Volume XLII-3
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3, 45-52, 2018
https://doi.org/10.5194/isprs-archives-XLII-3-45-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3, 45-52, 2018
https://doi.org/10.5194/isprs-archives-XLII-3-45-2018
© Author(s) 2018. This work is distributed under
the Creative Commons Attribution 4.0 License.

  30 Apr 2018

30 Apr 2018

PRACTICAL APPROACH FOR HYPERSPECTRAL IMAGE PROCESSING IN PYTHON

L. Annala, M. A. Eskelinen, J. Hämäläinen, A. Riihinen, and I. Pölönen L. Annala et al.
  • Faculty of Information Technology, University of Jyvaskyla, Finland

Keywords: Python, Data analysis, Hyperspectral imaging, Image processing, Machine learning, Open source

Abstract. Python is a very popular programming language among data scientists around the world. Python can also be used in hyperspectral data analysis. There are some toolboxes designed for spectral imaging, such as Spectral Python and HyperSpy, but there is a need for analysis pipeline, which is easy to use and agile for different solutions. We propose a Python pipeline which is built on packages xarray, Holoviews and scikit-learn. We have developed some of own tools, MaskAccessor, VisualisorAccessor and a spectral index library. They also fulfill our goal of easy and agile data processing. In this paper we will present our processing pipeline and demonstrate it in practice.