International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Volume XLII-3/W7
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W7, 25–32, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W7-25-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W7, 25–32, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W7-25-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  01 Mar 2019

01 Mar 2019

A FIT-FOR-PURPOSE ALGORITHM FOR ENVIRONMENTAL MONITORING BASED ON MAXIMUM LIKELIHOOD, SUPPORT VECTOR MACHINE AND RANDOM FOREST

A. Jamali A. Jamali
  • Faculty of Surveying Engineering, Apadana Institute of Higher Education, Shiraz, Iran

Keywords: Image classification, Earth Observation, Support Vector Machine, Random Forest, R

Abstract. Due to concerns of recent earth climate changes such as an increase of earth surface temperature and monitoring its effect on earth surface, environmental monitoring is a necessity. Environmental change monitoring in earth sciences needs land use land cover change (LULCC) modelling as a key factor to investigate impact of climate change phenomena such as droughts and floods on earth surface land cover. There are several free and commercial multi/hyper spectral data sources of Earth Observation (EO) satellites including Landsat, Sentinel and Spot. In this paper, for land use land cover modelling (LULCM), image classification of Landsat 8 using several mathematical and machine learning algorithms including Support Vector Machine (SVM), Random Forest (RF), Maximum Likelihood (ML) and a combination of SVM, ML and RF as a fit-for-purpose algorithm are implemented in R programming language and compared in terms of overall accuracy for image classification.