Puddle · Live overlay★ Flagship Live Demo
Liquid Leak Detection
Puddle, spray & visible-vapor detection
Tank battery & dike floorWellhead & stuffing boxHeaders, separators & SWDGathering lines & manifolds
AI Vision Analytics
Industrial camera analytics that run at the edge
VisionAery turns the cameras already on your sites into real-time industrial camera analytics — liquid leak detection, fire & smoke detection, and flare monitoring — running on edge GPU hardware at the site and feeding your existing VMS and SCADA through the same alarming paths as third-party analytics.
A Twin Eagle Solutions company — backed by 25+ years of industrial field, network, and camera deployments.
01 · VisionAery Analytics
Edge AI built for industry
Computer Vision Analytics
Real-time detection trained on real industrial imagery — running at the edge.
Puddle · Live overlay
Smoke · Live overlay Live Demo
Fire & Smoke Detection
Visible flame, glow & smoke plume detection
Pad equipment & enginesCompressor stationsTank battery & dikeProcessing skids & yards
Flare · Live overlay Live Demo
Flare Monitoring
Ignition, presence & smoke verification
Flare stack ignition statePilot-flame presenceVisible smoke / combustion qualityTip integrity & wind-out events
Alarm Verification
The reasoning layer that separates real releases from nuisance alarms before they reach an operator.
_1781316097898-CGTh9HRL.png)
95%+ LLD AccuracyAVA LLD — Alarm Verification Agent for Liquid Leak Detection
Reasoning that tells a real leak from a rain puddle
AVA LLD is the Liquid Leak Detection build of AVA, our Alarm Verification Agent. It reasons about every alarm the analytic raises and decides whether it's a true release, a false alarm, or a nuisance — the rain puddles and old stains that look just like the real thing. Deployed correctly, it raises Liquid Leak Detection accuracy to 95%+ in alarm detections, so operators stop chasing noise and real leaks stop getting buried. AVA is purpose-built for Liquid Leak Detection today, with more analytics on the roadmap.
True-positive vs. nuisanceReasoning over context95%+ LLD accuracyOn-prem · edge-fit
02 · Aery Device Manager
Your fleet. Your console.
Your rules.
Edge AI is only useful if you can actually run it across a fleet. VisionAery ships a management plane that pushes model and config updates to one site, one region, or every device at once — staged, versioned, and rollback-ready. The whole control plane runs on hardware you own, behind your firewall.
Online · Reporting
Edge Hardware
Compute
GPU + CPU
Cooling
Fanless
Network
Dual-NIC
visionaery / fleet
On-Prem
Devices
248
online
Sites
63
regions: 4
Uptime
99.97%
30d
Live Fleet
Eagle Ford · Pad 14-Av2.4.1synced 2m
Permian · Tank Battery 7v2.4.1synced 3m
DJ Basin · Compressor 3v2.4.0update pending
Bakken · SWD-22v2.4.1synced 1m
Haynesville · Pad 9v2.4.1synced 4m
Illustrative · VisionAery Fleet ConsoleAll regions healthy
01
Push models & configs fleet-wide
02
Enable & tune analytics per camera
03
Runs on hardware you own
04
Live device health & OTA updates
03 · Analytic Data Integration
In-house experts wire analytics into your VMS and SCADA
Every VisionAery analytic pushes alarms, overlays, and event clips into the video-management and control systems you already operate — keep the same operator workflows, user permissions, and retention policies you have today.
VMS Support
Alarms, overlays & searchable bookmarks
SCADA Support
Telemetry & alarm payloads delivered natively over MQTT
The VisionAery app speaks MQTT natively for telemetry and alarm payload data. For older systems that don't yet accept MQTT, we ship a prebuilt MQTT → Modbus TCP converter so those payloads land in your existing control stack without a rebuild.
04 · Site Survey & System Design — visualized
Site design matters as much as the analytic itself
The best analytic in the world can't fix a camera pointed at the wrong thing. Every VisionAery deployment starts with an aerial site survey and an FOV-overlay design — each colored cone is a planned camera with the lens, height, and aim required to cover what the analytics need to see, with no blind spots and no wasted pixels.
FOV Plan
Coverage MapVisionAery field crew validating an AI deployment on a live site.
Field-testedSchedule a POC
Prove it on your sites — before you commit
A VisionAery field crew comes to a representative pad of yours, stands up the analytics on existing or temporary cameras, and runs them against your real scenes — your equipment, your lighting, your weather, your operating cadence. You see the detections and the alarm path end-to-end before any commitment.
- On-pad camera survey and AOI selection by a VisionAery engineer
- Live VisionAery inference on your actual scenes
- VMS bookmarks + SCADA alarm preview wired into your existing systems
- POC report with detection samples, false-positive notes, and a deployment plan
Frequently asked questions about VisionAery
See how VisionAery performs on your sites
Try the live Liquid Leak Detection, Fire & Smoke, and Flare Monitoring demos against your own footage — or talk through a full deployment with our team.