Skip to main content

AI workflows

AI-Driven MBES Classification at Production Scale

HydroStack leverages advanced deep learning to automatically classify MBES soundings into distinct, meaningful categories - eliminating repetitive cleanup and QA/QC bottlenecks from your workflow.

By automating sounding classification, HydroStack:

  • Reduces manual post-processing effort
  • Improves consistency and data confidence
  • Accelerates delivery timelines
  • Produces insight-ready datasets faster

Built for production environments, HydroStack processes millions of MBES points in seconds, enabling high-throughput classification at scale. The result: clean, delivery-ready outputs available in a fraction of traditional processing time.

Trained on diverse open-source MBES datasets, the classifier has been exposed to a wide range of seabed conditions and survey scenarios - enhancing its robustness across varied operational contexts.

North Sea windfarm survey examples

These screenshots show pile and cable survey data before and after classification and cleanup, illustrating how the classifier supports faster downstream processing.