Building Damage Level Research Articles

K-Medoids and Support Vector Machine in Predicting the Level og Building Damage in Earthquake Insurance Modeling

Yogyakarta, an Indonesian province prone to earthquakes, frequently suffers extensive damage to buildings, necessitating insurance coverage to mitigate potential losses. This study aims to forecast earthquake insurance premiums by predicting building damage levels resulting from earthquakes. Utilizing data from buildings affected by the June 30, 2023, earthquake in Yogyakarta, we employ K-Medoids Clustering and Support Vector Machine (SVM) to predict two categories of building damage: minor (labelled as 1) and heavy (labelled as 2). The total premiums for minor damage range from approximately USD 86.55 to USD 288.50, while for heavy damage, they range from USD 120.05 to USD 400.18 using the K-Medoids algorithm. Meanwhile, premiums for minor damage range from USD 83.14 to USD 277.13, and for heavy damage, they range from USD 223.67 to USD 745.55 using the SVM algorithm.

Time-series satellite remote sensing reveals gradually increasing war damage in the Gaza Strip.

War-related urban destruction is a significant global concern, impacting national security, social stability, people's survival and economic development. The effects of urban geomorphology and complex geological contexts during conflicts, characterized by different levels of structural damage, are not yet fully understood globally. Here we report how integrating deep learning with data from the independently developed LuoJia3-01 satellite enables near real-time detection of explosions and assessment of different building damage levels in the Israel-Palestine conflict. We found that the damage continually increased from 17 October 2023 to 2 March 2024. We found 3747 missile craters with precision positions and sizes, and timing on vital infrastructure across five governorates in the Gaza Strip on 2 March 2024, providing accurate estimates of potential unexploded ordnance locations and assisting in demining and chemical decontamination. Our findings reveal a significant increase in damage to residential and educational structures, accounting for 58.4% of the total-15.4% destroyed, 18.7% severely damaged, 11.8% moderately damaged and 12.5% slightly damaged-which exacerbates the housing crisis and potential population displacement. Additionally, there is a 34.1% decline in the cultivated area of agricultural land, posing a risk to food security. The LuoJia3-01 satellite data are crucial for impartial conflict monitoring, and our innovative methodology offers a cost-effective, scalable approach to assess future conflicts in various global contexts. These first-time findings highlight the urgent need for an immediate ceasefire to prevent further damage and support the release of hostages and subsequent reconstruction efforts.

An Improved Instance Segmentation Method for Fast Assessment of Damaged Buildings Based on Post-Earthquake UAV Images.

Quickly and accurately assessing the damage level of buildings is a challenging task for post-disaster emergency response. Most of the existing research mainly adopts semantic segmentation and object detection methods, which have yielded good results. However, for high-resolution Unmanned Aerial Vehicle (UAV) imagery, these methods may result in the problem of various damage categories within a building and fail to accurately extract building edges, thus hindering post-disaster rescue and fine-grained assessment. To address this issue, we proposed an improved instance segmentation model that enhances classification accuracy by incorporating a Mixed Local Channel Attention (MLCA) mechanism in the backbone and improving small object segmentation accuracy by refining the Neck part. The method was tested on the Yangbi earthquake UVA images. The experimental results indicated that the modified model outperformed the original model by 1.07% and 1.11% in the two mean Average Precision (mAP) evaluation metrics, mAPbbox50 and mAPseg50, respectively. Importantly, the classification accuracy of the intact category was improved by 2.73% and 2.73%, respectively, while the collapse category saw an improvement of 2.58% and 2.14%. In addition, the proposed method was also compared with state-of-the-art instance segmentation models, e.g., Mask-R-CNN and YOLO V9-Seg. The results demonstrated that the proposed model exhibits advantages in both accuracy and efficiency. Specifically, the efficiency of the proposed model is three times faster than other models with similar accuracy. The proposed method can provide a valuable solution for fine-grained building damage evaluation.

Computer vision-based post-earthquake inspections for building safety assessment

Assessing the safety of earthquake-affected buildings is a critical structural health monitoring task that facilitates the timely response to present dangers and reduces potential threats to life and property. However, post-earthquake time constraints and harsh environmental conditions mean that images and videos taken on-site are frequently affected by poor resolution and blurriness, which may negatively affect the accuracy and usefulness of artificial intelligence image recognition tools. In this study, a HybridGAN model was developed that incorporates ESRGAN for resolution improvement and DeblurGANv2 for blurriness improvement. Additionally, a transfer learning U-Net (TF-Unet) was integrated to detect building components (i.e., columns and structural walls), classify building damage types, and identify building damage levels. Based on recognition results from three case studies and the relevant Taiwan codes, an automated system for building safety evaluation was proposed. The model was trained to directly classify and recognize the level of building component damage. The mean Intersection over Union (mIoU) results for the column and structural wall using the testing dataset were 81.326 % and 57.009 %, respectively. The pre-trained model was used to predict three case studies to test the capability of TF-Unet to handle real-word datasets.

Building Damage Degree Recognition Based on Temporal Attention Features

Natural disasters pose significant harm to society. As an important place for social activities and economic development, the degree of damage to building areas is directly related to disaster loss assessment and emergency rescue. Remote sensing image data, characterized by its wide coverage and multi-temporal features, provides important data support for post-disaster loss assessment. However, imaging differences caused by factors such as shooting time, imaging angle, and different sensors can interfere with the extraction of damage features and loss assessment. This paper proposes a Dual-Exchange-Attention U-Net (DERU-Net) model, which transforms the identification of building damage levels into intra-class semantic change detection. The DFMA feature attention fusion module is introduced to enhance the ability of dual-temporal feature extraction and achieve end-to-end assessment of building damage. The proposed method is comprehensively evaluated and tested on the xBD dataset. Experimental results show that compared with other methods, the DERU-Net proposed in this paper exhibits better stability and evaluation accuracy in assessing the degree of building damage.

The seismic resilience-based methodology of regional building function recovery assessment

Regional assessment of the functional recovery time of urban buildings is critical for decision-making after an earthquake, both as a key indicator of seismic disaster risk and for evaluating urban resilience. However, existing studies are insufficient for assessing the functional recovery time of buildings at the city scale. To address this issue, this study proposes a new methodology for predicting the functional recovery time of an area. This approach comprehensively considers performance-based repair objectives, various damage states, and the affected building area. It offers two improvements over traditional methods based on the assumption of unlimited restoration resources within the disaster area. First, the method establishes a standard for categorizing damaged buildings into repairable and reconstructive buildings using the building damage level parameter. The existing moderate damage states are considered rough and are further divided into three detailed damage states. The distribution functions of the three new states relative to the existing moderate damage are derived from actual earthquake cases. Second, a new parameter, the reconstruction rate, which characterizes the rate of reconstruction time per unit area of a building, is proposed. The repair time per unit area of a house in a repairable state, i.e., the repair rate, is discounted based on this new parameter using the damage state as an indicator. To illustrate the methodology in more detail, we constructed an earthquake scenario for the northeastern margin of the Tibetan Plateau and then provided a detailed assessment of the postearthquake recovery time of Zhangye under the earthquake scenario. This study aims to provide valuable insights for the government to formulate appropriate predisaster prevention policies and rational postdisaster recovery and reconstruction.

Intelligent assessment of building damage of 2023 Turkey-Syria Earthquake by multiple remote sensing approaches

A catastrophic Mw7.8 earthquake hit southeast Turkey and northwest Syria on February 6th, 2023, leading to more than 44 k deaths and 160 k building collapses. The interpretation of earthquake-triggered building damage is usually subjective, labor intensive, and limited by accessibility to the sites and the availability of instant, high-resolution images. Here we propose a multi-class damage detection (MCDD) model enlightened by artificial intelligence to synergize four variables, i.e., amplitude dispersion index (ADI) and damage proxy (DP) map derived from Synthetic Aperture Radar (SAR) images, the change of the normalized difference built-up index (NDBI) derived from optical remote sensing images, as well as peak ground acceleration (PGA). This approach allows us to characterize damage on a large, tectonic scale and a small, individual-building scale. The integration of multiple variables in classifying damage levels into no damage, slight damage, and serious damage (including partial or complete collapses) excels the traditional practice of solely use of DP by 11.25% in performance. Our proposed approach can quantitatively and automatically sort out different building damage levels from publicly available satellite observations, which helps prioritize the rescue mission in response to emergent disasters.

ANALISIS IDENTIFIKASI KONDISI STRUKTUR BANGUNAN

In order to carry out maintenance and upkeep activities, the first thing that must be done is to identify the condition of the building while paying attention to the various causes of damage that occurs and plans for future use of the building. The Master of Management Building, Faculty of Economics and Business, Pancasila University is a building that is more than 20 years old and requires repairs. Structural damage investigations are needed to assess building damage to determine the level of damage as an initial step in the process of identifying the level of building damage. This activity aims to assess the level of damage at the case study location. The initial stage of assessing the level of damage is carried out by carrying out visual observations of all building components. This research uses a descriptive qualitative approach which focuses on identifying the condition of the building structure. The data collection process was carried out by field observation and interviewing respondents (staff, employees, lecturers, work units at the study location, and faculty leaders) to find out the building's previous function and future plans. Analysis of the condition of the building structure and formulation of follow-up recommendations is prepared by involving experts (building structural experts). The research results showed that from visual observations, no damage was found, especially to the structural elements of the building. However, to plan to restore the building's function, a number of follow-up efforts are required, such as carrying out tests using the Non-Destructive Test (NDT) method, strengthening several structural elements including installing foundations on additional structural parts, and replacing the shape and material of the roof covering.

Research on the Evaluation Method for Key Buildings Based on Seismic Damage Index Modification—Taking the Yangzonghai Area Seismic Fortification Planning of Kunming City as an Example

This article focuses on the evaluation of key buildings in seismic mitigation planning, based on the modified seismic damage index. In the evaluation process, a computational model for key building evaluation is proposed, including the coefficients for model modification. The weights of the model modification coefficients are quantitatively calculated, followed by the application of the model for evaluation and conclusion derivation. Taking the "Yangzonghai Regional Seismic Design Planning" as an example, in the key building evaluation section of the specific mitigation planning, the average seismic damage index of group buildings in Yunnan Province's zone of VIII degree is used as the basis for calculation. The average seismic damage index of group buildings is adjusted with weighted factors such as building structure, construction year, site type, compliance with regulations, and adoption of seismic isolation technology. The factors are further adjusted using expert scoring. Finally, the seismic damage index for each important building is calculated. By referring to the corresponding building damage level associated with the seismic damage index, the potential damage levels that each important building may experience during a seismic event of the designated design level are determined. This study provides preliminary guidance for the renovation and improvement of important buildings and serves as an initial reference for decision-making by relevant government departments.

Classification of Earthquake Damage of Buildings in Songyuan Area Based on Image Recognition

Since the emergence of human beings on the earth, various disasters have been accompanied. Among many natural disasters, the earthquake is undoubtedly one of the most threatening disasters. This project uses Res Net-50 model for deep learning and image recognition of building structural damage. Through the program to assess the local earthquake damage, given the feasible standards to facilitate a unified understanding of the earthquake situation, thereby improving the efficiency of disaster relief. Through experiments, the accuracy of the training set of the two classifications finally reached about 89.3 %, and the prediction accuracy of the test set finally reached about 71.4 %, Through the identification of post-earthquake building damage in Songyuan area, it can be learned that the accuracy of the software identification binary classification task is 73.21 %. Experiments show that taking photos can be used to predict the damage level of buildings in a certain area, and seismic damage identification can provide basis and support for post-disaster rescue and reconstruction and economic loss assessment.

MODEL DECISION ANALYSIS IN DETERMINING BUILDING QUALITY BASED ON THE ANALYTICAL HIERARCHY PROCESS

Building infrastructure consisting of several types of alternatives that provide the need for selection of various kinds of assessments for ranking in carrying out repairs. The purpose of this study is to determine the priority order and condition of building quality, which is useful as a reference in improvement activities related to seeing the satisfaction of policy makers. Problems with building quality can result from various factors, one of which is related to facilities, building repairs and their handling which are feared to affect productivity in assessments and influence policy making. Primary data can be used as research data to be included in the decision model and direct observations can be made at the research location for each building in Dilhokseumawe. This study uses the Analytical Hierarchy Process (AHP) model, to determine the weight of the components obtained from the results of the interest assessment of building components which include the level of building damage, building age, ease of access, building structure, roof structure, building safety level. The results of the analysis show that factors the main consideration is improving the level of building security with a value of 0.297858847, the second ranking is the Building Structure factor with a priority value of 0.257685234%, the third ranking is the Ease of Access factor with a priority value of 0.128154963 and the Fourth is Building Age with a value of 0.076683671%. Finally the Roof Structure with a priority value of 0.072065172%. The results of this study can be used as recommendations for decision making in view of the quality of buildings and buildings.
 

THE ACTUAL CONDITIONS AND ISSUES CONCERNING THE RECONSTRUCTION COST AND PERIOD OF THE HOUSE RECONSTRUCTION PROCESS AFTER THE KUMAMOTO EARTHQUAKE

In this study, in order to contribute to the work of supporting the reconstruction of the lives of future disaster victims, we ascertained the actual level of building damage and public support for victims in Mashiki Town, which was damaged by the 2016 Kumamoto Earthquake. We then investigated and figured out the actual situation of the reconstruction process of households that have repaired or rebuilt their house through a questionnaire survey to focus on the time taken, cost of repair and rebuilding. We also clarified the level of understanding of the support system and ease of creating application forms.

Short-wall paste continuous mining and continuous backfilling for controlling industrial square surface subsidence

To facilitate coal mining under the industrial square in Western China, short-wall paste continuous mining and continuous backfilling (PCMCB) is proposed to release the overlaid resources and handle coal gangue. First, the concept of short-wall PCMCB is presented, and the principle of surface subsidence is analyzed. Second, the mechanical model of the branch roadway and strength of the backfill body are established via mechanical derivations. Third, the proportion of backfill materials is calculated. Finally, field applications are carried out in the Dingjiaqu coal mine. The results show the following: 1) Based on the simply supported beam model of the roof and strength design model of the backfill body, the reasonable width of the branch roadway was determined as 5 m and late strength of the backfill body should exceed 2.62 MPa in the Dingjiaqu coal mine. 2) According to the design strength of the backfill body and backfill material proportion, the reasonable ratio of cement: fly ash: gangue: water was determined as 1:1.82:4.17:1.78. 3) In accordance with the equivalent mining height theory, the critical filling rate of the goaf should be more than 90% to ensure that the deformation is within the first level of building damage. 4) The field application shows that subsidence could be effectively controlled by backfilling, the overall surface subsidence was very small, and the maximum subsidence value (77 mm) was far lower than the deformation extremes of the building protection requirements. This study provides practical guidance for the exploitation of overlaid resources under similar shallow burial conditions, which can help in the disposal of coal gangue with considerable social and economic benefits with popularization and application.

Rapid assessment of the public buildings after the February 25th, 2022 West Pasaman Earthquake

The earthquake that occurred on February 25th, 2022 (6.2 M) and the aftershock in the West Pasaman Regency, West Sumatra Province has caused damages to infrastructure and community settlements. One of those affected by the Pasaman Earthquake 2022 are public buildings. To determine the level of public building damage, an assessment should be carried out before reoccupying the buildings. In this study, a rapid assessment of damaged public buildings was conducted after the Pasaman Earthquake by the Disaster Study Center of Andalas University team. The assessed buildings include the buildings of Balerong Hall, Ibnu Sina Hospital, Grand Mosque, Population and Civil Registration Agency (Disdukcapil) office, Regional Public Hospital, and Elementary School (SDN) 22 of Talamau. The results of the assessment show that in general, these public buildings have non-structural damages, such as cracks in the walls and ceiling crashes from their positions. Meanwhile, in the buildings of Elementary School (SDN) 22 Talamau, Grand Mosque, and Disdukcapil, it is found that there are some structural damages observed, such as cracks on the columns. The main factor causing the damage is the low quality of the concrete in the columns. Recommendations for repairing the damaged building are also discussed in this study.

Machine learning remote sensing using the random forest classifier to detect the building damage caused by the Anak Krakatau Volcano tsunami

In Indonesia, tsunamis are frequent events. In 2000–2016, there were 44 tsunami events in Indonesia, with financial losses reaching 43.38 trillion. In 2018, a tsunami occurred in the Sunda Strait due to the eruption of the Anak Krakatau Volcano, which caused many fatalities and much building damage. This study aimed to detect the building damage in the Labuan District, Banten Province. Machine learning methods were used to detect building damage using random forest with object-based techniques. No previous research has combined selected predictors into scenarios; hence, the novelty of this study is combining various random forest predictors to identify the extent of building damage using 14 predictor scenarios. In addition, field surveys were conducted two years and nine months after the tsunami to observe the changes and efforts made. The results of the random forest classification were validated and compared with three datasets, namely xBD, Copernicus, and field survey data. The results of this study can help classify the level of building damage using satellite imagery to improve mitigation in tsunami-prone areas.

The potential for deflagration to detonation transition (DDT) - Lessons from LPG tanker transportation accident

On June 13, 2020, an LPG tanker leaked, followed by a vapor cloud explosion on the highway section near Liangshan Village, Wenling City, China. The accident killed 20 people and injured more than 170, causing severe damage to the surrounding area and significant impact to the society. This work studied the large vapor cloud explosion of the LPG transportation accident in terms of the dynamic phenomena of explosion, building damage level, and mechanism of DDT. The accident analysis indicates that blast shock wave front and reaction front were moving together at a speed higher than the speed of sound; the explosion overpressure of the accident is consistent with the calculation using the detonation approach of TNO multi-energy method; the blockage level and flame acceleration distance caused by the shrubs, and the quantity of release LPG provide enough conditions for DDT occurrence. Therefore, it is believed that DDT has occurred in this accident. The findings of this work also contribute to providing scientific recommendations on preventing DDT accidents, including redesign the vegetation at incident-prone section of the highway to eliminate the condition of DDT occurrence, and optimize the transportation route to avoid highly populated areas.

SimEx: A Tool for the Rapid Evaluation of the Effects of Explosions

The dynamic response of structural elements subjected to blast loading is a problem of growing interest in the field of defense and security. In this work, a novel computational tool for the rapid evaluation of the effects of explosions, hereafter referred to as SimEx, is presented and discussed. The classical correlations for the reference chemical (1 kg of TNT) and nuclear (106 kg of TNT) explosions, both spherical and hemispherical, are used together with the blast wave scaling laws and the International Standard Atmosphere (ISA) to compute the dynamic response of Single-Degree-of-Freedom (SDOF) systems subject to blast loading. The underlying simplifications in the analysis of the structural response follow the directives established by UFC 3-340-02 and the Protective Design Center Technical Reports of the US Army Corps of Engineers. This offers useful estimates with a low computational cost that enable in particular the computation of damage diagrams in the Charge Weight–Standoff distance (CW–S) space for the rapid screening of component (or building) damage levels. SimEx is a computer application based on Matlab and developed by the Fluid Mechanics Research Group at University Carlos III of Madrid (UC3M). It has been successfully used for both teaching and research purposes in the Degree in Security Engineering, taught to the future Guardia Civil officers at the Spanish University Center of the Civil Guard (CUGC). This dual use has allowed the development of the application well beyond its initial objective, testing on one hand the implemented capacities by undergraduate cadets with the end-user profile, and implementing new functionalities and utilities by Masters and PhD students. With this experience, the application has been continuously growing since its initial inception in 2014 both at a visual and a functional level, including new effects in the propagation of the blast waves, such as clearing and confinement, and incorporating new calculation assistants, such as those for the thermochemical analysis of explosive mixtures; crater formation; fragment mass distributions, ejection speeds and ballistic trajectories; and the statistical evaluation of damage to people due to overpressure, body projection, and fragment injuries.

Building Damage Assessment Based on Siamese Hierarchical Transformer Framework

The rapid and accurate damage assessment of buildings plays a critical role in disaster response. Based on pairs of pre- and post-disaster remote sensing images, effective building damage level assessment can be conducted. However, most existing methods are based on Convolutional Neural Network, which has limited ability to learn the global context. An attention mechanism helps ameliorate this problem. Hierarchical Transformer has powerful potential in the remote sensing field with strong global modeling capability. In this paper, we propose a novel two-stage damage assessment framework called SDAFormer, which embeds a symmetric hierarchical Transformer into a siamese U-Net-like network. In the first stage, the pre-disaster image is fed into a segmentation network for building localization. In the second stage, a two-branch damage classification network is established based on weights shared from the first stage. Then, pre- and post-disaster images are delivered to the network separately for damage assessment. Moreover, a spatial fusion module is designed to improve feature representation capability by building pixel-level correlation, which establishes spatial information in Swin Transformer blocks. The proposed framework achieves significant improvement on the large-scale building damage assessment dataset—xBD.

A novel attention-based deep learning method for post-disaster building damage classification

Although several past studies proposed deep learning methods to extract pre-disaster building footprints for post-disaster management using remote sensing techniques, building damage classification after a disaster is rarely discussed. Building damage classification using deep learning is a challenging task because imagery data of damaged buildings are limited publicly and damage levels are hard to judge from images. Most building damage classification papers with deep learning methods only judge buildings as collapse or not. Moreover, most deep learning models are designed for non-damaged building segmentation, which might not be suitable for damaged building segmentation. In order to solve the problems, including lack of imagery data, detailed damage levels, and suitable deep learning model for damaged building segmentation, this research aims to propose and evaluate a novel deep learning model to classify building damage into four levels quickly with in-house labelled damaged building information. This research proposes the Squeeze-and-Excitation dual High-Resolution Network (HRNet) model to be trained with open access xBD dataset and the 2010 Haiti Earthquake images labelled at the University of New South Wales. HRNet is adopted twice in the model because two steps need to be completed, building localization and then damage classification. Four comparative experiments are implemented with seven metrics, including combination F1, localization F1, localization precision, localization recall, damage F1, damage precision, and damage recall. The first experiment is to choose where the best place is to add SE channel attention block in the model. The Squeeze-and-Excitation (SE) block is added at four different places in the dual HRNet backbone to have a comparison with the model without SE. According to analyses of the results, the SE-PRE model has the best performance. The second experiment shows that a larger input size of images increases the processing time but performs better. The third and fourth experiments reveal that the Sigmoid function is better than Hard-Sigmoid in SE, and the transfer learning with pre-trained ImageNet weights may not be suitable for the proposed model. The theoretical contribution is proposing a novel deep learning model for building damage classification. Another contribution of this research is creating a Haiti Earthquake image dataset with four building damage levels. The practical contribution is providing a safe and speedy post-disaster building damage classification method with minimal fieldwork to support rescue teams.

Sentinel-1 and Sentinel-2 data fusion to distinguish building damage level of the 2018 Lombok Earthquake

Following an earthquake, Building Damage Assessment (BDA) is crucial in detecting areas that need immediate rescue actions and planning for the best evacuation strategy. Remote sensing has been widely used for BDA. The availability of Sentinel-1 and Sentinel-2 images that are freely accessible could enhance remote sensing applications for building damage classification caused by natural disasters. The accuracy of using the two satellite images for BDA mapping in tropical regions is still uncertain. This research aims to assess Sentinel-1 and Sentinel-2 data fusion performance for BDA mapping following the catastrophic 2018 Lombok Earthquake. Pre- and post-earthquake images of Sentinel-1 and Sentinel-2 were selected, preprocessed, and integrated using the random forest classifier. Three data scenarios were used in this study, i.e., Sentinel-1, Sentinel-2, and the fusion of both datasets. The results demonstrated that all models showed excellent performance in classifying destroyed buildings compared to severely and moderately damaged buildings. This is due to the medium spatial resolution of the images could not identify the damage level in detail. Sentinel-1 and Sentinel-2 data fusion provided the highest overall accuracy value (62.4%) compared to other single dataset models. According to the feature importance analysis result, GLCM Mean, GLCM Variance, and NIR Band played crucial roles in the damage classification. In conclusion, the fusion of Sentinel-1 and Sentinel-2 can provide the best model for destroyed building mapping. However, severely and moderately building damaged mapping can only be accurately mapped using higher spatial resolution images.