DarkVision provides palm‑sized ultra‑high‑resolution 3D acoustic imaging systems that capture volumetric data at up to 3 Tbps and convert it into detailed point‑cloud visualizations. Cloud‑based AI analytics automatically detect and size defects as small as 0.01 in, enabling operators in oil & gas, aerospace, and heavy manufacturing to assess asset integrity quickly and reduce downtime.
Funding
Funding not disclosed
Founders
Product
Problem
Industrial assets such as pipelines, downhole equipment, and aerospace components require regular inspection to ensure safety and performance, but existing nondestructive testing methods often lack the resolution and data fidelity needed to detect small or early‑stage defects, especially in harsh environments.
Solution
DarkVision delivers ultra‑high‑resolution 3D acoustic imaging using palm‑sized systems that combine solid‑state, ultra‑dense 2D transducer arrays with proprietary silicon chips. The hardware captures volumetric acoustic data at rates up to 3 Tbps, generating terabytes of raw signals per scan. Cloud‑based signal processing and AI‑driven analytics transform this data into high‑fidelity point clouds and visualizations, automatically identifying and quantifying defects as small as 0.01 in. The resulting actionable insights enable engineers and asset owners to assess integrity quickly, prioritize maintenance, and reduce downtime.
Target Audience
Primary customers are operators and engineers in sectors such as oil & gas (downhole and pipeline), aerospace, and heavy manufacturing who need precise, rapid inspection of critical infrastructure.
Features
- Solid‑state arrays with up to 4,000+ transducers per array, supporting full‑waveform imaging modes
- Real‑time digital signal processing capable of 614 Gb/s data throughput
- Cloud‑native machine‑learning pipelines that analyze terabytes of acoustic data to detect and size defects
- High‑fidelity visualization engine producing photorealistic 3D point clouds and sub‑millimeter defect maps
- Capability to operate under extreme conditions (up to 15,000 psi pressure and 302 °F temperature)
- Extensive labeled acoustic dataset (>10,000 TB) supporting continuous model improvement