Funded by the European Union – Next Generation EU
The use of prefabricated concrete components is now common in the construction of bridges and viaducts. Ensuring the quality of these often very large elements is essential for the safety of infrastructures. However, final inspections can be complex and costly.
The project developed with Immodrone was created to address this need by introducing an innovative system based on Computer Vision and Artificial Intelligence to automate inspections and support technicians with objective analyses. By using drones and AI, the aim is to make inspections faster, more precise, and reliable, contributing to the digitalization of the infrastructure sector.
The developed solution is based on a smart visual inspection system capable of detecting anomalies in prefabricated components. Through drones equipped with cameras, images of bolts, rivets, or other objects are captured, which, in addition to documenting their integrity, are processed through software. This data feeds Deep Learning models, trained to identify defects such as cracks, surface irregularities, or damage. The result is an objective, precise, and scalable system that supports technicians in verifying the quality of the artifacts.
The developed model was tested on the test set, demonstrating a good detection capability of the defined classes, even under variable framing and image angle conditions. In particular, it was observed that some classes are more easily identifiable from certain perspectives, thanks to greater visibility and clarity of structural elements.
Funded by the European Union – Next Generation EU
Don’t miss any updates on tenders, job openings, and CIM news directly in your feed.
"*" indicates required fields