TY - JOUR
T1 - Performance evaluation of road and building classifiers on VHR images
AU - Simler, C.
AU - Beumier, C.
PY - 2010
Y1 - 2010
N2 - A method is proposed for building and road detection on VHR multispectral aerial images of dense urban areas. Spatial and spectral features of segmented areas are classified using a 3-class SVM integrating some a priori and contextual information to handle unclassified patterns and conflicts. Geometrical object features and additional information improve the classification accuracy in the difficult case where many building roofs are grey like the roads and have similar geometry. Also, road network regularization is suggested to improve the classification accuracy.
AB - A method is proposed for building and road detection on VHR multispectral aerial images of dense urban areas. Spatial and spectral features of segmented areas are classified using a 3-class SVM integrating some a priori and contextual information to handle unclassified patterns and conflicts. Geometrical object features and additional information improve the classification accuracy in the difficult case where many building roofs are grey like the roads and have similar geometry. Also, road network regularization is suggested to improve the classification accuracy.
KW - Classifier combination
KW - Mean shift
KW - Support vector machine
KW - Very high spatial resolution image
UR - http://www.scopus.com/inward/record.url?scp=84924012120&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:84924012120
SN - 1682-1750
VL - 38
JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
IS - 4C7
T2 - Geographic Object-Based Image Analysis, GEOBIA 2010
Y2 - 29 June 2010 through 2 July 2010
ER -