Edgegenix Edge AI

Vision AI at the edge.
No cloud required.

Edgegenix Edge AI runs sub-40ms vision models directly on the asset — smoke, wildfire, object detection, and thermal fire mapping — fully offline. Detects. Decides. Reports when connected.

Request a live demoSee how it works ↓
Pipeline

Five stages. All on-device.

From raw sensor input to verified finding — every step runs locally on the Edgegenix AI Box.

01 Capture

RGB, thermal, and depth frames from camera or drone feed at up to 60fps

02 Preprocess

Hardware-accelerated normalise, resize, and channel fusion before inference

03 Infer

YOLO-class models run on-device NPU — sub-40ms end-to-end latency

04 Detect

Confidence scoring, class labelling, bounding box generation, and deduplication

05 Sync

Verified findings timestamped and forwarded to Cloud AI when network available

Use Cases

Four mission-critical scenarios.

Edge AI is purpose-built for environments where cloud dependency is not an option.

Wildfire & Smoke Detection

Detects smoke plumes and fire ignition from fixed towers, vehicles, and drones. On-device fusion with weather and fuel load data. First alerts before a 000 call is made.

SMOKE_V3 · 82% mAP

Object Detection

Classifies vehicles, people, wildlife, and equipment with >95% accuracy. Computes collision and intrusion paths entirely on-device for safety and access control.

4 CLASSES · >95% mAP

Traffic Sign Detection

Reads and classifies road signs, speed limits, and regulatory markings in real time. Deployed in autonomous vehicle trials and remote mine site haul roads.

SIGN_V2 · 94% mAP

Thermal Fire & Escape Route

Dual RGB+thermal input maps active fire zones, hotspot progression, and safe evacuation corridors — displayed in real time on truck-mounted screens without network.

THERMAL · 38ms LATENCY
Use Case 01

Wildfire & smoke detection.

SMOKE_V3 runs a YOLO-class detector fused with wind speed, relative humidity, and fuel dryness index. Every detection includes spread-risk scoring, not just a bounding box.

01RGB+NIR

Camera or drone frame at up to 4K

02Fuse

Wind, humidity, fuel dryness co-read from onboard sensors

03SMOKE_V3

On-device YOLO detection model — 38ms

04Risk Score

Spread probability and confidence score computed locally

05Alert

Finding dispatched to Cloud AI and emergency contacts

38msLatency
82%mAP Score
<2%False Alarm
100mMin Range
OfflineMode

The on-device risk engine fuses real-time camera output with localised weather data pulled from onboard sensors — eliminating false positives from dust, steam, and haze without a network call. When spread risk exceeds threshold, the finding is written to local storage with full timestamp, GPS, and environmental context, then forwarded to Cloud AI when connectivity returns.

Deployed across tower networks in NSW and WA, the system has triggered verified alerts up to 47 minutes before the first 000 call in field trials.

Detection Event

SMOKE PLUME — VERIFIED

Primary ignition detected. Wind ENE 28km/h. Fuel dryness index 4.1. Spread probability HIGH.

Confidence91%
Spread RiskHIGH
WindENE 28km/h
Fuel DI4.1 / 5.0
Latency38ms
ConnectivityOFFLINE
Use Case 02

Object detection & classification.

OBJ_V4 classifies four primary object classes with >95% mean average precision at 60fps. Intrusion path and collision risk are computed on-device — no cloud lookup required.

01RGB Frame

Wide-angle camera at full resolution

02Preprocess

Crop, normalise, channel stack via NPU

03OBJ_V4

Multi-class YOLO on NPU — 22ms

04Path Compute

Intrusion and collision vector on-device

05Alert / Log

Triggered alert + timestamped finding stored

Vehicle96.4%Cars, trucks, heavy machinery
Person95.1%Pedestrians, workers, intruders
Wildlife94.7%Cattle, kangaroo, feral animals
Equipment95.8%Plant, PPE, hazardous items
Use Case 03

Traffic sign detection.

SIGN_V2 reads and classifies regulatory, warning, and advisory road signage in real time — deployed in autonomous vehicle trials and remote mine site haul roads at >100km/h approach speed.

01Front Camera

Forward-facing wide-angle RGB input

02SIGN_V2

Sign classifier on NPU — 28ms

03OCR Layer

Speed value and text extraction

04Vehicle CAN

Speed limit fed to vehicle control bus

Speed Limit Signs

Reads numeric speed values from standard and non-standard sign formats including zone markers

Warning Signs

Detects sharp curves, intersection ahead, animal crossing, and school zone markers

Regulatory Signs

Stop, give way, no entry, one-way — all classified with occlusion tolerance up to 40%

Mine Site Signage

Custom sign classes for haul road intersections, blast zones, and exclusion areas

Lane & Road Markings

Classifies lane types, chevrons, and painted road instructions alongside signage

Night & Low Visibility

Operates under headlight illumination with IR-enhanced input for 24-hour coverage

Use Case 04

Thermal fire mapping &
escape route generation.

Dual RGB+thermal cameras on the truck feed THERM_V2, which maps active fire perimeters, hotspot progression, and dynamically computes safe evacuation corridors — all without connectivity.

Fire Perimeter Alert

ACTIVE FIRE — THERM_V2

Two active hotspots detected. Perimeter expanding NE at approx 12m/min. Escape route generated: NW exit via Track C.

Hotspot 1432°C
Hotspot 2318°C
Expand Rate12m/min NE
Escape RouteNW / Track C
Latency38ms
ConnectivityOFFLINE
Hardware

Truck-Mounted Setup

Dual RGB+FLIR thermal array feeds the Edge AI Box mounted in cab — no external power, runs on vehicle 12V.

ThermalFLIR Boson+
RGB1080p 60fps
ComputeAI Box Gen2
Power12V / 35W
Platform

Part of the complete Edgegenix intelligence stack.

See It In Action

Edge AI detecting smoke before the first call is made.

Request a live demonstration on your hardware or environment — wildfire, thermal, object detection, or custom use case.

Request a live demo