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

Anjali Raj,Adway Mitra,Manjira Sinha

2024-06-12

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.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is: how to use deep - learning techniques to accurately identify and map slums (or informal settlements) from high - resolution remote - sensing images, in order to provide reliable data to support urban planning and policy - making, thereby improving the living conditions of slum residents.

Specifically, this review article explores the following aspects:

1. **Challenges in data acquisition**: Although traditional ground surveys are reliable, they are costly and time - consuming. Therefore, a more efficient and cost - effective method is needed to obtain the spatial distribution information of slums.

2. **The need for technological progress**: With the progress of remote - sensing technologies such as satellites and unmanned aerial vehicles, as well as the development of artificial intelligence, especially deep learning, how to apply these new technologies to the identification and classification of slums is an important issue.

3. **Model selection and optimization**: Although deep - learning models such as convolutional neural networks (CNNs) perform well in slum detection, there is no universal best model. Therefore, it is necessary to study effective model architectures in different geographical contexts and explore how to adjust them for specific situations.

4. **Data limitations and model interpretability**: Current challenges include issues such as insufficient data and difficult - to - interpret models. To this end, strategies to overcome these obstacles need to be proposed.

By solving the above problems, this research aims to provide scientific basis and technical support for achieving the United Nations Sustainable Development Goals (SDGs), especially Goal 11 (Sustainable Cities and Communities), Goal 1 (No Poverty), and Goal 3 (Good Health and Well - being).