Project Details


Managing offshore corrosion is a subject that is heavily regulated and combined with several industry standards that must be followed to ensure the structure integrity of the assets. Yet, whilst, and perhaps even because, of those regulations, the current process of corrosion management for offshore assets remains time consuming, unstructured, inefficient, and rather subjective.

The PACMAN project develops and demonstrates the application of predictive corrosion management. The driver in this project is to drastically reduce the number of manual hours, which today makes corrosion detection too time consuming and inefficient. The tools for doing this are: automatic position tagging, improved corrosion detection through more advanced camera technology, improved machine learning methods for visual detection of corrosion in 2D images and automated transfer of 2D image corrosion findings to a 3D interpretation for more efficient handling of these corrosion findings. All of these tools are used in the predictive corrosion management. Essentially, predictive corrosion management is about severity classification of the found corrosion.

The project involves the development and maturement of an automatic corrosion detection and prediction program. The program should then via AI and machine learning be capable of conducting autonomous corrosion predictions based on numerous inputs, be that in terms of images or specific sensor technology.
Short titlePACMAN
Effective start/end date01/02/202231/01/2024

Collaborative partners

  • Semco Maritime
  • IPU
  • MM Survey
  • EUDP
  • Energy Cluster Denmark

UN Sustainable Development Goals

In 2015, UN member states agreed to 17 global Sustainable Development Goals (SDGs) to end poverty, protect the planet and ensure prosperity for all. This project contributes towards the following SDG(s):

  • SDG 7 - Affordable and Clean Energy
  • SDG 11 - Sustainable Cities and Communities
  • SDG 12 - Responsible Consumption and Production


  • Corrosion Management
  • Deep Learning
  • Computer Vision


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.