Improving global scale land cover classifications with multi-directional POLDER data and a decision tree classifier

Evaluation of the Landsat-5 TM and Landsat-7 ETM+ surface reflectance products

“…However, BRDFAdjustment has low impact on blue band (P decreases by 25% from no-Adjustment to BRDF-Adjustment) while the impact is moderate on green (40%) and red (44%) bands and high on infrared bands (N 45%). These findings could suggest that BRDF effects increase with the wavelength, which is in agreement to previous studies (de Colstoun & Walthall, 2006;Mahtab, Sridhar, & Navalgund, 2009). Concerning Landsat-7 ETM +, P does not vary much between the two first levels of adjustment since most of MODIS observations are derived from Terra that are acquired in a similar geometry as Landsat-7.…”

Predicting land cover using GIS, Bayesian and evolutionary algorithm methods

“…The goal of this work is to show that evolutionary methods can provide a useful tool for creating an objective Bayesian network structure, by applying this optimisation to a real-life situation in which several data sources are used to map the land cover. We do not mean to state that other data-mining methodologies, such as decision trees (Brown De Colstoun and Walthall, 2006;Pal, 2006;McCarty et al, 2007) or neural networks (Kuplich, 2006;Cots-Folch et al, 2007) are less effective for producing land cover maps. Instead, our intention is to demonstrate the utility of linking evolutionary approaches with Bayesian systems in providing a classification system capable of flexibility in data type usage, and to show that a Bayesian system can be evolved to produce optimal or near-optimal network designs based on the datasets available.…”

A Method for Estimating the Aerodynamic Roughness Length with NDVI and BRDF Signatures Using Multi-Temporal Proba-V Data

“…One potential solution is to bring in multi-angular optical remote sensing [23][24][25][26]. Sunlight hitting the vegetation canopy is scattered unevenly because of surface roughness, which is related to the canopy's shape and height.…”