Document Type
Data
Publication Date
4-21-2026
Abstract
This data was primarily used for the development and testing of preprocessing and artificial intelligence damage detection algorithms for Alaskan arctic roads. They provide a high-resolution, geospatial scan of almost 95 kilometers of Arctic roadways along the Dalton, Steese, and Nome Highways in Alaska. The data used was collected using a ground vehicle and uncrewed aircraft mounted LiDAR sensors between 2022 and 2024. The dataset provides a spatial resolution exceeding 300 pts/m2 and < 2 cm vertical accuracy with a total data volume just over 16GB. The data is broken into three regions including Nome, Prudhoe Bay, and Yukon. Each region contains multiple datasets representing the reference output for multiple stages of the processing including slicing into 100m segments, segmentation of the surrounding non-road data, and filtering the remaining noise/objects above the roads surface. Damage was also identified, categorized, and labeled within the Nome region due to the extensive damage from Typhoon Merbok.
File Types
LAZ, CSV, ZIP
DOI
10.31356/data039
Recommended Citation
Naima Kaabouch and Artificial Intelligence Research Center. "LiDAR Datasets for Development and Testing of Preprocessing and Artificial Intelligence Damage Detection of Alaskan Arctic Roads" (2026). Datasets. 39.
https://commons.und.edu/data/39
Comments
This material is based upon work supported by the Broad Agency Announcement Program and the Cold Regions Research and Engineering Laboratory (ERDC-CRREL) under Contract No. W913E524C0017.