scholarly journals AUTOMATED BUILDING SEGMENTATION AND DAMAGE ASSESSMENT FROM SATELLITE IMAGES FOR DISASTER RELIEF

2021 ◽  
Vol XLIII-B3-2021 ◽  
pp. 741-748
Author(s):  
X. Yuan ◽  
S. M. Azimi ◽  
C. Henry ◽  
V. Gstaiger ◽  
M. Codastefano ◽  
...  
Keyword(s):  
Damage Assessment ◽  
Building Damage ◽  
Accurate Information ◽  
Generalization Capability ◽  
Damage Classification ◽  
Performance Impact ◽  
Humanitarian Crisis ◽  
Disaster Data ◽  
The Impact

Abstract. After a natural disaster or humanitarian crisis, rescue forces and relief organisations are dependent on fast, area-wide and accurate information on the damage caused to infrastructure and the situation on the ground. This study focuses on the assessment of building damage levels on optical satellite imagery with a two-step ensemble model performing building segmentation and damage classification trained on a public dataset. We provide an extensive generalization study on pre- and post-disaster data from the passage of the cyclone Idai over Beira, Mozambique, in 2019 and the explosion in Beirut, Lebanon, in 2020. Critical challenges are addressed, including the detection of clustered buildings with uncommon visual appearances, the classification of damage levels by both humans and deep learning models, and the impact of varying imagery acquisition conditions. We show promising building damage assessment results and highlight the strong performance impact of data pre-processing on the generalization capability of deep convolutional models.

scholarly journals Building Damage Detection Using U-Net with Attention Mechanism from Pre- and Post-Disaster Remote Sensing Datasets

2021 ◽  
Vol 13 (5) ◽  
pp. 905
Author(s):  
Chuyi Wu ◽  
Feng Zhang ◽  
Junshi Xia ◽  
Yichen Xu ◽  
Guoqing Li ◽  
...  
Keyword(s):  
Damage Assessment ◽  
Large Scale ◽  
Binary Classification ◽  
Open Data ◽  
Building Damage ◽  
Attention Mechanism ◽  
Large Scale Dataset ◽  
Data Program ◽  
The Impact ◽  
Post Disaster

The building damage status is vital to plan rescue and reconstruction after a disaster and is also hard to detect and judge its level. Most existing studies focus on binary classification, and the attention of the model is distracted. In this study, we proposed a Siamese neural network that can localize and classify damaged buildings at one time. The main parts of this network are a variety of attention U-Nets using different backbones. The attention mechanism enables the network to pay more attention to the effective features and channels, so as to reduce the impact of useless features. We train them using the xBD dataset, which is a large-scale dataset for the advancement of building damage assessment, and compare their result balanced F (F1) scores. The score demonstrates that the performance of SEresNeXt with an attention mechanism gives the best performance, with the F1 score reaching 0.787. To improve the accuracy, we fused the results and got the best overall F1 score of 0.792. To verify the transferability and robustness of the model, we selected the dataset on the Maxar Open Data Program of two recent disasters to investigate the performance. By visual comparison, the results show that our model is robust and transferable.

scholarly journals Comparative analysis of damage to sewer rehabilitation coatings

2018 ◽  
Vol 45 ◽  
pp. 00041
Author(s):  
Andrzej Kuliczkowski ◽  
Stanisław Nogaj
Keyword(s):  
Damage Assessment ◽  
Technical Condition ◽  
Classification Methods ◽  
Damage Classification ◽  
Reliable Assessment ◽  
Sewer Rehabilitation ◽  
Underground Infrastructure ◽  
Original Part ◽  
Near Future

Technologies for the trenchless rehabilitation of pipelines using various types of coatings have been used for almost half a century. Considering that the assumed life expectancy of such renewed pipelines is 50 years, it will be necessary to assess their technical condition in the near future. The aim of this article is to attempt to answer the question "Do existing damage classification methods allow for the full and reliable assessment of the sewers already renewed with rehabilitation coatings?". The scope of the article, and its original part, is to describe how the problem of damage assessment of rehabilitation coatings has been included in various methods of classification of underground infrastructure pipelines, and conducting a comparison that compares these methods in terms of the damages described. An interpretation of the results of the research on rehabilitation coatings operated in various time periods, starting from those recently applied to those operating for over 30 years, was also made. The result of the analysis is to present the differences and deficiencies in the damage classification methods discussed.

scholarly journals DEEP LEARNING FOR AUTOMATIC BUILDING DAMAGE ASSESSMENT: APPLICATION IN POST-DISASTER SCENARIOS USING UAV DATA

2021 ◽  
Vol V-1-2021 ◽  
pp. 113-120
Author(s):  
A. Calantropio ◽  
F. Chiabrando ◽  
M. Codastefano ◽  
E. Bourke
Keyword(s):  
Deep Learning ◽  
Damage Assessment ◽  
Central Italy ◽  
Remote Sensing Data ◽  
Building Damage ◽  
Research Field ◽  
Future Research ◽  
Damage Classification ◽  
Research Perspectives ◽  
Post Disaster

Abstract. During the last few years, the technical and scientific advances in the Geomatics research field have led to the validation of new mapping and surveying strategies, without neglecting already consolidated practices. The use of remote sensing data for damage assessment in post-disaster scenarios underlined, in several contexts and situations, the importance of the Geomatics applied techniques for disaster management operations, and nowadays their reliability and suitability in environmental emergencies is globally recognized. In this paper, the authors present their experiences in the framework of the 2016 earthquake in Central Italy and the 2019 Cyclone Idai in Mozambique. Thanks to the use of image-based survey techniques as the main acquisition methods (UAV photogrammetry), damage assessment analysis has been carried out to assess and map the damages that occurred in Pescara del Tronto village, using DEEP (Digital Engine for Emergency Photo-analysis) a deep learning tool for automatic building footprint segmentation and building damage classification, functional to the rapid production of cartography to be used in emergency response operations. The performed analyses have been presented, and the strengths and weaknesses of the employed methods and techniques have been outlined. In conclusion and based on the authors' experience, some operational suggestions and best practices are provided and future research perspectives within the same research topic are introduced.

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

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>

scholarly journals TIPOLOGI KERUSAKAN BANGUNAN AKIBAT CUACA EKSTREM STUDI KASUS CUACA EKSTREM KABUPATEN BANTUL 2019

2020 ◽  
Vol 2 (2) ◽  
pp. 31-36
Author(s):  
Rachmat Wahyu Prabowo
Keyword(s):  
The Elderly ◽  
Extreme Weather ◽  
Building Damage ◽  
Vulnerable Groups ◽  
Extreme Weather Events ◽  
Graphical Form ◽  
Disaster Data ◽  
City District ◽  
The Impact ◽  
Building Elements

The threat of disaster is unavoidable in Indonesia. In 2018, for example, 2,564 disasters occurred in which no city / district in Indonesia was free from threats. Statistical data shows that more than 90% of disasters in Indonesia are included in the hydrometereological disaster category, which has continued to increase over the past 15 years. Even in the early half of 2020, there were 256 disasters with 99% of them being hydrometereological disasters such as floods, landslides and tornadoes. The existence of extreme weather events in Bantul Regency in 2019 shows that the dangers of hydrometereology can cause considerable harm to the community. There needs to be a study to find out how to mitigate hydrometereological disasters which are the most common hazards in Indonesia. Analysis of building damage and disaster data needs to be carried out to determine the characteristics of the impact of extreme weather on buildings, to identify building elements that are vulnerable to damage, to provide anticipatory measures and alertness to extreme weather in Indonesia. Building damage data is grouped and processed in graphical form to read information patterns as material for analysis. The results of the study show that there are things that must be done in the face of extreme weather including: choosing the quality of materials and the strength of the construction of the building, attention to building elements, especially the back of the building (kitchen & bathroom), walls, and foundations, high vigilance in the area steep contours and highlands, need to be vigilant with all elements of society in areas of high potential for extreme weather, and the need to pay attention to vulnerable groups of the elderly and children.

scholarly journals Transfer Learning for Improving Seismic Building Damage Assessment

2022 ◽  
Vol 14 (1) ◽  
pp. 201
Author(s):  
Qigen Lin ◽  
Tianyu Ci ◽  
Leibin Wang ◽  
Sanjit Kumar Mondal ◽  
Huaxiang Yin ◽  
...  
Keyword(s):  
Damage Assessment ◽  
Historical Data ◽  
Data Transfer ◽  
Rapid Assessment ◽  
Model Performance ◽  
Building Damage ◽  
Assessment Model ◽  
Training Data ◽  
Superior Performance ◽  
The Impact

The rapid assessment of building damage in earthquake-stricken areas is of paramount importance for emergency response. The development of remote sensing technology has aided in deriving reliable and precise building damage assessments of extensive areas following disasters. It is well documented that convolutional neural network methods have superior performance in earthquake building damage assessment compared with traditional machine learning methods. However, deep learning models require a large number of samples, and sufficient numbers of samples are usually not available in the newly earthquake-stricken areas rapidly enough. At the same time, the historical samples inevitably differ from the new earthquake-affected areas due to the discrepancy of regional building characteristics. For this purpose, this study proposes a data transfer algorithm for evaluating the impact of a single historical training sample on the model performance. Then, beneficial samples are selected to transfer knowledge from the historical data for facilitating the calibration of the new model. Four models are designed with two earthquake damage building datasets and the performance of the models is compared and evaluated. The results show that the data transfer algorithm proposed in this work improves the reliability of the building damage assessment model significantly by filtering samples from the historical data that are suitable for the new task. The performance of the model built based on the data transfer method on the test set of new earthquakes task is approximately 8% higher in overall accuracy compared with the model trained directly with the new earthquake samples when the training data for the new task is only 10% of the historical data and is operating under the objective of four classes of building damage. The proposed data transfer algorithm has effectively enhanced the precision of the seismic building damage assessment in a data-limited context. Thus, it could be applicable to the building damage assessment of new disasters.

Object-Oriented Image Understanding and Post-Earthquake Damage Assessment for the 2003 Bam, Iran, Earthquake

2005 ◽  
Vol 21 (1_suppl) ◽  
pp. 225-238 ◽  
Author(s):  
Luca Gusella ◽  
Beverley J. Adams ◽  
Gabriele Bitelli ◽  
Charles K. Huyck ◽  
Alessandro Mognol
Keyword(s):  
High Resolution ◽  
Satellite Imagery ◽  
Damage Assessment ◽  
Object Oriented ◽  
Image Understanding ◽  
Statistical Characteristics ◽  
Damage Classification ◽  
Bam Earthquake ◽  
Earthquake Damage Assessment

This paper presents a methodology for quantifying the number of buildings that collapsed following the Bam earthquake. The approach is object rather than pixel-oriented, commencing with the inventory of buildings as objects in high-resolution QuickBird satellite imagery captured before the event. The number of collapsed structures is computed based on the unique statistical characteristics of these objects/buildings within the “after” scene. A total of 18,872 structures were identified within Bam, of which the results suggest that 34% collapsed—a total of 6,473. Preliminary assessments indicate an overall accuracy for the damage classification of 70.5%.

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

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.

Forced displacement in Colombia: Magnitude and causes

2009 ◽  
Vol 4 (1) ◽  
Author(s):  
Ana Maria Ibanez
Keyword(s):  
State Policy ◽  
Forced Migration ◽  
Refugee Camps ◽  
Illegal Drugs ◽  
Armed Groups ◽  
Forced Displacement ◽  
Humanitarian Crisis ◽  
Individual Basis ◽  
The Impact

The article describes the magnitude, geographical extent,  and causes of forced population displacements in Colombia. Forced migration in Colombia is a war strategy adopted by armed groups to strengthen territorial strongholds, weaken civilian support to the enemy, seize valuable lands, and produce and transport illegal drugs with ease. Forced displacement in Colombia today affects 3.5 million people. Equivalent to 7.8 percent of Colombia's population, and second worldwide only to Sudan, this shows the magnitude of the humanitarian crisis the country is facing. The phenomenon involves all of Colombia's territory and nearly 90 percent of the country's municipalities expel or receive population. In contrast to other countries, forced migration in Colombia is largely internal. Illegal armed groups are the main responsible parties, migration does not result in massive refugee streams but occurs on an individual basis, and the displaced population is dispersed throughout the territory and not focused in refugee camps. These characteristics pose unique challenges for crafting state policy that can effectively mitigate the impact of displacement.

Sign in / Sign up

Export Citation Format

Share Document