Automating Subsea Asset Inspection for Corrosion/Pit Detection by Machine Learning

Any offshore operation involves the utilization and maintenance of a number of subsea assets, and corrosion inspection is one of the important components. Visual inspection plays a crucial step towards corrosion inspection of subsea asset(s). This visual inspection involves analysis of images of the subsea asset revealing the surface condition of the asset(s), which includes corrosion, pits and cracks. Such image analysis helps to understand the condition of asset(s) and strategize further integrity management practices. However, analyzing these images still remains a labour-intensive and time-consuming task, which is so far manually performed in the industry. In addition, possibility for human errors is another important drawback in the manual image analysis method. Therefore, automating these image analyses by machine learning application will not only make the process faster, but automation will also increase the accuracy. Our company is currently working on developing this automated inspection software application, which will ultimately become a flagship product of the company. This will enhance the revenue growth of the company and will help relieve the financial impact of the COVID pandemic.

Faculty Supervisor:

Stephen Czarnuch

Student:

Partner:

qualiTEAS Inc

Discipline:

Engineering

Sector:

Professional, scientific and technical services

University:

Memorial University of Newfoundland

Program: