Synthetic aperture radar (SAR) imagery is invaluable for assessing disaster-induced changes due to its capacity to capture detailed surface information despite varying environmental conditions. However, the scarcity of high-resolution (HR) SAR imagery before disasters presents significant challenges for accurately recognizing changes in damaged buildings, particularly in scenarios requiring pre- and post-disaster image pairs for machine-learning methods that rely on large samples. To address this challenge, our study proposes an innovative solution utilizing simulated SAR imagery generated through ray tracing-based SAR simulation, we generated high-quality predisaster SAR images that closely replicate the scattering properties of authentic SAR imagery using the same sensor orientation as postdisaster image. In this study, to evaluate the feasibility of using simulated SAR images for deep learning-based building damage classification, we investigate three different scenarios: 1) authentic pre- and post-disaster image pairs; 2) simulated predisaster images and authentic postdisaster images; and 3) postdisaster images alone. The methodology was applied in Mashiki, Japan, which was heavily impacted by the 2016 Kumamoto earthquake. Our classification results showed that the simulated predisaster SAR data produced outcomes comparable with those of using authentic image pairs and were clearly superior to the approach that utilized only postevent images. These findings illustrate that simulated SAR imagery is a reliable alternative when authentic predisaster data are unavailable, enabling fast, and accurate damage assessments to support emergency decision-making.

Urban areas in Lima, Peru, grow fast and with almost no control. Furthermore, resources to monitor urban areas are limited in Peru. Thus, the evolution of exposure of urban areas to natural hazards is highly uncertain. In this paper, we report a practical use of deep learning-based land use and land cover mapping to quantify, with unprecedented accuracy, the increment in exposure against floods in the District of Ate, Lima, Peru. We use Google Satellite imagery recorded in 2002 and 2023, generate land use maps, and quantify the increment of urban areas in the floodplain of the Rimac River. Results show an increment of exposure to floods of about 10.3% after 21 years.

The lack of pre-disaster data often poses challenges for reliable building damage predictions in disaster response using remote sensing. To address this limitation, we propose a framework that leverages Synthetic Aperture Radar (SAR) simulators to generate high-resolution simulated SAR images. Assessing the realism of these simulated images is crucial for their reliability in representing authentic SAR data. However, obtaining pre-disaster SAR images for city-scale areas can be challenging.In this study, we utilized Google 3D Tiles and Blender-GIS model to recreate scenes for simulating pre-disaster SAR images. We conducted a comprehensive analysis, evaluating the similarity between simulated and authentic SAR images through similarity assessments. Preliminary results suggest that the SAR simulator successfully produces distinct signatures for different 3D models in simulated SAR imagery. This approach holds promise for use in disaster response scenarios.

Remote sensing and deep learning have been integrated to solve multiple problems, including building damage assessment. With rapid development in both fields, deep learning and remote sensing can play a greater role in damage mapping, specifically in rapid damage assessment, to support emergency response efforts. Deep learning model evaluation is generally based on a statistical split separating training and testing sets of the same data distribution. Although this enables the evaluation of the model performance, this scheme does not disclose the ability of the model to perform in data obtained from different distributions, which is often the case in real-context disaster emergency response. This study evaluates the model generalization in emergency response scenarios. The results show that the current deep learning model has a high performance in in-domain testing yet experiences a drop of up to 53(%) in F1in realistic applications. Future studies should focus on enhancing the model transferability, including using domain adaptation techniques and harnessing multi-modal features.

Deep learning models are generally less able to maintain their performance in out-of-domain testing. Model transferability is crucial, especially when a model needs to be applied to a new dataset, such as in disaster emergency response, where the training samples are scarce. To solve the aforementioned issues, we propose a semi-supervised framework to improve model generalization by utilizing unlabelled samples from the target domain. The framework consists of two main steps: model initialization, which incorporates past events, and iterative fine-tuning. The latter step relies heavily on the pseudo-labels inferred with high confidence from the former step. We tested our framework on the 2024 Noto Peninsula Earthquake. Our framework shows an improvement in the model generalization indicated by higher scores in the tuned model compared to the initial model. The effect is even greater when the local context from the past event is included in the initial learning step. In this case, the score has increased by about 21% from 0.62 to 0.75. The proposed framework offers a promising solution for rapid disaster damage mapping.

After a major disaster, a rapid assessment of building damage is highly required for emergency response and prompt recovery. Remote sensing technologies have been widely applied for building damage mapping. Combining machine-learning algorithms (e.g., deep learning) and satellite images has recently demonstrated success in boosting damage recognition methods. Although previous techniques have shown great success, they primarily adopt supervised settings, often requiring a minimum number of training samples to achieve acceptable accuracy. Moreover, previous methods also are developed for specific target areas, which makes it challenging to apply them to other regions in case of future disasters. This paper presents a novel unsupervised approach for building damage mapping, focusing on collapsed structures, using modern convolutional neural network (CNN) models and high-resolution remote sensing imagery. We apply our mapping framework to revise the building damage following the 2007 Peru-Pisco Earthquake and the recent 2023 Turkey and Syria Earthquakes.

Current flood mapping methods combine remote sensing and machine learning technologies to estimate the inundation area. Although these methods have shown great success, they mainly focus on the flood extent without additional information on the inundation depth. However, knowing the inundation level can significantly benefit first responders and rescue efforts. Recent advances in machine learning have boosted the development of advanced methods for disaster management. This paper integrates modern convolutional neural network (CNN) models and physics-based numerical simulation to develop a novel framework for automatic flood inundation depth mapping. Our framework builds a synthetic training dataset using numerical flood simulation in four geographical regions. Then, it trains CNN models to understand the nonlinear relationship between inundation depth and topographic features. Our experiments, designed to evaluate the strength of our methodology in a real-world application, demonstrate that it can estimate flood depth with acceptable accuracy (Root-Mean-Squared Error=0.2) in unseen areas during training. These results indicate that a worldwide flood inundation mapping could be achieved by including key areas with representative topographic features.

We introduce OpenEarthMap, a benchmark dataset, for global high-resolution land cover mapping. OpenEarth-Map consists of 2.2 million segments of 5000 aerial and satellite images covering 97 regions from 44 countries across 6 continents, with manually annotated 8-class land cover labels at a 0.25–0.5m ground sampling distance. Se-mantic segmentation models trained on the OpenEarth-Map generalize worldwide and can be used as off-the-shelf models in a variety of applications. We evaluate the performance of state-of-the-art methods for unsupervised domain adaptation and present challenging problem settings suitable for further technical development. We also investigate lightweight models using automated neural architecture search for limited computational resources and fast mapping. The dataset is available at https: //open-earth-map.org.

We propose a framework that estimates the inundation depth (maximum water level) and debris-flow-induced topographic deformation from remote sensing imagery by integrating deep learning and numerical simulation. A water and debris-flow simulator generates training data for various artificial disaster scenarios. We show that regression models based on Attention U-Net and LinkNet architectures trained on such synthetic data can predict the maximum water level and topographic deformation from a remote sensing-derived change detection map and a digital elevation model. The proposed framework has an inpainting capability, thus mitigating the false negatives that are inevitable in remote sensing image analysis. Our framework breaks limits of remote sensing and enables rapid estimation of inundation depth and topographic deformation, essential information for emergency response, including rescue and relief activities. We conduct experiments with both synthetic and real data for two disaster events that caused simultaneous flooding and debris flows and demonstrate the effectiveness of our approach quantitatively and qualitatively. Our code and data sets are available athttps://github.com/nyokoya/dlsim.

Disasters such as earthquakes, hurricanes, and flooding are responsible for large-scale infrastructure damages and loss of human lives. Immediately after disaster strikes, one of the most critical and difficult tasks is accurately assessing the extent and severity of the disaster. This task is especially challenging in areas isolated by the disaster; in such cases, remote sensing information provides the best alternative to tackle this problem. This paper presents a damage mapping framework using remote sensing imagery acquired from previous disasters. The proposed deep learning-based framework is trained to learn features related to building damage using imagery from previous disasters that were collected from different regions around the world. Then, it is tested to recognize damage from a different urban environment.

This paper presents a novel application of remote sensing data and machine learning technologies for damage classification in a real-world cross-domain application. The proposed methodology trains models to learn the building damage characteristics recorded in the 2011 Tohoku Tsunami from multi-sensor and multi-temporal remote sensing images. Then, the trained models are tested in the recent 2018 Sulawesi Tsunami. Additionally, a simulation of high-resolution SAR image was carried to deal with missing data modality. Our initial results show that the ResNet-derived features from optical images acquired after the disaster together with moderate- and high-resolution synthetic aperture radar (SAR) post-event intensity data showed significant accuracy in classifying two levels of tsunami-induced damage, with an average f-score of approximately 0.72. Taking into account that no training data from the 2018 Sulawesi Tsunami was used, our methodology shows excellent potential for future implementation of a rapid response system based on a database of building damage constructed from previous majors disasters.

We present a nonlocal low-rank denoising algorithm for synthetic aperture radar (SAR) image stacks. The method extends the widely known DespecKS algorithm by integrating low-rank approximation, outlier removal, and total variation (TV) regularization into the estimation process. Preliminary experiments shows increased robustness against outliers and comparable performance to state-of-the-art stack despeckling algorithms.

This paper presents building damage mapping based on transfer learning techniques. Due to the different spatial resolutions of optical (WorldView, 0.5m) and SAR (Sentinel-1, 10m), we adopt different methods: pixel-level for moderate-resolution SAR images, and patch-level for very high-resolution optical images. For SAR images, the performance of fast unsupervised transfer learning methods, such as overall centroid alignment (OCA) and CORrelation ALignment (CORAL), are investigated. For the optical images, two public databases are used to predict the building damage mapping of Palu with WorldView-3 images via ResNet50. Experimental results indicate the effectiveness of transfer learning on the building damage mapping using different data sources.

On September 19, 2017, the Mw7.1 Puebla Earthquake caused significant destruction in several cities in central Mexico. In this paper, two pre- and one post-event ALOS2-PALSAR2 data were used to detect the damaged area around Izucar de Matamoros town in Mexico. First, we identify the built-up areas using pre-event data. Second, we evaluate the earthquake-induced damage areas using an RGB color-coded image constructed from the pre- and co-event coherence images. Our analysis showed that the green and red bands display a great potential to discriminate the damaged areas.

Near real-time building damage mapping is an indispensable prerequisite for governments to make decisions for disaster relief. With high-resolution synthetic aperture radar (SAR) systems, such as TerraSAR-X, the provision of such products in a fast and effective way becomes possible. In this letter, a deep learning-based framework for rapid regional tsunami damage recognition using post-event SAR imagery is proposed. To perform such a rapid damage mapping, a series of tile-based image split analysis is employed to generate the data set. Next, a selection algorithm with the SqueezeNet network is developed to swiftly distinguish between built-up (BU) and nonbuilt-up regions. Finally, a recognition algorithm with a modified wide residual network is developed to classify the BU regions into wash away, collapsed, and slightly damaged regions. Experiments performed on the TerraSAR-X data from the 2011 Tohoku earthquake and tsunami in Japan show a BU region extraction accuracy of 80.4% and a damage-level recognition accuracy of 74.8%, respectively. Our framework takes around 2 h to train on a new region, and only several minutes for prediction.

In this study, a practical methodology was presented to map damaged buildings using high resolution synthetic aperture radar (SAR) images and post-event building damage data from the 2013 Super Typhoon Haiyan, in Tacloban city, the Philippines. To detect destroyed structures, we focused on the changes in the radar signal within footprints of buildings between pre- and post-event SAR images. The method was tested using a 1.0 m resolution COSMO-SkyMed SAR images taken over Tacloban city, the Philippines. The method proves, with 73% accuracy in this case, to be suitable for estimating destroyed buildings.

In this study, the extent of the flooded areas by the Super Typhoon Haiyan in the Philippines were extracted using ASTER VNIR images taken over Tacloban city in the Visayas. In order to constraint the affected area, we employed the normalize difference vegetation and water indices (NDVI and NDWI) from the pre- and post-event images. The extension of the flooded area was determined by comparing the index characteristics before and after the event. A phase-based change detection method indices was applied to classify the affected area into three classes according to the changes between the pre- and post-images. Through NDWI the flooded areas were detected despite the moderate resolution of ASTER images. In addition, the phase-based analysis successfully detected level of change within the affected area that may be correlated to the damage observed on field surveys. The results from the phase-based analysis were verified with damage levels obtained through visual damage inspection using high resolution satellite images.

A strong typhoon “Haiyan” affected Southeast Asia on November 8, 2013, caused gigantic destruction in the Philippines. In this study, two pre- and one post-event COSMO-SkyMed SCSB data were used to detect the damaged area around Tacloban City, Leyte Island. First, the severe damaged areas were detected according to the difference between the pre- and post-event speckle divergence values. Then the pre- and co-event coherence (NDCI) and correlation coefficient (NDCOI) were calculated from the three temporal data. The relationships between the four building damage levels and NDCI or NDCOI value were obtained by introducing the visual interoperation result. Using this relationship, the possibility of each damage class was estimated in the whole urban area.