The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVIII-4/W1-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVIII-4/W1-2022, 321–328, 2022
https://doi.org/10.5194/isprs-archives-XLVIII-4-W1-2022-321-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVIII-4/W1-2022, 321–328, 2022
https://doi.org/10.5194/isprs-archives-XLVIII-4-W1-2022-321-2022
 
06 Aug 2022
06 Aug 2022

EARTH OBSERVATION DATACUBES MULTI-VISUALIZATION TOOLBOX

C. Muller1, A. Lestrade1, M. Marty1, A. Sadiku1, J. Neijt2, Y. Voumard2, and S. Gobron1 C. Muller et al.
  • 1Department of Engineering, University of Applied Sciences West Switzerland (HES-SO), Neuchâtel, Switzerland
  • 2Solenix Schweiz GmbH, Härkingen, Switzerland

Keywords: Earth Observation, Data Visualization, Raycasting, 3D Rendering, Real-time visualization, GPGPU, WebGL

Abstract. Too much information often kills information. With the increasing number of satellites and their ever-increasing performance, new tools must be made available to deal with this data onslaught. We noticed that a number of computer graphics tools were largely under-exploited to help scientists better interpret and find relevant information in large datasets. A modern approach to run large processes efficiently is the use of GPUs, but nowadays the emphasis is often put on the parallel processing of geospatial datasets rather than focusing on their visualization. Considering geospatial data using GPU resources for intermediate computation and visualization is this paper main contribution. Given the increasing interest in interacting directly with this data using Web pages or Notebooks, this article presents tools allowing a program to run on the GPU and display, using the WebGL API, these matrices of data, also called datacubes. This paper shows a range of models applicable to datacubes deployed in the context of terrestrial observation. The end goal is to display on a PC very large (i.e. 10243) datacubes rendered on the fly and in real time. Furthermore, results show our models can process large amounts of data and render them in real time. All these innovative rendering models are assembled in a toolbox dedicated to datacube visualization. Finally, we give several examples of how to use this toolbox which enables the retrieval of raw data from an external server to real-time rendering on a local Web page.