Machine learning algorithms such as Random Forest (RF) are being increasingly applied on traditionally geographical topics such as population estimation. Even though RF is a well performing and generalizable algorithm, the vast majority of its implementations are still ‘aspatial’ and may not address spatial heterogenous processes. At the same time, remote sensing (RS) data which are commonly used to model population can be highly spatially heterogeneous. From this scope, we present a novel geographical implementation of RF, named Geographical Random Forest (GRF) as both an predictive and exploratory tool to model population as a function of RS covariates. GRF is a disaggregation of RF into geographical space in the form of local sub-models. From the first empirical results, we conclude that GRF can be more predictive when an appropriate spatial scale is selected to model the data, with reduced residual autocorrelation and lower Root Mean Square Error and Mean Absolute Error values. Finally, and of equal importance, GRF can be used as an effective exploratory tool to visualize the relationship between dependent and independent variables, highlighting interesting local variations and allowing for a better understanding of the processes that may be causing the observed spatial heterogeneity.

In this letter, the recently developed extreme gradient boosting (Xgboost) classifier is implemented in a very high resolution (VHR) object-based urban land use-land cover application. In detail, we investigated the sensitivity of Xgboost to various sample sizes, as well as to feature selection (FS) by applying a standard technique, correlation-based FS. We compared Xgboost with benchmark classifiers such as random forest (RF) and support vector machines (SVMs). The methods are applied to VHR imagery of two sub-Saharan cities of Dakar and Ouagadougou and the village of Vaihingen, Germany. The results demonstrate that Xgboost parameterized with a Bayesian procedure, systematically outperformed RF and SVM, mainly in larger sample sizes.

This study evaluates the impact of four feature selection (FS) algorithms in an object-based image analysis framework for very-high-resolution land use-land cover classification. The selected FS algorithms, correlation-based feature selection, mean decrease in accuracy, random forest (RF) based recursive feature elimination, and variable selection using random forest, were tested on the extreme gradient boosting, support vector machine, K-nearest neighbor, RF, and recursive partitioningclassifiers, respectively. The results demonstrate that the selection of an appropriate FS method can be crucial to the performance of a machine learning classifier in terms of accuracy but also parsimony. In this scope, we propose a new metric to perform model selection named classification optimization score (COS) that rewards model simplicity and indirectly penalizes for increased computational time and processing requirements using the number of features for a given classification model as a surrogate. Our findings suggest that applying rigorous FS along with utilizing the COS metric may significantly reduce the processing time and the storage space while at the same time producing higher classification accuracy than using the initial dataset.

Up-to-date and reliable land-use information is essential for a variety of applications such as planning or monitoring of the urban environment. This research presents a workflow for mapping urban land use at the street block level, with a focus on residential use, using very-high resolution satellite imagery and derived land-cover maps as input. We develop a processing chain for the automated creation of street block polygons from OpenStreetMap and ancillary data. Spatial metrics and other street block features are computed, followed by feature selection that reduces the initial datasets by more than 80%, providing a parsimonious, discriminative, and redundancy-free set of features. A random forest (RF) classifier is used for the classification of street blocks, which results in accuracies of 84% and 79% for five and six land-use classes, respectively. We exploit the probabilistic output of RF to identify and relabel blocks that have a high degree of uncertainty. Finally, the thematic precision of the residential blocks is refined according to the proportion of the built-up area. The output data and processing chains are made freely available. The proposed framework is able to process large datasets, given that the cities in the case studies, Dakar and Ouagadougou, cover more than 1000 km² in total, with a spatial resolution of 0.5 m.

This study presents the development of a semi-automated processing chain for urban object-based land-cover and land-use classification. The processing chain is implemented in Python and relies on existing open-source software GRASS GIS and R. The complete tool chain is available in open access and is adaptable to specific user needs. For automation purposes, we developed two GRASS GIS add-ons enabling users (1) to optimize segmentation parameters in an unsupervised manner and (2) to classify remote sensing data using several individual machine learning classifiers or their prediction combinations through voting-schemes. We tested the performance of the processing chain using sub-metric multispectral and height data on two very different urban environments: Ouagadougou, Burkina Faso in sub-Saharan Africa and Liège, Belgium in Western Europe. Using a hierarchical classification scheme, the overall accuracy reached 93% at the first level (5 classes) and about 80% at the second level (11 and 9 classes, respectively).