Posters
Come meet the poster presenters to ask them questions and discuss their work
We would like to invite you to come and see the posters at our upcoming conference. The posters will showcase a diverse range of research topics, and will give delegates an opportunity to engage with the authors and learn more about their work. Whether you are a seasoned researcher or simply curious about the latest developments in your field, we believe that the posters will offer something of interest to everyone. So please join us at the conference and take advantage of this opportunity to learn and engage with your peers in the academic community. We look forward to seeing you there!
PO271: Blade inspections with drones - can a simple photography calculation method improve the reliability of damage diagnosis?
Michel Greiner, Expert Systems Engineering, Omexom Renewable Energies Offshore GmbH
Abstract
Although the use of drones to inspect rotor blades has been common practice since several years, the question of how to reliably capture damage with a drone camera still remains. This question became even more important after the introduction of artificial intelligence (AI) as a tool to assist the process. Guided by the experience of several inspection campaigns with drones and the evaluation of thousands of images, the team of experts at Omexom Offshore addressed the question of whether small damages such as cracks or scratches (also known as deviations) leave "traces" in the captured image and can therefore be identified as damages - either by the human eye (screening of images on a monitor) or by an AI tool. The underlying goal was to find out how the camera hardware (sensor type, objective, etc.) and image capturing parameters (the angle of the image, aperture values, distance, lighting, etc.) influence the quality of the images. Based on this, camera hardware and parameters are to be defined so that deviations of a given size reliably leave "traces" in an image. The investigation was carried out by using stickers designed with precisely defined geometric patterns. The stickers were applied to the surface of rotor blades and, in a second step, images of them were captured by drones. The captured images were then analysed by a human on a computer to investigate the relationship between certain geometric patterns and their respective trace on the final image, as well as the influence of the camera hardware and the parameter setup. Finally, a consistent quantification of the camera hardware and parameter setup in relation to the visible traces of a given geometry from the sticker to the image was established. The result is a significative improvement in the reliability of diagnosis based on drone images.
