BLDNet: A Semi-supervised Change Detection Building Damage Framework using Graph Convolutional Networks and Urban Domain Knowledge (2022-01-25T00:00:00.000000Z)

TL;DR

BLDNet is presented, a novel graph formulation for building damage change detection and enable learning relationships and representations from both local patterns and non-stationary neighborhoods in a semi-supervised framework with few annotated data.

Abstract

Change detection is instrumental to localize damage and understand destruction in disaster informatics. While convolutional neural networks are at the core of recent change detection solutions, we present in this work, BLDNet, a novel graph formulation for building damage change detection and enable learning relationships and representations from both local patterns and non-stationary neighborhoods. More specifically, we use graph convolutional networks to efficiently learn these features in a semi-supervised framework with few annotated data. Additionally, BLDNet formulation allows for the injection of additional contextual building meta-features. We train and benchmark on the xBD dataset to validate the effectiveness of our approach. We also demonstrate on urban data from the 2020 Beirut Port Explosion that performance is improved by incorporating domain knowledge building meta-features.

Authors

Ali Ismail

1 papers

M. Awad

1 papers

References66 items

Building damage assessment for rapid disaster response with a deep object-based semantic change detection framework: From natural disasters to man-made disasters

Semisupervised Change Detection Using Graph Convolutional Network

Transferability of Convolutional Neural Network Models for Identifying Damaged Buildings Due to Earthquake

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

TL;DR

Abstract

Authors

References66 items

Building damage assessment for rapid disaster response with a deep object-based semantic change detection framework: From natural disasters to man-made disasters

Semisupervised Change Detection Using Graph Convolutional Network

Transferability of Convolutional Neural Network Models for Identifying Damaged Buildings Due to Earthquake

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

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

Technical Solution Discussion for Key Challenges of Operational Convolutional Neural Network-Based Building-Damage Assessment from Satellite Imagery: Perspective from Benchmark xBD Dataset

Assessing out-of-domain generalization for robust building damage detection

Assessment of Convolutional Neural Network Architectures for Earthquake-Induced Building Damage Detection based on Pre- and Post-Event Orthophoto Images

Preliminary yield estimation of the 2020 Beirut explosion using video footage from social media

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

SemiCDNet: A Semisupervised Convolutional Neural Network for Change Detection in High Resolution Remote-Sensing Images

Graph Convolutional Networks for Hyperspectral Image Classification

Deep Learning-Based Identification of Collapsed, Non-Collapsed and Blue Tarp-Covered Buildings from Post-Disaster Aerial Images

Deep Learning for Change Detection in Remote Sensing Images: Comprehensive Review and Meta-Analysis

Object-based change detection for VHR remote sensing images based on a Trisiamese-LSTM

Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges

Nonlocal Graph Convolutional Networks for Hyperspectral Image Classification

Building Disaster Damage Assessment in Satellite Imagery with Multi-Temporal Fusion

Change Detection in Multisource VHR Images via Deep Siamese Convolutional Multiple-Layers Recurrent Neural Network

Machine learning for buildings’ characterization and power-law recovery of urban metrics

PGA-SiamNet: Pyramid Feature-Based Attention-Guided Siamese Network for Remote Sensing Orthoimagery Building Change Detection

A novel hybrid machine learning approach for change detection in remote sensing images

Structural Building Damage Detection with Deep Learning: Assessment of a State-of-the-Art CNN in Operational Conditions

xBD: A Dataset for Assessing Building Damage from Satellite Imagery

Building Damage Detection in Satellite Imagery Using Convolutional Neural Networks

Toward Generalized Change Detection on Planetary Surfaces With Convolutional Autoencoders and Transfer Learning

Graph convolutional network for multi-label VHR remote sensing scene recognition

Change Detection of Surface Water in Remote Sensing Images Based on Fully Convolutional Network

Improving plankton image classification using context metadata

Siamese graph convolutional network for content based remote sensing image retrieval

End-to-End Change Detection for High Resolution Satellite Images Using Improved UNet++

Building Instance Change Detection from Large-Scale Aerial Images using Convolutional Neural Networks and Simulated Samples

Improve Image Classification Tasks Using Simple Convolutional Architectures with Processed Metadata Injection

Automatic Post-Disaster Damage Mapping Using Deep-Learning Techniques for Change Detection: Case Study of the Tohoku Tsunami

Detecting Large-Scale Urban Land Cover Changes from Very High Resolution Remote Sensing Images Using CNN-Based Classification

Transferred Deep Learning for Sea Ice Change Detection From Synthetic-Aperture Radar Images

Multi-Label Image Recognition With Graph Convolutional Networks

Change detection techniques for remote sensing applications: a survey

Fast Graph Representation Learning with PyTorch Geometric

Land-Use Change Detection with Convolutional Neural Network Methods

Sea Ice Change Detection in SAR Images Based on Convolutional-Wavelet Neural Networks

Change detection based on Faster R-CNN for high-resolution remote sensing images

Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networks

Semi-Supervised Clustering With Multiresolution Autoencoders

Learning Spectral-Spatial-Temporal Features via a Recurrent Convolutional Neural Network for Change Detection in Multispectral Imagery

Change Detection Based on Deep Siamese Convolutional Network for Optical Aerial Images

Forest Change Detection in Incomplete Satellite Images With Deep Neural Networks

Automatic recognition of landslide based on CNN and texture change detection

Semi-Supervised Classification with Graph Convolutional Networks

Learning a Transferable Change Rule from a Recurrent Neural Network for Land Cover Change Detection

Deep Residual Learning for Image Recognition

A new approach for surface water change detection: Integration of pixel level image fusion and image classification techniques

Adam: A Method for Stochastic Optimization

Improved forest change detection with terrain illumination corrected Landsat images

Detection of urban expansion in an urban-rural landscape with multitemporal QuickBird images

A Split-Based Approach to Unsupervised Change Detection in Large-Size Multitemporal Images: Application to Tsunami-Damage Assessment

Use of Remote Sensing Applications and its Implications to the Society

Multi‐temporal Mapping of Burned Forest over Canada Using Satellite‐based Change Metrics

Monitoring large areas for forest change using Landsat: Generalization across space, time and Landsat sensors

Two algorithms for constructing a Delaunay triangulation

Individual Comparisons by Ranking Methods

“Probable Error of a Mean, The”

Lebanon: Beirut port explosions - aug

High contrast: We chart beirut’s ever-changing architecture scene, Aug 2018

Visualizing Data using t-SNE

A Mathematical Theory of Communication

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names