Skip to main content

Case studies

Examples of field deployment

The industries and signal types changed, but every program followed the same standard from problem framing through operational validation.

Environment & ForestryUltrasonic tree diagnostics

Korea Forest Research Institute

Ultrasonic non-destructive diagnostics for earlier pine wilt disease detection

Duration

Research project

Ultrasonic NDT scan

Diagnostic method

On-site interpretation

Data processing

Portable field unit

Equipment form

Challenge

Early-stage pine wilt shows almost no visible symptoms, so diagnosis often came too late. Existing methods needed specialists and long inspection times, which made them hard to run at scale.

Approach

We built ultrasonic NDT equipment and a field reading interface that tracks changes in ultrasonic propagation inside trunks. Data is captured and read on site, without an external network.

Results

  • Earlier detection via ultrasonic analysis
  • Single on-site scan and read workflow
  • Less dependence on specialists
  • Detection of internal change in visually healthy trees
Logistics & PortCrane wire monitoring

Major Korean port operator (anonymized)

Real-time fatigue and anomaly monitoring for STS crane wires

Duration

Pilot completed

STS cranes

Target equipment

Continuous real-time

Collection mode

Challenge

STS crane wires accumulate fatigue fast under repeated heavy loads, but there was almost no real-time visibility beyond scheduled inspections.

Approach

We developed a system for continuous wire condition monitoring.

Results

  • Continuous real-time wire status monitoring
  • Fatigue buildup tracking even during operations
Security & ControlVoice-based identity verification

Company C

Speaker verification in high-noise environments

Duration

Pilot completed

Voiceprint + liveness

Auth method

Software-only layer

Deployment

Challenge

Conventional voice-based speaker verification had frequent misidentifications. Password-only access was hard to manage in multi-environment setups.

Approach

We integrated a noise-robust speaker verification model with liveness detection into the login flow. It runs as software-only — no additional hardware needed.

Results

  • Speaker verification that operates in high-noise conditions
  • Natural integration into the existing login flow