Subpixel urban land cover estimation: comparing cubist, random forests, and support vector regression

  title={Subpixel urban land cover estimation: comparing cubist, random forests, and support vector regression},
  author={J. T. Walton},
  journal={Photogrammetric Engineering and Remote Sensing},
  • J. T. Walton
  • Published 2008
  • Geography
  • Photogrammetric Engineering and Remote Sensing
  • Three machine learning subpixel estimation methods (Cubist, Random Forests, and support vector regression) were applied to estimate urban cover. Urban forest canopy cover and impervious surface cover were estimated from Landsat-7 ETMimagery using a higher resolution cover map resampled to 30 m as training and reference data. Three different band combinations (reflectance, tasseled cap, and both reflectance and tasseled cap plus thermal) were compared for their effectiveness with each of the… CONTINUE READING
    127 Citations

    Figures and Tables from this paper.

    Subpixel Mapping of Urban Areas Using EnMAP Data and Multioutput Support Vector Regression
    • 16
    Improving sub-pixel imperviousness change prediction by ensembling heterogeneous non-linear regression models
    • 4
    • Highly Influenced
    • PDF


    Urban cover mapping using digital, high-spatial resolution aerial imagery
    • 102
    • Highly Influential
    Percent tree cover estimation using regression tree method: a case study of Africa with very-high resolution QuickBird images as training data
    • 10
    Estimation and vicarious validation of urban vegetation abundance by spectral mixture analysis
    • 148
    • PDF
    Estimating impervious surface distribution by spectral mixture analysis
    • 689
    • PDF
    Random forest classifier for remote sensing classification
    • 1,209