Snow Surface
Snow surface research focuses on understanding snow's physical properties and its impact across various domains, from hydrology and avalanche safety to autonomous vehicle navigation and remote sensing. Current research employs machine learning algorithms, including convolutional neural networks (CNNs) like U-Net and Segformer, and autoencoders, to improve snow detection and classification in diverse data sources such as satellite imagery, drone footage, and even ski-mounted sensors. These advancements are crucial for enhancing water resource management, improving safety in challenging weather conditions, and advancing our understanding of snow's role in Earth's systems and beyond, including exoplanet studies. The development of robust datasets, like the Nordic Vehicle Dataset, is also a key area of focus to support the training and evaluation of these models.
Papers
Nordic Vehicle Dataset (NVD): Performance of vehicle detectors using newly captured NVD from UAV in different snowy weather conditions
Hamam Mokayed, Amirhossein Nayebiastaneh, Kanjar De, Stergios Sozos, Olle Hagner, Bjorn Backe
A Method for Classifying Snow Using Ski-Mounted Strain Sensors
Florian McLelland, Floris van Breugel