security, RBAC & MFA
Three enforced roles, JWT authentication, tamper-resistant audit log with CSV export, dormant TOTP MFA and request hardening.
09
Health & monitoring
Host CPU, RAM, disk and GPU metrics, per-camera circuit breaker, engine liveness heartbeat, tamper and blind-spot detection.
10
QRT interface
PIN-gated tablet UI with compact alert cards, read-only live map, team GPS heartbeat and nearest-drone dispatch.
11
Live streaming
Low-latency WebRTC primary with HLS fallback and MJPEG for drone feeds; scroll-zoom, mode toggle and recording controls.
One platform, not five integrations
Detection, sensors, drones, mapping, recording and access control are one system with one audit log — because in an accreditation review, five vendors is five arguments.
Engineering detail
Sized, measured and deployable on hardware you already own.
Numbers from the running system, not a benchmark sheet.
At a glance
Scale and response
80 configured cameras on a single GPU node, with roughly 15–20 concurrently active 1440p streams — CPU-bound, not GPU-bound. Up to four autonomous drones over MAVLink and ArduPilot.
- Store · PostgreSQL 16 with SQLite as native development fallback
- Stack · Python / FastAPI · React / Vite · Ultralytics YOLO
AI model stack
Eight models, fused
Weighted Box Fusion across a primary detector, a transformer detector, open-vocabulary and grounding models, thermal and sliced-tile inference, plus a counter-UAS tracker. Approximately 28 GB VRAM for the full stack.
Security & deployment
Built for the accreditation review
Three-role RBAC, JWT authentication, a tamper-resistant audit log, dormant TOTP MFA and encrypted camera credentials — the controls an assessor asks for, present before they ask.
- Deploy · Native · Docker (CPU and GPU) · Windows via WSL2 · offline and air-gapped · systemd service
Weighted Box Fusion across eight models
≈28 GB VRAM for the full stackPrimaryTransformerOpen-vocabGroundingThermal / IRSliced-tileC-UASPose
Motion gating and a global tile budget keep the ensemble inside the compute envelope — which is why 80 configured cameras run on a single node without the GPU becoming the bottleneck.
Measured, in the field
What the detection stack actually returns.
Figures from the classified perimeter programme this engine was proven on. The client cannot be named; the numbers can.
97.8%
Detection accuracy
ARGUS, air-gapped
0.18%
False positive rate
Below the alert-fatigue floor
100%
Offline capable
Tested with the NIC down
94.3%
Acoustic accuracy at −5 dB
SENTINEL, sub-50 ms
Deployment targets