Context-Aware Slum Mapping in Sub-Saharan Africa Using Sentinel-1 Texture and Local Climate Zones

Jul 16, 2026

Accurate mapping of informal settlements in Sub-Saharan African cities is constrained by optical imagery failing to separate Informal Settlements (LCZ 7) from spectrally similar formal Compact Low-Rise areas (LCZ 3). The study introduces a context-aware, reproducible Optical–SAR framework that fuses Sentinel-2 spectral cues with Sentinel-1 structural signals using an adapted LCZ taxonomy. A three-tier SAR strategy—calibrated backscatter, GLCM textures, and physics-guided features—improves delineation across Nairobi and Eldoret.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

Introduction

Study goal and motivation

Why optical-only fails in SSA

SAR as structural contrast

Proposed framework

Three-tier Optical–SAR fusion

Feature engineering and texture metrics

Experiments and evaluation

Datasets and reference areas

Stratified hold-out and season-aware ablation

Results and findings

Accuracy by season

Confusion reduction (LCZ 3 vs LCZ 7)

Cross-city transfer considerations

Context-Aware Slum Mapping in Sub-Saharan Africa Using Sentinel-1 Texture and Local Climate Zones Peterson Chepkilot, Student Member, IEEE, Babak Memar, Student Member, IEEE, and Paolo Gamba, Fellow, IEEE arXiv :2607 .07532v 1 [ cs .CV] 8 Jul 2026 Abstract—Accurate mapping of informal settlements remains a major challenge in Sub-Saharan African (SSA) cities because optical imagery often fails to distinguish Informal Settlements (defined here as LCZ 7) from spectrally similar formal Compact Low-Rise areas (LCZ 3). This study presents a context-aware, reproducible Optical–SAR framework that improves informal settlement delineation using Sentinel-2 spectral features and Sentinel-1 structural information within an adapted Local Climate Zone (LCZ) taxonomy. We implement a three-tier SAR integration strategy: calibrated backscatter, GLCM textures, and a physics-guided feature engineered to capture the high structural disorder and weak radar return characteristic of SSA informal settlements. Using reference data across Nairobi and Eldoret (Kenya), we evaluate performance via a stratified hold-out protocol and a season-aware ablation study. Results show that SAR textures provide the dominant performance gain for LCZ 7 detection. The Optical–SAR model achieves overall accuracy of 0.816 (dry) and 0.807 (wet), significantly outperforming the WUDAPT baseline (OA 0.704) and reducing the critical LCZ 3 ↔ LCZ 7 confusion to ∼7% . Seasonal analysis indicates that while optical-only separability varies with phenology, SAR-derived textures stabilize informal settlement mapping across seasons. These findings demonstrate that the incorporation of SAR-derived features yields consistent improvements for urban morphology mapping in data-scarce environments across seasons and across the evaluated source cities, while cross-city transfer remains limited without local adaptation strategies. Index Terms—Informal settlements, slum mapping, Local Climate Zones (LCZ), Sentinel-1 SAR, SAR texture, Sentinel-2, Optical–SAR fusion, WUDAPT, Sub-Saharan Africa. I. INTRODUCTION RAPID urbanization across Sub-Saharan Africa (SSA) is generating highly heterogeneous urban fabrics characterized by the coexistence of formal planned neighborhoods and extensive informal settlements. Reliable urban morphology mapping at fine spatial scales is therefore essential for infrastructure planning, environmental monitoring, and comparative urban research. The Local Climate Zone (LCZ) framework [1] has become the de facto standard for describing urban form in a climatically meaningful and internationally comparable manner. Despite its global adoption, LCZ mapping performance degrades in many SSA cities due to morphological complexity and material heterogeneity that challenge optical-only Peterson Chepkilot is with the Department of Civil, Building and Environmental Engineering, Sapienza University of Rome, 00184 Rome, Italy (e-mail: [petersonkipkurui.chepkilot@uniroma1.it](petersonkipkurui.chepkilot@uniroma1.it)). Babak Memar and Paolo Gamba are with the Department of Electrical, Computer and Biomedical Engineering, University of Pavia, 27100 Pavia, Italy (e-mail: [babak.memar.it@gmail.com](babak.memar.it@gmail.com); [paolo.gamba@unipv.it](paolo.gamba@unipv.it)). approaches. A fundamental limitation is the persistent classification confusion between Compact Low-Rise (LCZ 3) and Lightweight Low-Rise (LCZ 7), the latter typically corresponding to informal settlements. LCZ 7 areas are predominantly characterized by dense, low-rise structures with corrugated iron roofing, often in varying states of oxidation. In contrast, LCZ 3 areas more frequently include asbestos sheets, concrete slabs, tile roofing, and occasionally metal roofs arranged in more regular block configurations. Although construction typologies differ, spatial aggregation at Sentinel-2 resolution together with seasonal compositing reduces effective spectral separability between LCZ 3 and

FAQs

What problem does the study address in Sub-Saharan Africa urban mapping?

It addresses difficulty in accurately distinguishing informal settlements (LCZ 7) from spectrally similar formal Compact Low-Rise areas (LCZ 3) using optical imagery alone.

How does the proposed framework fuse optical and SAR data?

It uses an adapted LCZ taxonomy and a three-tier SAR integration strategy combining calibrated backscatter metrics, Sentinel-1 GLCM texture features, and a physics-guided feature engineered to capture structural disorder and weak radar return.

Which component provides the main performance improvement and how is it evaluated?

SAR textures provide the dominant gain for LCZ 7 detection. Performance is assessed using stratified hold-out validation on reference data from Nairobi and Eldoret, plus a season-aware ablation study.

Jul 16, 2026

Upload by Evangeline