Computer vision for corrosion detection


Principal investigator

Research areas


Engineered products from cars, planes, bridges and buildings, are comprised of materials that have been crafted by humans. Inotherwords, we put a heck of a lot of energy into metal production, only to have nature want the energy back, and causing metals to corrode. Corrosion costs the world over a trillion dollars a year (and rising), which is more than the whole GDP of Australia. In order to better protect against the impact of corrosion, and to also revolutionise its detection, computer vision can play an important role. In this project, the use of IBM WatsonTM for corrosion detection (from images) will be explored.


To produce a classifier and detector that will allow the community to further advance its utilisation of computer vision for good.


Ability to code (even if very elementary skills) would be beneficial.

An interest to learn Python.

Passion for data science.


computer vision, machine learning, AI, data science, corrosion, fun, rust

Updated:  1 June 2019/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing