Mapping of Dwellings in IDP/Refugee Settlements from Very High-Resolution Satellite Imagery Using a Mask Region-Based Convolutional Neural Network

Abstract

Earth-observation-based mapping plays a critical role in humanitarian responses by providing timely and accurate information in inaccessible areas, or in situations where frequent updates and monitoring are required, such as in internally displaced population (IDP)/refugee settlements. Manual information extraction pipelines are slow and resource inefficient. Advances in deep learning, especially convolutional neural networks (CNNs), are providing state-of-the-art possibilities for automation in information extraction. This study investigates a deep convolutional neural network-based Mask R-CNN model for dwelling extractions in IDP/refugee settlements. The study uses a time series of very high-resolution satellite images from WorldView-2 and WorldView-3. The model was trained with transfer learning through domain adaptation from nonremote sensing tasks. The capability of a model trained on historical images to detect dwelling features on completely unseen newly obtained images through temporal transfer was investigated. The results show that transfer learning provides better performance than training the model from scratch, with an MIoU range of 4.5 to 15.3%, and a range of 18.6 to 25.6% for the overall quality of the extracted dwellings, which varied on the bases of the source of the pretrained weight and the input image. Once it was trained on historical images, the model achieved 62.9, 89.3, and 77% for the object-based mean intersection over union (MIoU), completeness, and quality metrics, respectively, on completely unseen images.

Self-Supervised Variational Autoencoder for Unsupervised Object Counting from Very-High-Resolution Satellite Imagery: Applications in Dwelling Extraction in FDP Settlement Areas

“…Currently, deep learning models have shown great performance in classification [10], segmentation [11], [12] and object detection tasks [13]- [15] which paves the way for automatic information retrieval pipelines. Benefiting from advances in deep learning for computer vision, there are promising works dedicated to dwelling extraction from temporary settlements for humanitarian emergency response [16]- [21]. Despite the strong performance of supervised computer vision models in various fields, they have known limitations that constrain their full-fledged usage in operational humanitarian emergency response.…”

Unsupervised Domain Adaptation for Instance Segmentation: Extracting Dwellings in Temporary Settlements Across Various Geographical Settings

“…Though these developments resulted in the generation of global building footprint datasets [14], [15], FDP settlements are still less represented in terms of geographic coverage and provision of information with detailed spatio-temporal granularity [16]. As a result, recent promising works have focused on dwelling extraction in temporary settlements [17], [18], [19], [20], [21], building extraction in complex urban settings for humanitarian applications [22], and FDP settlement densification and spatial dynamic analysis [1].…”

Impact of Digital Twins and Metaverse on Cities: History, Current Situation, and Application Perspectives

“…The comprehensive development of the digital industry and the use of data as new energy is an important means to change the dependence of economic growth on natural energy. The combination of digitalization with energy-saving and low-carbon technologies is the focus, which can effectively promote the rational utilization of urban resources [11,12]. Digital Twins (DTs) cities make full use of government networks, data exchange platforms, and modern information technology to give full play to the monitoring and screening ability, business operation guidance ability, information transparency, collaborative support, and decision support of information systems.…”