Title :
Integration of multi-seasonal Landsat 8 and TerraSAR-X data for urban mapping: An assessment
Author :
Villa, Paolo ; Fontanelli, Giacomo ; Crema, Alberto
Author_Institution :
Inst. for Electromagn. Sensing of the Environ., Milan, Italy
fDate :
March 30 2015-April 1 2015
Abstract :
Accurate land cover maps provide critical information to scientists and decision-makers involved in urban monitoring and management. Satellite remote sensing can be used for producing mid-resolution urban maps at regional scale, especially when integrating multispectral optical information with SAR data. Starting from processing of Landsat 8 and TerraSAR-X multi-seasonal data (March-August 2014) covering a study area located in Lombardy region (Italy), we carried out an assessment of urban mapping performance using different non-parametric supervised classification algorithms and input features. The results show that best overall accuracy is generally reached with Random Forest (95.5%) and Support Vector Machines (93.6%), using both optical and SAR information. Adding X-band backscatter as input information produced an average accuracy improvement around 3%. Among various land cover classes, detection errors were concentrated on urban sparse fabric, and vegetated land cover, especially when SAR features are not used as input.
Keywords :
land cover; learning (artificial intelligence); pattern classification; remote sensing by radar; support vector machines; synthetic aperture radar; terrain mapping; vegetation; vegetation mapping; AD 2014 03 to 08; Italy; Lombardy region; SAR features; SAR information; X-band backscatter; average accuracy improvement; decision-makers; detection errors; input features; input information; land cover classes; land cover maps; midresolution urban mapping; multiseasonal Landsat 8 data; multiseasonal TerraSAR-X data; multispectral optical information; nonparametric supervised classification algorithms; optical information; random forest; regional scale; satellite remote sensing; scientists; support vector machines; urban management; urban mapping performance; urban monitoring; urban sparse fabric; vegetated land cover; Feature extraction; Feedforward neural networks; Monitoring; Radio frequency; Radiometry; Spatial databases; Vegetation mapping;
Conference_Titel :
Urban Remote Sensing Event (JURSE), 2015 Joint
Conference_Location :
Lausanne
DOI :
10.1109/JURSE.2015.7120474