Feature Extraction and Selection of Sentinel-1 Dual-Pol Data for Global-Scale Local Climate Zone Classification

Abstract

The concept of the local climate zone (LCZ) has been recently proposed as a generic land-cover/land-use classification scheme. It divides urban regions into 17 categories based on compositions of man-made structures and natural landscapes. Although it was originally designed for temperature study, the morphological structure concealed in LCZs also reflects economic status and population distribution. To this end, global LCZ classification is of great value for worldwide studies on economy and population. Conventional classification approaches are usually successful for an individual city using optical remote sensing data. This paper, however, attempts for the first time to produce global LCZ classification maps using polarimetric synthetic aperture radar (PolSAR) data. Specifically, we first produce polarimetric features, local statistical features, texture features, and morphological features and compare them, with respect to their classification performance. Here, an ensemble classifier is investigated, which is trained and tested on already separated transcontinental cities. Considering the challenging global scope this work handles, we conclude the classification accuracy is not yet satisfactory. However, Sentinel-1 dual-Pol SAR data could contribute the classification for several LCZ classes. According to our feature studies, the combination of local statistical features and morphological features yields the best classification results with 61.8% overall accuracy (OA), which is 3% higher than the OA produced by the second best features combination. The 3% is considerably large for a global scale. Based on our feature importance analysis, features related to VH polarized data contributed the most to the eventual classification result.

The Human Influence Experiment (Part 2): Guidelines for Improved Mapping of Local Climate Zones Using a Supervised Classification

“…Second, due to the fact that less then 10% of the participants lived in Berlin or were familiar with the city, it was not possible to investigate the influence of local knowledge on the classification results. Even though, currently research is done investigating the potential of continental-scale LCZ maps [26][27][28], the question on local knowledge remains important. In this respect, it might be better to include more cities.…”

Revealing Kunming’s (China) Historical Urban Planning Policies Through Local Climate Zones

“…Qui et al (2018) [35,36] particularly focused on data/feature choice including optical data such as Landsat-8 and Sentinel-2, as well as additional datasets such as the Global Urban Footprint (GUF), the OpenStreetMap (OSM) layers buildings and land use, and the Visible Infrared Imager Radiometer Suite (VIIRS)-based Nighttime Light (NTL). [37,38], on the other hand, explored the potential of Sentinel-1 polarimetric SAR data for LCZ classification. Despite the rapid advancements in spatio-temporal LCZ mapping, to date, the large number of LCZ maps created for cities and regions around the world (see, e.g., references in [28,33,35,[39][40][41]) are typically static in time.…”

OpenSARUrban: A Sentinel-1 SAR Image Dataset for Urban Interpretation

“…The categories related to this study are mostly man-made structures, except for the category of vegetation. It has been demonstrated that VH-polarized data contributes the most to the classification of man-made structures [67]. The second viewpoint is to explain the phenomenon from the perspective of visual appearance.…”