Dar Es Salaam Very-High-Resolution Land Cover Map

This is a very-high-resolution land cover map of Dar es Salaam derived from satellite imagery (Pleiades, 0.5m resolution). The majority of the area is classified from a 2016 (July) image while a small part of it from two images collected in January and March 2018, respectively.

 The pixel values related to the following legend:

5=tree
8=shadow
3=artificial ground surface
4=low vegetation
2=water
7=bare ground
1=building
113=high elevated buildings
112=medium elevated buildings
111=low elevated buildings

The Out of Bag error of the product is 6,38% with the following class errors:

Building = 0.035826

Water = 0.049934

Artificial Ground Surface = 0.077108

Low Vegetation = 0.108709

Tall Vegetation = 0.062278

Bare Ground = 0.13803

Shadow = 0.019872

References:

[1] Grippa, Taïs, Moritz Lennert, Benjamin Beaumont, Sabine Vanhuysse, Nathalie Stephenne, and Eléonore Wolff. 2017. “An Open-Source Semi-Automated Processing Chain for Urban Object-Based Classification.” Remote Sensing 9 (4): 358. https://doi.org/10.3390/rs9040358.

[2] Grippa, Tais, Stefanos Georganos, Sabine G. Vanhuysse, Moritz Lennert, and Eléonore Wolff. 2017. “A Local Segmentation Parameter Optimization Approach for Mapping Heterogeneous Urban Environments Using VHR Imagery.” In Proceedings Volume 10431, Remote Sensing Technologies and Applications in Urban Environments II., edited by Wieke Heldens, Nektarios Chrysoulakis, Thilo Erbertseder, and Ying Zhang, 20. SPIE. https://doi.org/10.1117/12.2278422.

[3] Georganos, Stefanos, Taïs Grippa, Moritz Lennert, Sabine Vanhuysse, and Eleonore Wolff. 2017. “SPUSPO: Spatially Partitioned Unsupervised Segmentation Parameter Optimization for Efficiently Segmenting Large Heterogeneous Areas.” In Proceedings of the 2017 Conference on Big Data from Space (BiDS’17).

This dataset was produced in the frame of  REACT (http://react.ulb.be), funded by the Belgian Federal Science Policy Office (BELSPO).