Artificial intelligence-based crack detection system for concrete structures

Artificial intelligence-based crack detection system for concrete structures
As both the number of companies that rely on drone-assisted inspection as well as the amount of data collected in concrete damage assessment are increasing, the need for development of Artificial Intelligent (AI) tools that will assist building and infrastructure inspectors to quickly retrieve meaningful insights from data is essential. To this end, the Danish Technological Institute works towards developing an automated solution for the benefit of drone-based inspection of concrete structures. The goal is to create a first generation of an automated crack detection system based on AI (particularly, Deep Learning). The initial version of the system - featured in this video - was developed and demonstrated in a controlled environment at DTI. Today, our crack detection system is being further developed and it will be validated in late-2018. The innovation behind merging reasonably mature technology of drones with cutting edge AI aims at providing a safer, faster and more productive inspection; creating new possibilities for increased effectiveness in infrastructure asset management by making periodic structure monitoring and/or damage assessment economically feasible.