Photogrammetric and light detection and ranging data analysis for digital terrain modeling in linear infrastructure

In the article, elevation data along the axes of a dirt road and an asphalt road in flat terrain were compared.For this purpose, data obtained from airborne laser scanning and from a photogrammetric survey using an unmanned aerial vehicle were used.The acquired data, in the form of a 1 × 1 m grid of terrain elevation points, were applied to create a digital terrain model in the form of a grid or triangles.The grid of elevation points was also resampled to resolutions of 2 m, 5 m, and 10 m.The analyses performed showed very good agreement of the acquired measurement data in the case of the asphalt road (particularly for grids with resolutions up to 5 m).Due to the uneven surface of the dirt road, photogrammetric data and the digital terrain model based on triangles provided a better representation of its terrain.For elevation differences between ALS and UAV data along road centrelines, the gravel road showed higher errors (RMSE from 0.09 m to 0.13 m) than the asphalt road (RMSE 0.04 m).The study introduces a new comparative workflow that provides a practical solution for linear infrastructure designers, supporting the selection of appropriate data sources and spatial resolutions depending on the project characteristics.