Code and Data package of the article "Fast Digital Terrain Model Anomaly Computation of LiDAR Data"

R code and data to apply the method describe in the paper "Fast computation of digital terrain model anomalies based on LiDAR data for geoglyph detection in the Amazon" https://doi.org/10.1080/2150704X.2022.2109942 .

The method compute and enhance the elevation anomalies of soil surface using digital terrain models (DTMs) at different spatial resolution obtained with the LiDAR point cloud, the LAS files (.las). The method is designed to normalize the elevation of each pixel in relation to its neighbours and to remove the influence of the main elevation in the image contrast. It takes advantage of open source softwares R and Rstudio [1,2] and of the lidR package [3] to speed up the DTMs computation for the LAS files. It is designed to require the minimum number of operations, in order to ease the application on large sets of LiDAR point-cloud files.

An example is given for application to a single LAS file and an other example show the parallelization of the method for several LAS files.

The three point cloud datasets (.las files) in the ./LAS directory were simulated for a fake landscape of two hills containing four geoglyphs, two walls on the left (a square and a circle) and two trenches (a circle and a square). To simulate the noise of DTM below forest, a positive noise generate from a Gamma distribution of shape parameter of and 1 scaled between 0 and 3 was added to the elevation value, for a random selection of 50% of the DTM pixels. We generated three LAS files, one with walls of +2m and trenches of -3m (000001.las) , one with walls of +1m and trenches of -1.5m (000002.las) and the last one with walls of +0.25m and trenches of -0.50m (000003.las).

[1] R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria,
2020.

[2] RStudio Team. RStudio: Integrated Development Environment for R. RStudio, PBC., Boston, MA, 2020.

[3] Roussel, J.R.; Auty, D. Airborne LiDAR Data Manipulation and Visualization for Forestry Applications. https://jean-romain.github.io/lidRbook/dtm.html#tin, 2020. R package version 3.0.4.