Rapid identification of damaged buildings using incremental learning with transferred data from historical natural disaster cases

Ge, Jiayi; Tang, Hong Kim; Yang, Naisen; Hu, Yijiang

doi:10.1016/j.isprsjprs.2022.11.010

Cited by 27 publications

(9 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the free availability of Landsat and Sentinel satellite data, as well as the powerful geospatial cloud platforms (e.g., Google Earth Engine, GEE), more international studies work towards mapping at a finer spatial resolution (e.g., decametric) over long periods and large geographic, and providing more detailed building-related products such as the Impervious Surface Area (ISA), and the Human Settlement Footprint (HSF). For LUCC, ISA and HSF, it is witnessed a series of global mapping efforts in recent years, such as FROM_GLC (Gong et al, 2013), GAIA (Gong et al, 2020a), GISA-10m (Huang et al, 2022), GHSL (Corbane et al, 2021) and WSF (Marconcini et al, 2020b).…”

Section: Building-related Productsmentioning

confidence: 99%

“…The building rooftop area is an essential indicator of human activity (Huang et al, 2021a), sustainable urbanization (Appolloni et al, 2021;Burke et al, 2021), building energy modeling (Byrne et al, 2015;Chen et al, 2022), urban planning (Nadal et al, 2017) and quick response to natural disasters (Chen et al, 2022;Ge et al, 2023) in the recent years. Such dataset has thus become pivotal in a range of policy decisions by the government, such as arranging the correlation between economic development and demographic growth, and how and where to implement public service.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CBRA: The first multi-annual (2016–2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with Super-resolution Segmentation from Sentinel-2 imagery

Liu

Lyu

2023

Preprint

View full text Add to dashboard Cite

Abstract. Large-scale and multi-annual maps of building rooftop area (BRA) are crucial for addressing policy decisions and sustainable development. In addition, as a fine-grained indicator of human activities, BRA could contribute to urban planning and energy modelling to provide benefits to human well-being. However, it is still challenging to produce large-scale BRA due to the rather tiny size of individual buildings. From the viewpoint of classification methods, conventional approaches utilize high-resolution aerial images (metric or sub-metric resolution) to map BRA; unfortunately, high-resolution imagery is both infrequently captured and expensive to purchase, making the BRA mapping costly and inadequate over a consistent spatio-temporal scale. From the viewpoint of learning strategies, there is a non-trivial gap that persists between the limited training references and the applications over geospatial variations. Despite the difficulties, existing large-scale BRA datasets, such as those from Microsoft or Google, do not include China, hence there are no full-coverage maps of BRA in China yet. In this paper, we first propose a deep-learning method, named Spatio-Temporal aware Super-Resolution Segmentation framework (STSR-Seg) to achieve robust super-resolution BRA extraction from relatively low-resolution imagery over a large geographic space. Then, we produce the multi-annual China building rooftop area dataset (CBRA) with 2.5 m resolution from 2016–2021 Sentinel-2 images. The CBRA is the first full-coverage and multi-annual BRA data in China. With the designed training sample generation algorithms and the spatio-temporal aware learning strategies, the CBRA achieves good performance with the F1 score of 62.55 % (+10.61 % compared with the previous BRA data in China) based on 250,000 testing samples in urban areas, and the recall of 78.94 % based on 30,000 testing samples in rural areas. Temporal analysis shows good performance consistency over years and the well agreement to other multi-annual impervious surface area datasets. The STSR-Seg will enable low-cost, dynamic and large-scale BRA mapping (https://github.com/zpl99/STSR-Seg). The CBRA will foster the development of BRA mapping and therefore provide basic data for sustainable research (Liu et al., 2023; https://doi.org/10.5281/zenodo.7500612).

show abstract

Section: Building-related Productsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

CBRA: The first multi-annual (2016–2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with Super-resolution Segmentation from Sentinel-2 imagery

Liu

Lyu

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…It is worth noting that the used datasets are benchmark datasets and have been employed in lots of research [15,17,23]. The ground truth of the Haiti-Earthquake dataset is available on [18] and the website 1 . In addition, the ground truth of the Bata-Explosion dataset is available on an open public website 2 .…”

Section: Data Inventorymentioning

confidence: 99%

“…For end-to-end building damage assessment, the deep object localization network and deep damage classification network were also merged into one semantic change detection network. Ge, et al [18] presented an incremental learning framework for classifying collapsed buildings. For this purpose, they used end-to-end gradient boosting networks with an assemble-decision strategy as an incremental learning framework.…”

Section: Introductionmentioning

confidence: 99%

BDD-Net+: A Building Damage Detection Framework Based on Modified Coat-Net

Seydi

Hasanlou

Chanussot

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

An accurate and fast assessment of damaged buildings following a disaster is critical for planning rescue and reconstruction efforts. The damage assessment by traditional methods is time-consuming and with limited performance. In this paper, we propose an end-to-end deep-learning network named Building Damage Detection Network-plus (BDD-Net+). The BDD-Net+ is based on a combination of convolution layers and transformer blocks. The proposed framework takes advantage of the multiscale residual convolution blocks and self-attention layers. The proposed framework consists of four main steps: (1) data preparation, (2) model training, (3) damage map generation and evaluation, and (4) the use of an explainable artificial intelligence (XAI) framework for understanding and interpretation of the operation model. Experimental results include two representative realworld benchmark datasets (i.e., the Haiti earthquake and the Bata explosion). The obtained results illustrate that BDD-Net+ achieves excellent efficacy in comparison with other state-ofthe-art methods. Furthermore, the visualization of the results by XAI shows that BDD-Net+ provides more interpretable and explainable results for damage detection than other studied methods.

show abstract

“…Only a few studies have labored on this point and encountered two major issues. First, the test areas were limited to neighborhoods or even smaller areas [27,28,29], and certain parts of those same (or very similar) areas were labeled and used for training [30], so adaptability problems were never realistically examined. Second, when making high-stakes decisions in disaster response management [31,32], it is essential to take into account known uncertainties in the estimation of uncertainty.…”

Section: Introductionmentioning

confidence: 99%

Deep Ensemble Learning for Rapid Large-Scale Post-Earthquake Damage Assessment---Application to 2023 Kahramanmaraş Earthquake Sequence

Soleimani-Babakamali,

Askari,

Heravi

et al. 2023

Preprint

View full text Add to dashboard Cite

Extensive impacted region and government damage surveys following the 2023 Kahramanmaraş earthquake sequence provided a unique opportunity to design and validate a rapid post-event damage assessment framework with satellite images, addressing the typical challenges (e.g., generalization to unseen areas) encountered in rapid regional assessment efforts. By analyzing images from the significantly impacted city of Antakya, we manually identified visual features and compared those with damage states in the survey. Then, an Artificial Intelligence-based framework was developed to automate segmentation and damage assessment processes for future test areas. The resulting tool, dubbed RAPID-A, is an uncertainty-aware ensemble of various deep segmentation models fed with public data released shortly after such incidents. Furthermore, the extent of the damages reported in the survey allowed us to propose and validate a strategy to identify various intensities of damage using a single model, considering the less severe damage as more uncertain. Test case studies in Gaziantep and Kahramanmaraş demonstrate that RAPID-A is generalizable, accurate, and efficient. The significance of this framework lies in its ability to assist in rapid initial damage evaluations before expert dispatch or to replace it in countries that lack the capacity to launch large-scale regional assessment programs to guide foreign support.

show abstract

Rapid identification of damaged buildings using incremental learning with transferred data from historical natural disaster cases

Cited by 27 publications

References 50 publications

CBRA: The first multi-annual (2016–2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with Super-resolution Segmentation from Sentinel-2 imagery

CBRA: The first multi-annual (2016–2021) and high-resolution (2.5 m) building rooftop area dataset in China derived with Super-resolution Segmentation from Sentinel-2 imagery

BDD-Net+: A Building Damage Detection Framework Based on Modified Coat-Net

Deep Ensemble Learning for Rapid Large-Scale Post-Earthquake Damage Assessment---Application to 2023 Kahramanmaraş Earthquake Sequence

Contact Info

Product

Resources

About