Deep Learning for Slum Mapping in Remote Sensing Images: A Meta-analysis and Review

TL;DR

The paper surveys a decade of deep learning applied to slum mapping from remote sensing data, highlighting a shift from traditional texture and OBIA methods to CNN-based and hybrid architectures integrated with GIS. It uses a PRISMA-guided meta-analysis of 40 studies (2014–2024) to reveal regional concentrations (notably Asia and Africa) and data-source diversity (e.g., Sentinel, WorldView, Google Earth). Key contributions include documenting common modeling choices (U-Net variants, transfer learning, multimodal fusion), prevalent loss functions (weighted cross-entropy and Dice), and evaluation metrics (OA, IoU, F1, Kappa), along with practical guidance for data preparation and GIS deployment. The findings emphasize context-specific model selection, data quality, and ethical considerations, and suggest future directions such as synthetic data generation, advanced hybrid architectures, and closer integration with urban planning policy.

Abstract

The major Sustainable Development Goals (SDG) 2030, set by the United Nations Development Program (UNDP), include sustainable cities and communities, no poverty, and reduced inequalities. However, millions of people live in slums or informal settlements with poor living conditions in many major cities around the world, especially in less developed countries. To emancipate these settlements and their inhabitants through government intervention, accurate data about slum location and extent is required. While ground survey data is the most reliable, such surveys are costly and time-consuming. An alternative is remotely sensed data obtained from very high-resolution (VHR) imagery. With the advancement of new technology, remote sensing based mapping of slums has emerged as a prominent research area. The parallel rise of Artificial Intelligence, especially Deep Learning has added a new dimension to this field as it allows automated analysis of satellite imagery to identify complex spatial patterns associated with slums. This article offers a detailed review and meta-analysis of research on slum mapping using remote sensing imagery from 2014 to 2024, with a special focus on deep learning approaches. Our analysis reveals a trend towards increasingly complex neural network architectures, with advancements in data preprocessing and model training techniques significantly enhancing slum identification accuracy. We have attempted to identify key methodologies that are effective across diverse geographic contexts. While acknowledging the transformative impact Convolutional Neural Networks (CNNs) in slum detection, our review underscores the absence of a universally optimal model, suggesting the need for context-specific adaptations. We also identify prevailing challenges in this field, such as data limitations and a lack of model explainability and suggest potential strategies for overcoming these.

Figures (10)

• Figure 1: Geographic distribution of the proportion of urban populations residing in slums by country unhabitatslum. The color gradient indicates the percentage, with darker shades representing higher proportions. The absence of colors denotes the unavailability of data.
• Figure 2: Comprehensive Workflow of Deep Learning for Slum Mapping. This diagram highlights the essential stages in the deep learning process used for effective slum detection and analysis
• Figure 3: Architecture of a Convolutional Neural Network (CNN)
• Figure 4: Encoder-Decoder structure of an Autoencoder.
• Figure 5: PRISMA flow diagram depicting the systematic selection process of the articles.