Pyoints: A Python package for point cloud, voxel and raster processing.

Abstract

The evolution of automated systems like autonomous robots and unmanned aerial vehicles leads to manifold opportunities in science, agriculture and industry. Remote sensing devices, like laser scanners and multi-spectral cameras, can be combined with sensor networks to all-embracingly monitor a research object. The analysis of such big data is based on geoinformatics and remote sensing techniques. Today, next to physically driven approaches, machine learning techniques are often used to extract relevant thematical information from the data sets. Analysis requires a fusion of the data sets, which is made difficult conceptually and technically by different data dimensions, data structures, and various spatial, spectral, and temporal resolutions.

Eulerian Parameter Inference: A Probabilistic Change of Variables for Model-Based Inference with High-Variability Data Sets

“…We present our EPI inference scheme (Figure 2) using a data set containing annual average temperatures y (i) of 3 168 weather stations around the world and a model s(q) = y that describes these temperatures as a function of the corresponding latitudes q (i) . The data was collected through the meteostat Python library [22] (Source: Meteostat) and is shown as orange dots in the upper left subgraph of Figure 2. Empirically, we assume that the temperature y is proportional to the location's cosine of the latitude q.…”

“…The datasets analysed during this study are available in the FAIRDOMHub repository accessible through this DOI. The data for the first application scenario was taken from the meteostat Python library [22] (Source: Meteostat). The data for the second example was taken from the publicly available file of the Robert Koch Institute [24].…”

Integration of Remote Sensing Data into a Composite Voxel Model for Environmental Performance Analysis of Terraced Vineyards in Tuscany, Italy

“…Both terrestrial and aerial autonomous robotic platforms are constantly collecting large quantities of geospatial data, which are centrally processed to support the execution of common orchard maintenance tasks, such as irrigation. The need for interdisciplinary data integration resulted in the publication of a pyoints python library [31], which bridges different representations of geometric point-based data, including point clouds and geo-referenced rasters, as well as the voxels required by the prototype of farming robots aimed at precision agriculture applications [32].…”

ALS as Tool to Study Preferred Stem Inclination Directions

“…The triangulation allows for a linear interpolation of the elevation within the given tile. In detail, to identify points representing the terrain, the points classified as ground are thinned with a duplicate point filter identifying local minima with a radius of 0.8 m (see filters.surface of Lamprecht [33]). Similarly, to identify points representing the canopy, the points classified as vegetation are thinned with the same duplicate point filter extracting local maxima with a radius of 0.8 m.…”