@article{fdi:010061283,
  title = {{L}arge-scale classification of water areas using airborne topographic lidar data},
  author = {{S}meeckaert, {J}. and {M}allet, {C}. and {D}avid, {N}. and {C}hehata, {N}esrine and {F}erraz, {A}.},
  editor = {},
  language = {{ENG}},
  abstract = {{A}ccurate {D}igital {T}errain {M}odels ({DTM}s) are inevitable inputs for mapping and analyzing areas subject to natural hazards. {T}opographic airborne laser scanning has become an established technique to characterize the {E}arth's surface: lidar provides 3{D} point clouds allowing for a fine reconstruction of the topography while preserving high frequencies of the relief. {F}or flood hazard modeling, the key step, before going onto terrain modeling, is the discrimination of land and water areas within the delivered point clouds. {T}herefore, instantaneous shorelines, river banks, and inland waters can be extracted as a basis for more reliable {DTM} generation. {T}his paper presents an automatic, efficient, and versatile workflow for land/water classification of airborne topographic lidar points, effective at large scales (>300 km(2)). {F}or that purpose, the {S}upport {V}ector {M}achine ({SVM}) method is used as a classification framework and it is embedded in a workflow designed for our specific goal. {F}irst, a restricted but carefully designed set of features, based only on 3{D} lidar point coordinates and flightline information, is defined as classifier input. {T}hen, the {SVM} learning step is performed on small but well-targeted areas thanks to a semiautomatic region growing strategy. {F}inally, label probability output by {SVM} is merged with contextual knowledge during a probabilistic relaxation step in order to remove pixel-wise misclassification. {R}esults show that a survey of hundreds of millions of points are labeled with high accuracy (>95% in most cases for coastal areas, and >90% for rivers) and that small natural and anthropic features of interest are still well classified even though we work at low point densities (0.5-4 pts/m(2)). {W}e also noticed that it may fail in water-logged areas. {N}evertheless, our approach remains valid for regional and national mapping purposes, coasts and rivers, and provides a strong basis for further discrimination of land-cover classes and coastal habitats.},
  keywords = {{L}idar ; {A}irborne ; {C}lassification ; {W}ater ; {S}upport vector machines ; {S}eashore ; {R}ivers ; {L}arge scale mapping},
  booktitle = {},
  journal = {{R}emote {S}ensing of {E}nvironment},
  volume = {138},
  numero = {},
  pages = {134--148},
  ISSN = {0034-4257},
  year = {2013},
  DOI = {10.1016/j.rse.2013.07.004},
  URL = {https://www.documentation.ird.fr/hor/fdi:010061283},
}