TopCAT and PySESA: Open-source software tools for point cloud decimation, roughness analyses, and quantitative description of terrestrial surfaces

Abstract:

The increasing use of high-resolution topography (HRT) constructed from point clouds obtained from technology such as LiDAR, SoNAR, SAR, SfM and a variety of range-imaging techniques, has created a demand for custom analytical tools and software for point cloud decimation (data thinning and gridding) and spatially explicit statistical analysis of terrestrial surfaces. We will present on a number of analytical and computational tools designed to quantify surface roughness and texture, directly from point clouds in a variety of ways (using spatial- and frequency-domain statistics). TopCAT (Topographic Point Cloud Analysis Toolkit; Brasington et al., 2012) and PySESA (Python program for Spatially Explicit Spectral Analysis) both work by applying a small moving window to (x,y,z) data to calculate a suite of (spatial and spectral domain) statistics, which are then spatially-referenced on a regular (x,y) grid at a user-defined resolution.

Collectively, these tools facilitate quantitative description of surfaces and may allow, for example, fully automated texture characterization and segmentation, roughness and grain size calculation, and feature detection and classification, on very large point clouds with great computational efficiency. Using tools such as these, it may be possible to detect geomorphic change in surfaces which have undergone minimal elevation difference, for example deflation surfaces which have coarsened but undergone no net elevation change, or surfaces which have eroded and accreted, leaving behind a different textural surface expression than before. The functionalities of the two toolboxes are illustrated with example high-resolution bathymetric point cloud data collected with multibeam echosounder, and topographic data collected with LiDAR.