The Proof

Every number on this page comes from verified testing and real deployments — not projections. Every source is attributed. Every layer is identified.

See Layer
500%
Detection improvement
ProHawk AI validated testing
Think Layer
2,000→10
False alarms / month
Vaidio — Intl. media company
See Layer
$7M
Hardware upgrades avoided
Major Hawaii utility · 800 cameras
See Layer
20×
Detection range extended
ProHawk AI performance data
Dell AI Innovation Lab & NVIDIA Metropolis Lab

Independent Lab Testing:
What Happens When AI Can't See

Dell AI Innovation Lab and NVIDIA Metropolis Lab tested standard AI analytics (Ipsotek VISuite) under simulated near-blackout conditions. The question: what happens to detection accuracy when the camera feed degrades? The answer changed the conversation.

Vehicle Detection — Near-Blackout Conditions

Tier 1 Verified
Vehicle ClassBaseline (Ideal)Degraded (No Restoration)After ProHawk Restoration
Cars492 detected (100%)324 detected (65%)452 detected (91%) ↑
BusesMost dramatic recovery2 detected (100%)0 detected (0%)4 detected (200%) ↑↑Exceeded original baseline
Trucks14 detected (100%)0 detected (0%)1 detected (7%)
Vans4 detected (100%)0 detected (0%)2 detected (50%) ↑
The critical finding: Without restoration, buses, trucks, and vans were completely invisible to AI — 0% detection across all three classes. After ProHawk restoration, every vehicle class became detectable again, with buses exceeding the original baseline.
300%
Improved object detection & tracking
Dell / NVIDIA Validation Labs
30×
Faster video stream processing
Dell / NVIDIA Validation Labs
3–4×
Clarity improvement in degraded images
All four patents combined
Four U.S. Patents

The Technology Behind Restoration —
In Plain Language

ProHawk AI is protected by four U.S. patents. Each solves a different visibility problem. Here's what they actually do — no technical jargon required.

US 8,724,196 B2

Recovering What Fog Hides

Your camera recorded a foggy parking lot. The cars, people, and license plates are in there — the fog just scattered the light before it reached you. This patent reverses the scattering and recovers the detail the sensor actually captured.

91% car detection in near-blackout
40% fewer false positives
US 8,780,235

Fixing Video Live — Not After the Fact

This isn't a batch process you run on yesterday's footage. It restores every frame as it streams — at 2 milliseconds per frame. The latency is imperceptible to the human eye. Your security team sees clean, restored video in real time.

300% object detection improvement
30× faster processing
US 9,082,163

Seeing Both Sides of the Loading Dock

Loading dock at noon: half in blinding sunlight, half in deep shadow. This technology balances both halves simultaneously. You see the person in the shadow AND the truck in the sunlight, in the same frame, at the same time.

3–4× clarity improvement
50% better facial recognition in low light
US 9,129,406 B2

Following Someone From Dark to Light

A camera at a building entrance sees bright sky and a dark doorway. Without this patent, you get one or the other. With it, you get both. A person walking from a dark hallway into sunlight stays visible the entire time — no blinding flash, no disappearing into shadow.

30% better anomaly detection
35% medical diagnostic accuracy
Real-World Deployments

Real-World Deployments You Can Verify

Two deployments, two completely different conditions, same restoration technology. These are not demonstrations or marketing claims — both are publicly documented with on-record confirmation from the operators involved.

Connecticut Division of Criminal Justice

Restored Evidence in Active Criminal Investigations

The Office of Inspector General uses ProHawk AI to restore degraded dashcam, body camera, and surveillance footage in legal forensics. Two documented cases, two outcomes.

Use-of-Force Investigation

An officer's altercation with a suspect who died four days later. ProHawk restored dashcam and body camera footage to reveal hidden details and reconstruct the sequence of events.

Outcome: Officer exonerated

AR-15 Shooting at Officer

A perpetrator fired three shots through an officer's windshield. ProHawk restored dashcam footage to identify distinguishing features and establish shot trajectory.

Outcome: Perpetrator identified and apprehended
Daubert standard compliant — admissible as evidence in U.S. federal courts. ProHawk restoration can also recover detail through tinted windows.
History Channel · The Curse of Oak Island

Restoration Through Subsea Turbidity, Live on Television

Season 10 premiere — the production team lowered an underwater camera 105 feet down a borehole into a flooded tunnel. The original footage was nearly opaque from sediment and darkness. ProHawk AI restored it to inspection-grade clarity.

The Conditions

Underwater void at 105ft depth, near-total darkness, dark turbid murky water, color distortion. Standard underwater camera, no specialized optics.

Outcome: Murky video transformed to high-resolution clarity

The Discovery

Restored footage revealed broken timbers, vertical wooden dowel fasteners, and a structural void — first-ever clear view of an underground structure in the show's history.

Outcome: New evidence in a 227-year-old mystery
A working television production used ProHawk to solve a real visibility problem on camera. The restoration was performed off-site and the result aired in the season premiere.
Proof by Industry

Results Across Verticals

Verified deployment data from six verticals where Partner One has documented results. The underlying restoration and analytics capabilities apply across every industry that operates cameras in degraded conditions.

Energy & Utilities

$7M
Camera upgrades avoided across 800 cameras See
$2M
IR illuminator savings See
Thermal camera range extended See
Infrared camera range extended See

Security & Public Safety

2,000→10
Monthly false alarms cut Think
40%
Reduction in AI false positives See
2 of 2
Connecticut DCJ cases — officer exonerated, perpetrator apprehended See

Manufacturing

$1M/yr
Operational savings, 12+ use cases Think
300%
AI object detection improvement See
3–4×
Clarity in dusty/steamy conditions See

Healthcare

ProHawk restoration applies to surgical video clarity in degraded lighting, medical imaging enhancement, and patient monitoring environments where visual input quality affects diagnostic and operational reliability.

Deployment proofs in active development.

Transportation

91%
Car detection in near-blackout (vs 65%) See
20%
Vehicle classification accuracy boost See
30×
Faster video processing See

Robotics & Automation

ProHawk's real-time restoration and high-speed processing apply to autonomous navigation and robotic vision systems where degraded visual input creates operational risk — dust, steam, low light, and rapidly changing conditions.

Deployment proofs in active development.

Why Both Layers Matter

Two Problems. Two Solutions.
One Architecture.

Your cameras are sending degraded video to analytics that struggle with it. That's not one problem — it's two problems making each other worse.

The foggy feed causes your existing analytics to misfire. The misfiring analytics cause your team to stop trusting the system. And the moment your team stops trusting it, the entire investment becomes overhead.

See Layer · ProHawk AI
40%
Reduction in false positives
ProHawk restores the image — it does not detect objects or trigger alerts. It clears fog, rain, darkness, and glare from the camera feed before any analytics platform processes it. In validated testing, this single change — same cameras, same analytics, same settings — reduced false positives by 40%.
Source: ProHawk AI · Patent US 8,724,196 B2
+
Think Layer · Vaidio
2,000→10
False alarms per month cut
Vaidio is the analytics platform — it detects objects, recognizes behavior, triggers alerts, and reasons about what it sees. In a separate deployment, replacing a competitor's analytics with Vaidio on the same cameras — no ProHawk, no restoration — reduced false alarms from 2,000 per month to 10.
Source: Vaidio · International media company deployment
If you fix only the input (ProHawk without Vaidio)

Your existing analytics now receive clean, restored video. They perform measurably better — 40% fewer false positives. But if those analytics have weak detection models, you've improved the situation without fully solving it.

If you fix only the analytics (Vaidio without ProHawk)

A world-class intelligence platform is now processing your feeds — dramatically better than most competitors. But even Vaidio can't detect what the camera literally cannot capture. On foggy mornings, it's still working with compromised input.

These are independent, verified results from different deployments. We don't combine them — because they each stand on their own. Partner One delivers both layers, because fixing only one still leaves the other problem in place.

When both layers are deployed, you've solved the input problem and the processing problem at the same time. Clean cameras feeding analytics smart enough to use them.

Where This Technology Began

The origins of ProHawk's restoration technology stem from work done for the Japanese nuclear power industry — environments where cameras operate in extreme radiation, heat, and contamination, and where visual clarity is not optional. That foundation gives ProHawk a unique understanding of real-world visual challenges that no competitor developed in a software lab can match.

See It Work on Your Cameras

You've seen what these layers do in deployment. Now find out what they do in your environment.

Start the Conversation →