Machine learning detects integrity issues in subsea video

Clarus Subsea’s iCUE leverages historical dataset of corrosion imagery.

Acteon unit Clarus Subsea Integrity has announced ‘iCUE’ an app to detect integrity anomalies in inspection videos taken of subsea assets such as subsea risers, pipelines or moorings. iCUE uses machine learning to identify corrosion and other defects in video footage obtained from ROV surveys.

Clarus provides subsea integrity engineering services and claims to have amassed a ‘vast’ knowledge base of subsea anomaly detection, monitoring and remediation. Its image base of corrosion and other anomalies provided a labelled dataset for the machine learning classifier. iCUE now provides operators with accurate identification of dangerous conditions that might otherwise go undetected. A 10x speed-up over human review is claimed.

Integrity managers can use the app to trend anomalies across multiple inspections conducted throughout the asset’s life and assess the risk/rewards of a possible extension for an asset that is approaching the end of its original design life. The company is now working on a real-time version to enable detection while surveying.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.