Volume XLII-4/W16
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W16, 283–289, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W16-283-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-4/W16, 283–289, 2019
https://doi.org/10.5194/isprs-archives-XLII-4-W16-283-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  01 Oct 2019

01 Oct 2019

EVALUATION OF ADVANCED DATA MINING ALGORITHMS IN LAND USE/LAND COVER MAPPING

A. Jamali1 and A. Abdul Rahman2 A. Jamali and A. Abdul Rahman
  • 1Faculty of Surveying Engineering, Apadana Institute of Higher Education, Shiraz, Iran
  • 2Faculty of Geoinformation and Real Estate, Universiti Teknologi Malaysia (UTM), Malaysia

Keywords: Land use Land cover, LULC, machine learning, data mining, image classification

Abstract. For environmental monitoring, land-cover mapping, and urban planning, remote sensing is an effective method. In this paper, firstly, for land use land cover mapping, Landsat 8 OLI image classification based on six advanced mathematical algorithms of machine learning including Random Forest, Decision Table, DTNB, Multilayer Perceptron, Non-Nested Generalized Exemplars (NN ge) and Simple Logistic is used. Then, results are compared in the terms of Overall Accuracy (OA), Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) for land use land cover (LULC) mapping. Based on the training and test datasets, Simple Logistic had the best performance in terms of OA, MAE and RMSE values of 99.9293, 0.0006 and 0.016 for training dataset and values of 99.9467, 0.0005 and 0.0153 for the test dataset.