Super-resolution techniques to enhance the low-resolution inspection images and video of drones
Reference persons GIAN PAOLO CIMELLARO
Description The quality of monitoring data is an important parameter that profoundly impacts the accuracy of AI-based defect detection algorithms. Any corruption of the monitoring data can degrade the system performance considerably. In the case of visual inspections using drones, the image and video quality is often compromised due to the use of low-cost cameras, motion blur induced by the UAS movements, and distortions caused by fisheye lenses. Moreover, physical inaccessibility, safety and security concerns often prevent the UAS from going from every perspective very close to the structure under investigation, resulting in low resolution imaging, which lacks the optical details necessary for accurate analytics.
This research will exploit the latest AI-based video super-resolution imaging techniques to reconstruct high-resolution pictures and videos from the original low resolution inputs. This will allow to maintain a large working distance and viewing perspective by flying far from the structure and thereby complete a given inspection task within the boundary of limited flight time. The developed algorithms should be lightweight, compatibly with the drones’ limited onboard computation capacity and competent for real-time vision-based damage identification. The algorithms will resort to a layered approach to systematically maximize information gain and minimize uncertainty per objects inspected and to reduce need of high-resolution computations, also through explorative close-in missions.
Deadline 17/11/2023 PROPONI LA TUA CANDIDATURA