@article{KorzeniowskaPfeiferLandtwing2018,
  author    = {Korzeniowska, Karolina and Pfeifer, Norbert and Landtwing, Stephan},
  title     = {Mapping gullies, dunes, lava fields, and landslides via surface roughness},
  series = {Geomorphology : an international journal on pure and applied geomorphology},
  volume    = {301},
  journal   = {Geomorphology : an international journal on pure and applied geomorphology},
  publisher = {Elsevier Science},
  address   = {Amsterdam},
  issn      = {0169-555X},
  doi       = {10.1016/j.geomorph.2017.10.011},
  pages     = {53 -- 67},
  year      = {2018},
  abstract  = {Gully erosion is a widespread and significant process involved in soil and land degradation. Mapping gullies helps to quantify past, and anticipate future, soil losses. Digital terrain models offer promising data for automatically detecting and mapping gullies especially in vegetated areas, although methods vary widely measures of local terrain roughness are the most varied and debated among these methods. Rarely do studies test the performance of roughness metrics for mapping gullies, limiting their applicability to small training areas. To this end, we systematically explored how local terrain roughness derived from high-resolution Light Detection And Ranging (LiDAR) data can aid in the unsupervised detection of gullies over a large area. We also tested expanding this method for other landforms diagnostic of similarly abrupt land-surface changes, including lava fields, dunes, and landslides, as well as investigating the influence of different roughness thresholds, resolutions of kernels, and input data resolution, and comparing our method with previously published roughness algorithms. Our results show that total curvature is a suitable metric for recognising analysed gullies and lava fields from LiDAR data, with comparable success to that of more sophisticated roughness metrics. Tested dunes or landslides remain difficult to distinguish from the surrounding landscape, partly because they are not easily defined in terms of their topographic signature.},
  language  = {en}
}
@article{KorzeniowskaKorup2017,
  author    = {Korzeniowska, Karolina and Korup, Oliver},
  title     = {Object-Based Detection of Lakes Prone to Seasonal Ice Cover on the Tibetan Plateau},
  series = {Remote sensing},
  volume    = {9},
  journal   = {Remote sensing},
  publisher = {MDPI},
  address   = {Basel},
  issn      = {2072-4292},
  doi       = {10.3390/rs9040339},
  pages     = {23},
  year      = {2017},
  language  = {en}
}
@article{KorzeniowskaBuehlerMartyetal.2017,
  author    = {Korzeniowska, Karolina and Buehler, Yves and Marty, Mauro and Korup, Oliver},
  title     = {Regional snow-avalanche detection using object-based image analysis of near-infrared aerial imagery},
  series = {Natural hazards and earth system sciences},
  volume    = {17},
  journal   = {Natural hazards and earth system sciences},
  publisher = {Copernicus},
  address   = {G{\"o}ttingen},
  issn      = {1561-8633},
  doi       = {10.5194/nhess-17-1823-2017},
  pages     = {1823 -- 1836},
  year      = {2017},
  abstract  = {Snow avalanches are destructive mass movements in mountain regions that continue to claim lives and cause infrastructural damage and traffic detours. Given that avalanches often occur in remote and poorly accessible steep terrain, their detection and mapping is extensive and time consuming. Nonetheless, systematic avalanche detection over large areas could help to generate more complete and up-to-date inventories (cadastres) necessary for validating avalanche forecasting and hazard mapping. In this study, we focused on automatically detecting avalanches and classifying them into release zones, tracks, and run-out zones based on 0.25m near-infrared (NIR) ADS80-SH92 aerial imagery using an object-based image analysis (OBIA) approach. Our algorithm takes into account the brightness, the normalised difference vegetation index (NDVI), the normalised difference water index (NDWI), and its standard deviation (SDNDWI) to distinguish avalanches from other land-surface elements. Using normalised parameters allows applying this method across large areas. We trained the method by analysing the properties of snow avalanches at three 4km\&\#8722;2 areas near Davos, Switzerland. We compared the results with manually mapped avalanche polygons and obtained a user's accuracy of >0.9 and a Cohen's kappa of 0.79-0.85. Testing the method for a larger area of 226.3km\&\#8722;2, we estimated producer's and user's accuracies of 0.61 and 0.78, respectively, with a Cohen's kappa of 0.67. Detected avalanches that overlapped with reference data by >80\% occurred randomly throughout the testing area, showing that our method avoids overfitting. Our method has potential for large-scale avalanche mapping, although further investigations into other regions are desirable to verify the robustness of our selected thresholds and the transferability of the method.},
  language  = {en}
}
@article{KorzeniowskaBuehlerMartyetal.2017,
  author    = {Korzeniowska, Karolina and B{\"u}hler, Yves and Marty, Mauro and Korup, Oliver},
  title     = {Regional snow-avalanche detection using object-based image analysis of near-infrared aerial imagery},
  series = {Natural hazards and earth system sciences : NHESS},
  volume    = {17},
  journal   = {Natural hazards and earth system sciences : NHESS},
  publisher = {Copernicus},
  address   = {G{\"o}ttingen},
  issn      = {1561-8633},
  doi       = {10.5194/nhess-17-1823-2017},
  pages     = {1823 -- 1836},
  year      = {2017},
  abstract  = {Snow avalanches are destructive mass movements in mountain regions that continue to claim lives and cause infrastructural damage and traffic detours. Given that avalanches often occur in remote and poorly accessible steep terrain, their detection and mapping is extensive and time consuming. Nonetheless, systematic avalanche detection over large areas could help to generate more complete and up-to-date inventories (cadastres) necessary for validating avalanche forecasting and hazard mapping. In this study, we focused on automatically detecting avalanches and classifying them into release zones, tracks, and run-out zones based on 0.25 m near-infrared (NIR) ADS80-SH92 aerial imagery using an object-based image analysis (OBIA) approach. Our algorithm takes into account the brightness, the normalised difference vegetation index (NDVI), the normalised difference water index (NDWI), and its standard deviation (SDNDWI) to distinguish avalanches from other land-surface elements. Using normalised parameters allows applying this method across large areas. We trained the method by analysing the properties of snow avalanches at three 4 km-2 areas near Davos, Switzerland. We compared the results with manually mapped avalanche polygons and obtained a user's accuracy of > 0.9 and a Cohen's kappa of 0.79-0.85. Testing the method for a larger area of 226.3 km-2, we estimated producer's and user's accuracies of 0.61 and 0.78, respectively, with a Cohen's kappa of 0.67. Detected avalanches that overlapped with reference data by > 80 \% occurred randomly throughout the testing area, showing that our method avoids overfitting. Our method has potential for large-scale avalanche mapping, although further investigations into other regions are desirable to verify the robustness of our selected thresholds and the transferability of the method.},
  language  = {en}
}