The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLII-2/W13
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W13, 1887–1892, 2019
https://doi.org/10.5194/isprs-archives-XLII-2-W13-1887-2019
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-2/W13, 1887–1892, 2019
https://doi.org/10.5194/isprs-archives-XLII-2-W13-1887-2019

  05 Jun 2019

05 Jun 2019

COMPARISON OF OBJECT BASED MACHINE LEARNING CLASSIFICATIONS OF PLANETSCOPE AND WORLDVIEW-3 SATELLITE IMAGES FOR LAND USE / COVER

A. Tuzcu1, G. Taskin2, and N. Musaoğlu1 A. Tuzcu et al.
  • 1Istanbul Technical University, Faculty of Civil Engineering, Department of Geomatics Engineering, 34469 Maslak, Istanbul, Turkey
  • 2Istanbul Technical University, Earthquake Engineering and Disaster Management Institute, 34469 Maslak, Istanbul, Turkey

Keywords: Remote sensing, Object Based Classification, Machine Learning, Land Use / Cover, Support Vector Machine

Abstract. The purpose of the study was to compare performance of the classification methods, that are Rule Based (RB) classifier and Support Vector Machine (SVM), of Planetscope and Worldview-3 satellite images in order to produce land use / cover thematic maps. Six classes, which are deep water, shallow water, vegetation, agricultural area, soil and saline soil, were considered. After performing the classification process, accuracy assessment was employed based on the error matrices. The results showed that, both of the classification methods and satellite data were adequate to classify the area. Besides, classification accuracy was improved when Worldview-3 satellite and SVM method were used. The classification accuracies of RB classification of Planetscope and Worldview-3 were %87 and %94 respectively and the classification accuracies of SVM classification of Planetscope and Worldview-3 were %93 and %96 respectively.