BDD-Net: A General Protocol for Mapping Buildings Damaged by a Wide Range of Disasters Based on Satellite Imagery

The timely and accurate recognition of damage to buildings after destructive disasters is one of the most important post-event responses. Due to the complex and dangerous situations in affected areas, field surveys of post-disaster conditions are not always feasible. The use of satellite imagery for disaster assessment can overcome this problem. However, the textural and contextual features of post-event satellite images vary with disaster types, which makes it difficult to use models that have been developed for a specific disaster type to detect damaged buildings following other types of disasters. Therefore, it is hard to use a single model to effectively and automatically recognize post-disaster building damage for a broad range of disaster types. Therefore, in this paper, we introduce a building damage detection network (BDD-Net) composed of a novel end-to-end remote sensing pixel-classification deep convolutional neural network. BDD-Net was developed to automatically classify every pixel of a post-disaster image into one of non-damaged building, damaged building, or background classes. Pre- and post-disaster images were provided as input for the network to increase semantic information, and a hybrid loss function that combines dice loss and focal loss was used to optimize the network. Publicly available data were utilized to train and test the model, which makes the presented method readily repeatable and comparable. The protocol was tested on images for five disaster types, namely flood, earthquake, volcanic eruption, hurricane, and wildfire. The results show that the proposed method is consistently effective for recognizing buildings damaged by different disasters and in different areas.

Download Full-text

Post-disaster assessment of landslides in southern Taiwan after 2009 Typhoon Morakot using remote sensing and spatial analysis

Natural Hazards and Earth System Science ◽

10.5194/nhess-10-2179-2010 ◽

2010 ◽

Vol 10 (10) ◽

pp. 2179-2190 ◽

Cited By ~ 66

Author(s):

F. Tsai ◽

J.-H. Hwang ◽

L.-C. Chen ◽

T.-H. Lin

Keyword(s):

Spatial Analysis ◽

Change Detection ◽

Satellite Images ◽

Regional Scale ◽

Typhoon Morakot ◽

Change Detection Analysis ◽

Detection Analysis ◽

Disaster Assessment ◽

Southern Taiwan ◽

Post Disaster

Abstract. On 8 August 2009, the extreme rainfall of Typhoon Morakot triggered enormous landslides in mountainous regions of southern Taiwan, causing catastrophic infrastructure and property damages and human casualties. A comprehensive evaluation of the landslides is essential for the post-disaster reconstruction and should be helpful for future hazard mitigation. This paper presents a systematic approach to utilize multi-temporal satellite images and other geo-spatial data for the post-disaster assessment of landslides on a regional scale. Rigorous orthorectification and radiometric correction procedures were applied to the satellite images. Landslides were identified with NDVI filtering, change detection analysis and interactive post-analysis editing to produce an accurate landslide map. Spatial analysis was performed to obtain statistical characteristics of the identified landslides and their relationship with topographical factors. A total of 9333 landslides (22 590 ha) was detected from change detection analysis of satellite images. Most of the detected landslides are smaller than 10 ha. Less than 5% of them are larger than 10 ha but together they constitute more than 45% of the total landslide area. Spatial analysis of the detected landslides indicates that most of them have average elevations between 500 m to 2000 m and with average slope gradients between 20° and 40°. In addition, a particularly devastating landslide whose debris flow destroyed a riverside village was examined in depth for detailed investigation. The volume of this slide is estimated to be more than 2.6 million m3 with an average depth of 40 m.

Download Full-text

Post-Disaster Urban Recovery Monitoring in Pisco After the 2007 Peru Earthquake Using Satellite Image

Journal of Disaster Research ◽

10.20965/jdr.2014.p1059 ◽

2014 ◽

Vol 9 (6) ◽

pp. 1059-1068 ◽

Cited By ~ 11

Author(s):

Tomoyo Hoshi ◽

◽

Osamu Murao ◽

Kunihiko Yoshino ◽

Fumio Yamazaki ◽

...

Keyword(s):

Satellite Imagery ◽

Urban Areas ◽

Satellite Image ◽

Recovery Process ◽

Visual Interpretation ◽

Field Surveys ◽

Recovery Processes ◽

The City ◽

Post Disaster

Pisco was the area most damaged by the 2007 Peru earthquake. The purpose of this research is to develop possibilities of using satellite imagery to monitor postdisaster urban recovery processes, focusing on the urban change in Pisco between 2007 and 2011. To this end, the authors carried out field surveys in the city in 2012 and 2013 and also examined previous surveys to determine that building reconstruction peaked between 2008 and 2009. After analyzing the five-year recovery process, the authors compared its reconstruction conditions by visual interpretation with those by image analysis using satellite image. An accuracy of 71.2% was achieved for the visual interpretation results in congested urban areas, and that for developed districts was about 60%. The result shows that satellite imagery can be a useful tool for monitoring and understanding post-disaster urban recovery processes in the areas in which conducting long-term field survey is difficult.

Download Full-text

Developing fragility functions for the areas affected by the 2009 Samoa earthquake and tsunami

Natural Hazards and Earth System Sciences Discussions ◽

10.5194/nhessd-2-1-2014 ◽

2014 ◽

Vol 2 (1) ◽

pp. 1-25

Author(s):

H. Gokon ◽

S. Koshimura ◽

K. Imai ◽

M. Matsuoka ◽

Y. Namegaya ◽

...

Keyword(s):

Satellite Images ◽

Hydrodynamic Force ◽

Deep Ocean ◽

Building Damage ◽

Fragility Functions ◽

Tsunami Propagation ◽

Tsunami Inundation ◽

Field Surveys ◽

High Resolution Satellite Images ◽

Tsunami Fragility

Abstract. Fragility functions in terms of flow depth, flow velocity and hydrodynamic force are developed to evaluate structural vulnerability in the areas affected by the 2009 Samoa earthquake and tsunami. First, numerical simulations of tsunami propagation and inundation are conducted to reproduce the features of tsunami inundation. To validate the results, flow depths measured in field surveys and waveforms measured by Deep-ocean Assessment and Reporting of Tsunamis (DART) gauges are utilized. Next, building damage is investigated by manually detecting changes between pre- and post-tsunami high-resolution satellite images. Finally, the data related to tsunami features and building damage are integrated using GIS, and tsunami fragility functions are developed based on the statistical analyses.

Download Full-text

Post-Disaster Building Damage Detection from Earth Observation Imagery Using Unsupervised and Transferable Anomaly Detecting Generative Adversarial Networks

Remote Sensing ◽

10.3390/rs12244193 ◽

2020 ◽

Vol 12 (24) ◽

pp. 4193

Author(s):

Sofia Tilon ◽

Francesco Nex ◽

Norman Kerle ◽

George Vosselman

Keyword(s):

Deep Learning ◽

Damage Detection ◽

Satellite Imagery ◽

Building Damage ◽

Training Data ◽

Training Dataset ◽

Generative Adversarial Networks ◽

Generative Adversarial Network ◽

Custom Made ◽

Post Disaster

We present an unsupervised deep learning approach for post-disaster building damage detection that can transfer to different typologies of damage or geographical locations. Previous advances in this direction were limited by insufficient qualitative training data. We propose to use a state-of-the-art Anomaly Detecting Generative Adversarial Network (ADGAN) because it only requires pre-event imagery of buildings in their undamaged state. This approach aids the post-disaster response phase because the model can be developed in the pre-event phase and rapidly deployed in the post-event phase. We used the xBD dataset, containing pre- and post- event satellite imagery of several disaster-types, and a custom made Unmanned Aerial Vehicle (UAV) dataset, containing post-earthquake imagery. Results showed that models trained on UAV-imagery were capable of detecting earthquake-induced damage. The best performing model for European locations obtained a recall, precision and F1-score of 0.59, 0.97 and 0.74, respectively. Models trained on satellite imagery were capable of detecting damage on the condition that the training dataset was void of vegetation and shadows. In this manner, the best performing model for (wild)fire events yielded a recall, precision and F1-score of 0.78, 0.99 and 0.87, respectively. Compared to other supervised and/or multi-epoch approaches, our results are encouraging. Moreover, in addition to image classifications, we show how contextual information can be used to create detailed damage maps without the need of a dedicated multi-task deep learning framework. Finally, we formulate practical guidelines to apply this single-epoch and unsupervised method to real-world applications.

Download Full-text

Interpretability in Convolutional Neural Networks for Building Damage Classification in Satellite Imagery

10.5194/egusphere-egu21-13873 ◽

2021 ◽

Author(s):

Thomas Chen

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Satellite Imagery ◽

Building Damage ◽

Classification Model ◽

Damage Classification ◽

Humanitarian Crisis ◽

Monthly Basis ◽

Loss Of Life ◽

Post Disaster

<p>Natural disasters ravage the world's cities, valleys, and shores on a monthly basis. Having precise and efficient mechanisms for assessing infrastructure damage is essential to channel resources and minimize the loss of life. Using a dataset that includes labeled pre- and post- disaster satellite imagery, the xBD dataset, we train multiple convolutional neural networks to assess building damage on a per-building basis. In order to investigate how to best classify building damage, we present a highly interpretable deep-learning methodology that seeks to explicitly convey the most useful information required to train an accurate classification model. We also delve into which loss functions best optimize these models. Our findings include that ordinal-cross entropy loss is the most optimal loss function to use and that including the type of disaster that caused the damage in combination with a pre- and post-disaster image best predicts the level of damage caused. We also make progress in the realm of qualitative representations of which parts of the images that the model is using to predict damage levels, through gradient class-activation maps. Our research seeks to computationally contribute to aiding in this ongoing and growing humanitarian crisis, heightened by climate change. Specifically, it advances the study of more interpretable machine learning models, which were lacking in previous literature and are important for the understanding of not only research scientists but also operators of such technologies in underserved regions.</p>

Download Full-text

Interpretability in Convolutional Neural Networks for Building Damage Classification in Satellite Imagery

10.20944/preprints202101.0053.v1 ◽

2021 ◽

Author(s):

Thomas Chen

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Satellite Imagery ◽

Building Damage ◽

Classification Model ◽

Damage Classification ◽

Humanitarian Crisis ◽

Monthly Basis ◽

Loss Of Life ◽

Post Disaster

Natural disasters ravage the world's cities, valleys, and shores on a monthly basis. Having precise and efficient mechanisms for assessing infrastructure damage is essential to channel resources and minimize the loss of life. Using a dataset that includes labeled pre- and post- disaster satellite imagery, we train multiple convolutional neural networks to assess building damage on a per-building basis. In order to investigate how to best classify building damage, we present a highly interpretable deep-learning methodology that seeks to explicitly convey the most useful information required to train an accurate classification model. We also delve into which loss functions best optimize these models. Our findings include that ordinal-cross entropy loss is the most optimal loss function to use and that including the type of disaster that caused the damage in combination with a pre- and post-disaster image best predicts the level of damage caused. Our research seeks to computationally contribute to aiding in this ongoing and growing humanitarian crisis, heightened by climate change.

Download Full-text

shorelines of the Holocene Megalake Chad (Africa, Sahara) investigated with very high resolution satellite imagery (Pléiades) : example of the Goz Kerki paleo-spit

Revue Française de Photogrammétrie et de Télédétection ◽

10.52638/rfpt.2014.114 ◽

2014 ◽

pp. 63-68 ◽

Cited By ~ 1

Author(s):

Mathieu Schuster ◽

Claude Roquin ◽

Abderamane Moussa ◽

Jean-François Ghienne ◽

Philippe Duringer ◽

...

Keyword(s):

High Resolution ◽

Satellite Imagery ◽

Satellite Images ◽

Field Surveys ◽

Large Size ◽

High Resolution Satellite Imagery ◽

Sahel Region ◽

Lake Deposits ◽

Humid Period ◽

Very High

Megalake Chad (350,000 km2), the largest paleo-lake of the Sahara-Sahel area, is one of the most emblematic marker of the hydroclimatic changes that occurred during the African Humid Period (AHP; ca. 11,500 — 5,000 years BP) in subtropical Africa. From field surveys, the existence of Megalake Chad is well demonstrated by widespread typical lake deposits. However, considering the very large size of this paleo-lake, it is best evidenced and understood from space. Conspicuous paleo-littoral features distributed along hundreds of kilometers are clearly visible on second generation satellite images. These features represent major archives of the Megalake Chad and of the climate during the AHP. This paper is the first attempt to investigate the paleo-littoral of Megalake Chad with very high resolution satellite imagery. A Pléiades scene (images and DEM) is used to characterize the fossil sand spit of the Goz Kerki, which is one of the most representative and best preserved littoral features of Megalake Chad. Thanks to Pléiades stereoscopic images the geomorphology and the lithology of this paleo-spit can now be detailed and the evolution of the paleo-bathymetry of Megalake Chad can be reconstructed. This brings new insights into the paleo-environments and paleo-climates of the Sahara-Sahel region.

Download Full-text

Beyond building damage: estimating and understanding non-recovery following disasters

10.21203/rs.3.rs-275410/v1 ◽

2021 ◽

Author(s):

Sabine Loos ◽

David Lallemant ◽

Feroz Khan ◽

Jamie McCaughey ◽

Robert Banick ◽

...

Keyword(s):

Population Density ◽

Secondary Data ◽

Building Damage ◽

Nepal Earthquake ◽

Wide Range ◽

Damage Data ◽

Recovery Resources ◽

Post Disaster ◽

2015 Nepal Earthquake

Abstract Following a disaster, crucial decisions about recovery resources often focus on immediate impact, partly due to a lack of detailed information on who will struggle to recover. Here we perform an analysis of surveyed data on reconstruction and secondary data commonly available after a disaster to estimate a metric of non-recovery or the probability that a household could not fully reconstruct within five years after an earthquake. Analyzing data from the 2015 Nepal earthquake, we find that non-recovery is associated with a wide range of factors beyond building damage, such as ongoing risks, population density, and remoteness. If such information were available after the 2015 earthquake, it would have highlighted that many damaged areas have differential abilities to reconstruct due to these factors. More generally, moving beyond damage data to evaluate and quantify non-recovery will support effective post-disaster decisions that consider pre-existing differences in the ability to recover.

Download Full-text

Statistical analysis and modeling to examine the exterior and interior building damage pertaining to the 2016 Kumamoto earthquake

Earthquake Spectra ◽

10.1177/87552930211035408 ◽

2021 ◽

pp. 875529302110354

Author(s):

Haoyi Xiu ◽

Takayuki Shinohara ◽

Masashi Matsuoka ◽

Munenari Inoguchi ◽

Ken Kawabe ◽

...

Keyword(s):

Decision Making ◽

Assessment Procedure ◽

Field Surveys ◽

The 2016 Kumamoto Earthquake ◽

Damage Grade ◽

Damage Grades ◽

Distinct Stage ◽

Survey Results ◽

The Relationship ◽

Post Disaster

After an earthquake occurs, field surveys are conducted by relevant authorities to assess the damage suffered by buildings. The field survey is essential as it ensures the safety of residents and provides the necessary information to local authorities for post-disaster recovery. In Japan, a primary (mandatory) exterior survey is conducted first, and a secondary (voluntary) interior survey is performed subsequently if the residents request a reinvestigation. However, a major challenge associated with field surveys is the substantial time cost of determining the damage grades. Moreover, an interior survey is performed only after receiving the reinvestigation request from occupants, which further delays the decision-making process. In addition, the risk of incorrect damage estimation during the exterior survey must be considered because underestimating the damage can endanger the residents. Therefore, in this study, a three-part analysis (Parts I–III), where each part corresponds to a distinct stage of the standard damage assessment procedure, was performed to characterize the relationship between the building parameters and damage grades at different stages. To further explore the possibility of accelerating decision-making, predictive modeling was performed in each part. The Part I results indicate that estimating the final damage grade for all buildings immediately after the exterior survey is similar to treating the exterior survey results as the final ones. The Part II results show that buildings that potentially require an interior survey can be predicted with reasonable accuracy after the exterior survey. In buildings for which reinvestigations have been requested, Part III demonstrates that the risk of underestimation in the exterior survey can be predicted reliably.

Download Full-text

Imaging ground surface deformations in post-disaster settings via small UAVs

Geoscience Letters ◽

10.1186/s40562-021-00194-8 ◽

2021 ◽

Vol 8 (1) ◽

Author(s):

Richard L. Ybañez ◽

Audrei Anne B. Ybañez ◽

Alfredo Mahar Francisco A. Lagmay ◽

Mario A. Aurelio

Keyword(s):

Surface Deformation ◽

Ground Surface ◽

Point Clouds ◽

Lateral Spreading ◽

Fault Movement ◽

Field Surveys ◽

3D Point Clouds ◽

Vertical Displacements ◽

Resolution Imaging ◽

Post Disaster

AbstractSmall unmanned aerial vehicles have been seeing increased deployment in field surveys in recent years. Their portability, maneuverability, and high-resolution imaging are useful in mapping surface features that satellite- and plane-mounted imaging systems could not access. In this study, we develop and apply a workplan for implementing UAV surveys in post-disaster settings to optimize the flights for the needs of the scientific team and first responders. Three disasters caused by geophysical hazards and their associated surface deformation impacts were studied implementing this workplan and was optimized based on the target features and environmental conditions. An earthquake that caused lateral spreading and damaged houses and roads near riverine areas were observed in drone images to have lengths of up to 40 m and vertical displacements of 60 cm. Drone surveys captured 2D aerial raster images and 3D point clouds leading to the preservation of these features in soft-sedimentary ground which were found to be tilled over after only 3 months. The point cloud provided a stored 3D environment where further analysis of the mechanisms leading to these fissures is possible. In another earthquake-devastated locale, areas hypothesized to contain the suspected source fault zone necessitated low-altitude UAV imaging below the treeline capturing Riedel shears with centimetric accuracy that supported the existence of extensional surface deformation due to fault movement. In the aftermath of a phreatomagmatic eruption and the formation of sub-metric fissures in nearby towns, high-altitude flights allowed for the identification of the location and dominant NE–SW trend of these fissures suggesting horst-and-graben structures. The workplan implemented and refined during these deployments will prove useful in surveying other post-disaster settings around the world, optimizing data collection while minimizing risk to the drone and the drone operators.

Download Full-text