Dataset for Aircraft Dent Detection utilizing Specular Reflection and Deep Learning

Abstract

The aircraft dent dataset is created as part of the Master Thesis titled "Aircraft Dent Detection utilizing Specular Reflection and Deep Learning" in collaboration with the German Aerospace Center (DLR e.V.) at the Institute of Maintenance, Repair, and Overhaul within the Robot-Assisted Inspection and Repair research group.

Surface dents are difficult to detect visually since they usually lack distinguishing colors, and can only be seen via shadows or reflections. To make dent detection easier and more reliable for deep learning models, a capturing setup utilizing specular reflections is tested. This necessitates the creation of a new dataset of aircraft surface images, where dents are made visible with the help of specular reflections. A new annotation method that makes use of an optical tracking system to automatically create annotations was developed to create the dataset. The dataset contains over 6000 labeled images of dents, making it one of the largest publicly available dataset in this domain.