Dakar land use map at street block level

This datatset contains a land use classification of Dakar (Senegal) at the street block level. It was created following the methodology presented in [1].

Description of the files:

"Dakar_landuse_shapefile.zip" : Shapefile of the street blocks extracted from OpenStreetMap using [2] with classification results in the attribute table.

"Dakar_landuse_style.zip" : Files for style of the shapefile.

Attribute table content:

"CAT", "GID" : ID of the street block

"PROB_ACS" : Probability to belong to class ACS

"PROB_AGRI" : Probability to belong to class AGRI

"PROB_BARE" : Probability to belong to class BARE

"PROB_DEPR" : Probability to belong to class DEPR

"PROB_PLAN" : Probability to belong to class PLAN

"PROB_VEG" : Probability to belong to class VEG

"FIRST_LABE" : Class with the highest classification probability

"SEC_LABEL" : Class with the second highest classification probability

"FIRST_PROB" : Value of the highest classification probability

"SEC_PROB" : Value of the second highest classification probability

"UNCERTAIN" : Difference between "FIRST_PROB" and "SEC_PROB"

"BUILT_PERC" : Percentage of the street blocks covered by built-up (from land cover map)

"MAP_LABEL" : Final classification label with uncertainty and different density classes

Legend classes label:

"AGRI" : Agricultural vegetation

"VEG" : Natural vegetation

"BARE" : Bare soils

"ACS" : Non-residential built-up (administrative, commercial, services, etc.)

"PLAN" : Planned residential built-up

"PLAN_LD" : Planned residential low density built-up

"DEPR" : Deprived residential built-up

"UNCERT" : Uncertain classification

References:

[1] Grippa, Tais, 2018, "Mapping urban land use at street block level using OpenStreetMap, remote sensing data and spatial metrics", ISPRS Int. J. Geo-Inf. 2018, 7(7), 246. https://doi.org/10.3390/ijgi7070246 

[2] Grippa, Tais. 2018. “Osm Street Blocks Extraction.” Zenodo. https://doi.org/10.5281/zenodo.1290637.

Funding: 

This dataset was produced in the frame of two research project : MAUPP (http://maupp.ulb.ac.be) and REACT (http://react.ulb.be), funded by the Belgian Federal Science Policy Office (BELSPO).