This is the authors' abstract. We don't add key points for this paper.
Light detection and ranging remote sensing technologies have a lot of applications, including robotics and autonomous vehicles, archeology and meteorology, geospatial mapping, etc. They provide the generation of 3D models, in particular, forested landscapes. In this research, we consider the representation of objects by coined hyper-height data cubes (HHDCs). HHDCs contain data ensuring the simple extraction of canopy height and digital terrain models. Each HHDC is a three-dimensional tensor of non-negative integers. The description of even small regions of the Earth and the construction of the corresponding digital model use a significant number of such structures. This leads to great memory, time, and traffic expenses that are required for storing and transferring data via communication channels. Data compression algorithms frequently address the issue of high resource costs. This paper analyzes several lossless compression methods, including bit packing, Golomb codes (GC), and run-length encoding (RLE), as well as the possibility of their combination, together with experimental verification for the considered application. As a result, we suggest combining GC with RLE. It is shown that it ensures higher memory savings than other considered methods when compressing a set of HHDCs simulated using the Smithsonian Environmental Research Center NEON data. Depending on the source data, the median value of the compression ratio is greater than 28.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Макарічев et al. (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: