Research internship: UAV lidar point cloud denoising (job offer closed)
6 months – Master 2 Research Internship.
Yellowscan, Montpellier
Contact: pol.kennel AT yellowscan-lidar.com
PhD opportunity at the end of the internship.
Context
YellowScan designs, develops, and produces mapping sensors for professional UAV (unmanned arial vehicle) applications. Fully integrated, ultra-light and easy to use, these highly automated lidar (light detection and ranging) sensors are used by customers worldwide in various fields: topography, environmental research, archaeology, civil engineering, and mining. To rapidly develop our business with a very high level of quality and services, we are actively looking for the best talents to reinforce our team. Today, we are offering a Master of Science internship to join our R&D team.
Project description
Lidar is a laser remote sensing technology widely used in geographic information systems. Achieving centimeter accuracy in the positioning of data collected during UAV lidar surveys is essential for mapping applications. Each georeferenced measurement, i.e., each point of the lidar point cloud, depends on the quality of the system’s instru-ments (laser scanner, inertial measurement unit, satellite positioning) and the acquisition conditions (weather, satellite reception, etc.).
The goal of this internship is to develop a state-of-the-art method for the reconstruction of local surfaces in the point cloud to allow the displacement and/or the deletion of points in order to eliminate the point cloud noise. The aim is to produce high quality data, i.e., with thin and smooth surfaces, while preserving fine struc-tures such as building edges, natural and/or complex objects (see figure below). This internship may lead to a PhD.
Skills
- Excellent skills in C++ and knowledge of Git
- Organization, methodology and autonomy
- Interest in new technologies
- Skills in computer vision, 3D, image processing, machine learning, statistics or applied mathematics.