KU and Lockheed Martin to collaborate on computer vision project

KU plans series of research projects with the US group

Abu Dhabi-based Khalifa University and Lockheed Martin have announced a new computer vision and pattern recognition - or CVPR - research project using AI reinforcement learning. This is the first in a series of university research projects following an MoU on research and development signed with the US global aerospace, defence and space group last year. 

The new CVPR research project will create a framework to integrate computer vision, machine learning, and expert systems, capable of accurately locating markings and symbols on aircraft and, potentially, marine craft. 

CVPR has so far been applied to biometrics, target identification, aimpoint refinement and automated control. The Khalifa University project will use a CVPR framework to inspect and verify measurements, in order to improve the speed and accuracy of aircraft logo inspections during the NATO aircraft production cycle. The project aims to significantly improve quality assurance for the sign-off of placement, spacing and positioning of logos.

The project will also research techniques to improve the ability of CVPR to accurately register standard symbology in imagery. These techniques could then be applied in any process that is attempting to accurately locate markings and symbols. 

The ongoing partnership between Khalifa University and Lockheed Martin is expected to create unique R&D opportunities, help evolve the country’s research ecosystem and increase the pipeline of highly skilled aerospace, defence, technology and security engineers. Machine intelligence, autonomy, microelectronics, structural and thermal materials, and air vehicles were all identified as potential areas of research in the 2020 agreement.

• Read the Khalifa University press release.