Skip to content
United Vision

LinkVision platform

Built by United Vision. Built for accuracy.

Explore the AI inspection platform that connects capture, analysis, spatial context, and professional reporting.

Core values

Why choose United Vision?

01

Safer

Drone-first capture reduces scaffolding and work at height while keeping every observation traceable.

02

Faster

AI-assisted review compresses inspection cycles from weeks to hours for each building.

03

More economical

Efficient capture and automated analysis reduce labour, access, and equipment costs.

04

More accurate

High-resolution imagery, AI detection, and expert review support consistent, measurable findings.

05

More standardised

Structured records and export-ready reports keep projects aligned from capture to delivery.

Product capabilities

What can our solution do?

A streamlined interface for faster reviews and clearer, more accurate reports.

AI detection and assessment

Multi-model detection identifies and classifies façade and infrastructure defects from high-resolution imagery and assesses their safety level.

Request demo

Depth and dimension measurement

Monocular depth estimation uses drone-camera parameters to measure position, width, height, and affected area.

Request demo

3D reconstruction and defect mapping

Generate spatial building models and place every reviewed finding in its precise 3D context.

Request demo

Thermal imaging analysis

Reveal thermal anomalies linked to moisture, delamination, insulation failure, and other hidden conditions.

Request demo

Privacy protection

Automatically detect and blur faces, windows, licence plates, and other sensitive visual details.

Request demo

Automated report generation

Turn reviewed findings into consistent Word and Excel deliverables with charts, indices, and visual evidence.

Request demo

Flexible deployment

One platform. Two ways to work.

Choose the setup that fits your team. Both versions share core inspection tools; historical inspection comparison is currently exclusive to local deployment.

Local version

Run LinkVision on your own Windows computer with a software licence. Alongside the core inspection tools shared with the cloud version, local deployment also supports historical inspection comparison.

  • Local installation
  • Licensed software
  • Real-time online authentication
  • Historical inspection comparison

For teams that prefer to run the software on their own equipment. The local version is slightly less convenient to operate than the cloud version.

Keep your computer connected to the internet during use for real-time licence authentication.

Local analysis workspace01 / 02
Local analysis workspace, 1 / 2

Recommended hardware

A 2 TB drive is estimated to hold data for around 100 buildings. Actual capacity depends on the amount of data per building; reserve additional storage for larger projects.

Operating system
Windows
Graphics card
NVIDIA GeForce RTX 4090
Processor
Intel Core i9
Memory
32 GB
Storage
2 TB or more

Core features

Residential building defects01 / 02
Concrete crack
Concrete spalling
Concrete delamination
Degraded plaster
Vegetation intrusion
Rust stain
Tile crack
Tile spalling
Water stain
Bulging
Mould contamination
Craquelure
Residential building defects, 1 / 2
01

AI defect detection

LinkVision combines automated detection with engineering review to turn large-scale drone imagery into prioritised, decision-ready findings.

  • Automated detectionIdentify 74+ building-defect types and an extensible library of infrastructure conditions.
  • Severity assessmentGrade findings by risk so teams can prioritise investigation and maintenance.
  • Precise measurementEstimate defect position, width, height, and area directly from source imagery.
  • Decision supportApply configurable standards and local assessment rules to every project.
Building thermal anomalies01 / 02
Façade hollowing
Façade hollowing — second view
Moisture intrusion
Moisture intrusion — second view
Curtain-wall water leakage
Insulation defect
Glass seal failure
Curtain-wall insulation failure
Envelope heat loss
Building thermal anomalies, 1 / 2
02

Thermal imaging analysis

Analyse infrared imagery alongside visible-light records to identify thermal anomalies on building façades. The combined view helps reveal moisture intrusion, delamination, insulation defects, and conditions that are difficult to confirm visually.

Vehicle privacy protection01 / 03
Vehicle privacy protection, 1 / 3
03

Privacy protection

99% privacy protection: a built-in workflow detects and obscures sensitive targets such as faces, windows, and licence plates. Teams can retain useful inspection evidence while supporting privacy and project-compliance requirements.

Lee Chung Street building model01 / 05
Lee Chung Street building model, 1 / 5
04

3D reconstruction and defect localisation

Automatically reconstruct a building from coordinated drone-image sequences, then map each defect into the 3D model. Floor-level positioning and elevation plans make each issue easier to locate. The spatial view supports clearer collaboration across project teams.

South elevation defect report01 / 03
South elevation defect report, 1 / 3
05

Automated report generation

Organise all AI detection results and descriptions in a structured workspace. Reviewers can verify and correct findings with live previews and floor-level positioning. Automatically generated reports include key information linked to locations in the 3D model for clear, readable results.

  • Defect indexFilter by type, location, and severity.
  • Detailed inspection dataInclude dimensions, confidence, evidence, and repair recommendations.
  • Visual analyticsSummarise results with charts and spatial views.
2D elevations and original photos across inspection periods01 / 02
2D elevations and original photos across inspection periods, 1 / 2
06

Historical inspection comparison

Compare inspection records for the same building across different periods using 2D elevations and 3D models, together with original photos and changes in defect counts within selected regions. Engineers can examine defect details and track building conditions to inform reinspection and maintenance assessments. This feature is currently available only in the local deployment version.

Core capabilities

Platform capabilities at a glance

A complete drone-based façade inspection workflow, from defect recognition to project deliverables.

CapabilityOur platform
AI defect detection

74+ defect types

Optical and infrared recognition.

Detection accuracy

90%–95%

Measured against manual review. Cloud and local deployments use the same assessment basis.

Field-test sample coverage

Samples across a broad range of building types

Samples come from real inspection projects, with coverage continuing to expand.

Multiple AI models

Coordinated AI models

Models share feature extraction, defect localization, and result reasoning to support recognition accuracy and output quality.

Severity assessment

AI-assisted severity assessment and defect descriptions

Dimension estimation

Image-based estimation of physical defect dimensions

Drone route planning

Sparse-point planning without a prebuilt 3D model

3D reconstruction

3D reconstruction and defect mapping, with floor-level positioning, elevation plans, and interactive review of full-façade 2D views and 3D models

Historical inspection comparison

Available in local deployment only

Compare 2D elevations, 3D models, original photos, and regional defect counts across inspection periods.

Automated reports

Structured inspection reports with a human review step

Privacy protection

Adaptive privacy blurring for faces, windows, doors, and other sensitive regions

Deployment options

Cloud platform

A web-based platform accessed directly in your browser, with no installation required.

Local deployment

Process and keep project data on your own device. A network connection is required for real-time license verification.

Recommended local hardware

RTX 4090 GPU · Intel Core i9 · 32 GB RAM · 2 TB storage or more

Approximately 100 buildings of project data, depending on project scale. Reserve additional storage for larger projects.

Processing time

4–8 hours in project field tests

Actual time depends on the inspection scope and hardware configuration.

Request demo

Tell us about your project and we will arrange a tailored LinkVision walkthrough.

A customer arranging a LinkVision demonstration
Built around your workflowSee LinkVision on your own use case.