45,774 Hits in 5.8 sec

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

Jinyuan Shao, Lina Tang, Ming Liu, Guofan Shao, Lang Sun, Quanyi Qiu
2020 Remote Sensing  
The use of satellite imagery for disaster assessment can overcome this problem.  ...  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.  ...  Acknowledgments: The authors would like to thank editors and the anonymous reviewers for their insightful suggestions and comments.  ... 
doi:10.3390/rs12101670 fatcat:ltszpxxs2zbtzbc4xwwynpvrba

Nazr-CNN: Fine-Grained Classification of UAV Imagery for Damage Assessment [article]

N. Attari and F. Ofli and M. Awad and J. Lucas and S. Chawla
2017 arXiv   pre-print
We propose Nazr-CNN1, a deep learning pipeline for object detection and fine-grained classification in images acquired from Unmanned Aerial Vehicles (UAVs) for damage assessment and monitoring.  ...  We show one such case of transfer learning to assess the level of damage in aerial images collected after a typhoon in Philippines.  ...  DEEP LEARNING FRAMEWORK Since our labeled data set is relatively small (1,085 images), we have built our deep learning pipeline (Nazr-CNN) using existing pre-trained networks as building blocks.  ... 
arXiv:1611.06474v2 fatcat:jh7lgi62jfgdbkga5mqtvmqrwe

Interpretability in Convolutional Neural Networks for Building Damage Classification in Satellite Imagery [article]

Thomas Y. Chen
2022 arXiv   pre-print
training data most accurately predicts the level of damage caused.  ...  We present a novel methodology of interpretable deep learning that seeks to explicitly investigate the most useful modalities of information in the training data to create an accurate classification model  ...  The blurred building of size 64 pixels would not be a useful data point for any deep learning model to learn from and yield accurate results.  ... 
arXiv:2201.10523v1 fatcat:jxkezulgdrdo3hpdhnf2gwwgee


A. Calantropio, F. Chiabrando, M. Codastefano, E. Bourke
2021 ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
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  ...  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  ...  Thanks to INGC for allowing the use of the Mozambique image dataset.  ... 
doi:10.5194/isprs-annals-v-1-2021-113-2021 fatcat:ijxtcyy3evguhchy5i226pbo6a

Deep Learning-Enabled Semantic Inference of Individual Building Damage Magnitude from Satellite Images

Bradley J. Wheeler, Hassan A. Karimi
2020 Algorithms  
We present a deep learning-based computer vision model to semantically infer the magnitude of damage to individual buildings after natural disasters using pre- and post-disaster satellite images.  ...  studies to show our methods and results for the importance and impact of different training parameters in deep learning for satellite imagery.  ...  Each building has a ground truth polygon that outlines its location in the image, and a ground truth label that assigns the building a semantic label from the joint damage scale.  ... 
doi:10.3390/a13080195 fatcat:o3a33wg4jnfadbxap5rbkbabam

RescueNet: A High Resolution UAV Semantic Segmentation Benchmark Dataset for Natural Disaster Damage Assessment [article]

Tashnim Chowdhury, Robin Murphy, Maryam Rahnemoonfar
2022 arXiv   pre-print
Deep learning based computer vision techniques can help in scene understanding, and help rescue teams with precise damage assessment.  ...  Unfortunately, available datasets for natural disaster damage assessment lack detailed annotation of the affected areas, and therefore do not support the deep learning models in total damage assessment  ...  His research interest includes Deep Learning, Machine Learning, Semantic Segmentation, Few Shot Learning, Meta Learning, and Bayesian Learning.  ... 
arXiv:2202.12361v1 fatcat:7hbi7h5xlveqbcos7hoiwnuxki

Earthquake-induced Building Damage Mapping Based on Multi-task Deep Learning Framework

Fang Chen, Bo Yu
2019 IEEE Access  
In this paper, a multi-task deep learning framework is proposed to map damaged and intact buildings from large-scale very high spatial resolution images.  ...  The development in earth observation using high spatial resolution images makes it possible to recognize damaged and intact buildings.  ...  The main task is to detect buildings in good shape and damaged and the extra task is to semantically segment images into multiple ground objects. The extra task is used to supplement the main task.  ... 
doi:10.1109/access.2019.2958983 fatcat:7fihfiatobhrnllese62h2by2m

The 2011 Tohoku Tsunami from the Sky: A Review on the Evolution of Artificial Intelligence Methods for Damage Assessment

Jérémie Sublime
2021 Geosciences  
The amount of devastation was so great that it took years to achieve a proper assessment of the economical and structural damage, with the consequences still being felt today.  ...  We focus on more than 15 studies that are compared and evaluated in terms of the data they require, the methods used, their degree of automation, their metric performances, and their strengths and weaknesses  ...  The authors use both a map of the building footprints to detect lateral shift, and a ground-truth for the damage assessment training.  ... 
doi:10.3390/geosciences11030133 fatcat:q4e4d73xxff3bcxeiik5byqr7e

BDD-Net: An End-to-End Multiscale Residual CNN for Earthquake-Induced Building Damage Detection

Seyd Teymoor Seydi, Heidar Rastiveis, Bahareh Kalantar, Alfian Abdul Halin, Naonori Ueda
2022 Remote Sensing  
In this study, we explore this fusion potential, which incorporates deep learning.  ...  The three main phases are: (1) data preprocessing and building footprint extraction based on building vector maps, (2) sample data preparation and data augmentation, and (3) model optimization and building  ...  Ground Truth Data In this research, the sample data was manually collected and divided into two classes: (1) Damage and (2) Non-damage. This sample was used to train BDD-Net.  ... 
doi:10.3390/rs14092214 fatcat:gmhwax4ohrcqhdj664tfuuxwem

BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images [article]

Yu Shen, Sijie Zhu, Taojiannan Yang, Chen Chen, Delu Pan, Jianyu Chen, Liang Xiao, Qian Du
2021 arXiv   pre-print
With a pair of pre- and post-disaster satellite images, building damage assessment aims at predicting the extent of damage to buildings.  ...  With the powerful ability of feature representation, deep neural networks have been successfully applied to building damage assessment.  ...  With a set of labeled data, deep learning-based methods can automatically learn image feature representations from low level to high level, without selecting hand-crafted image features.  ... 
arXiv:2105.07364v1 fatcat:v52hpkmi35dlzbsrrvwe7vpj6m

Towards Operational Satellite-Based Damage-Mapping Using U-Net Convolutional Network: A Case Study of 2011 Tohoku Earthquake-Tsunami

Yanbing Bai, Erick Mas, Shunichi Koshimura
2018 Remote Sensing  
In this article, a deep learning algorithm for the semantic segmentation of high-resolution remote-sensing images using the U-net convolutional network was proposed to map the damage rapidly.  ...  damage assessment accuracy and lag in timeliness, which dramatically reduces the significance and feasibility of extending the present method to practical operational applications.  ...  The HR images and ground-truth data of building damage were projected into the UTM/WGS84 geo-referenced coordinate system.  ... 
doi:10.3390/rs10101626 fatcat:t5h73sysfvctfmdxmtxt62xcva

Pyramid Pooling Module-Based Semi-Siamese Network: A Benchmark Model for Assessing Building Damage from xBD Satellite Imagery Datasets

Yanbing Bai, Junjie Hu, Jinhua Su, Xing Liu, Haoyu Liu, Xianwen He, Shengwang Meng, Erick Mas, Shunichi Koshimura
2020 Remote Sensing  
Most mainstream research on assessing building damage using satellite imagery is based on scattered datasets and lacks unified standards and methods to quantify and compare the performance of different  ...  Our proposed method achieves F1 scores of 0.90, 0.41, 0.65, and 0.70 for the undamaged, minor-damaged, major-damaged, and destroyed building classes, respectively.  ...  We thank the two reviewers for their helpful and constructive comments on our work. The author gratefully acknowledges the support of K.C. Wong Education Foundation, Hong Kong.  ... 
doi:10.3390/rs12244055 fatcat:undwujdacvd7zb7m4rdj6d6anu

Tsunami Damage Detection with Remote Sensing: A Review

Shunichi Koshimura, Luis Moya, Erick Mas, Yanbing Bai
2020 Geosciences  
Recent advances of remote sensing and technologies of image analysis meet the above needs and lead to more rapid and efficient understanding of tsunami affected areas.  ...  E.M. is for tsunami damage interpretation, L.M. is for machine learning method, and Y.B. is for deep learning method. All authors have read and agreed to the published version of the manuscript.  ...  Funding: This study was partly funded by the JSPS Kakenhi Program (17H06108 and 17H02050) and the Core Research Cluster of Disaster Science at Tohoku University.  ... 
doi:10.3390/geosciences10050177 fatcat:o6trnuw7efgvjp7zbnh3ev2nqm


X. Yuan, S. M. Azimi, C. Henry, V. Gstaiger, M. Codastefano, M. Manalili, S. Cairo, S. Modugno, M. Wieland, A. Schneibel, N. Merkle
2021 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
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  ...  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.  ...  These data form the basis for our ground truth and allowed a quantitative evaluation of our models on the city of Beria, Mozambique.  ... 
doi:10.5194/isprs-archives-xliii-b3-2021-741-2021 fatcat:yw7aihv3onetbb5i4mkipsespi

FloodNet: A High Resolution Aerial Imagery Dataset for Post Flood Scene Understanding

Maryam Rahnemoonfar, Tashnim Chowdhury, Argho Sarkar, Debvrat Varshney, Masoud Yari, Robin Murphy
2021 IEEE Access  
On the other hand the primary source of the ground-level imageries is social media [30] , these imageries lack geo-location tags [27] and suffers from data scarcity for deep learning training [11]  ...  [21] used crowd sourced images from social media which captured disaster sites from the ground level.  ... 
doi:10.1109/access.2021.3090981 fatcat:wacd4jsapzfuhhu5kjvkl3ab2m
« Previous Showing results 1 — 15 out of 45,774 results