Abstract
Mapping Earth’s surface and its rapid changes with remotely sensed data is a crucial task to understand the impact of an increasingly urban world population on the environment. However, the impressive amount of available Earth observation data is only marginally exploited in common classifications. In this study, we use the computational power of Google Earth Engine and Google Cloud Platform to generate an oversized feature set in which we explore feature importance and analyze the influence of dimensionality reduction methods to object-based land cover classification with Support Vector Machines. We propose a methodology to extract the most relevant features and optimize an SVM classifier hyperparameters to achieve higher classification accuracy. The proposed approach is evaluated in two different urban study areas of Stockholm and Beijing. Despite different training set sizes in the two study sites, the averaged feature importance ranking showed similar results for the top-ranking features. In particular, Sentinel-2 NDVI, NDWI, and Sentinel-1 VV temporal means are the highest ranked features and the experiment results strongly indicated that the fusion of these features improved the separability between urban land cover and land use classes. Overall classification accuracies of 94% and 93% were achieved in Stockholm and Beijing study sites, respectively. The test demonstrated the viability of the methodology in a cloud-computing environment to incorporate dimensionality reduction as a key step in the land cover classification process, which we consider essential for the exploitation of the growing Earth observation big data. To encourage further research and development of reliable workflows, we share our datasets and publish the developed Google Earth Engine and Python scripts as free and open-source software.
Highlights
Mapping Earth’s surface and its rapid changes with remotely sensed data are a crucial task to help understand the impact of an increasingly urban world population on the environment
The test demonstrated the viability of the methodology in a cloud-computing environment to incorporate dimensionality reduction as a key step in the land cover classification process, which we consider essential for the exploitation of the growing Earth observation big data
In the last couple decades, several studies have investigated the classification of urban scenes using remote sensing data
Summary
Mapping Earth’s surface and its rapid changes with remotely sensed data are a crucial task to help understand the impact of an increasingly urban world population on the environment. The successful use of remote sensing data for urban scene classification depends on several considerations (e.g., image spatial resolution, acquisition time, spectral information, classification scheme, etc.). Another factor of crucial importance and that has strong impact on the classification accuracy, in general, is the input features and their quality. The training samples are separated in the new space by a hyperplane (defined by the support vectors) that guarantee the largest margin between the classes It has been used successfully in different types of remote sensing applications—for example, classification [47,48,49], change detection [50,51,52], and in image fusion [53]. A further analysis of the feature selection allows us to evaluate the importance of individual features to the classification task
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have