Have a personal or library account? Click to login
Remote Sensing Building Damage Assessment Based on Machine Learning Cover
Open Access
|Sep 2024

Figures & Tables

Confusion matrix formal table

Prediction categoryTrue categoryPositive sampleNegative sample
Positive sampleTPFP
Negative sampleFNTN

Based on the building damage level table defined in this article

ClassDescription
0Undamaged
1Minor damage
2Major damage
3Destroyed

Training results on validation dataset

NameExplanationColor
F1The overall F1 value of the building damage assessment on the xBD validation setYellow
F1_LocF1 values for segmentation of building localization on the xBD validation setPurple
F1_DamF1 value for building damage classification on the xBD validation setGreen
F1_UndamF1 value for classification of undamaged buildings on the xBD validation setGrey
F1_MinF1 value for classification of minor damage buildings on the xBD validation setBlue
F1_MajF1 value for classification of major damage buildings on the xBD validation setOrange
F1_DesF1 value for classification of destroyed buildings on the xBD validation setRed

European disaster committee table for building damage assessment

Masonry ConstructionFortified BuildingsDamage Level
Undamaged
Minor Damaged
Medium Damaged
Major Damage
Destroyed

Training environment configuration table

Configuration informationDetail
Hardware ConfigurationNivdia RTX 3080 12G
LanguagePython 3.8
Main FramePytorch 2.1.0 Cuda11.8
Image information1024×1024 20248 photos
Optimization FunctionAdam
Loss Functioncross entropy loss
Epoch30
Training time12h
Language: English
Page range: 1 - 12
Published on: Sep 30, 2024
Published by: Xi’an Technological University
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2024 Jiawei Tang, Shengquan Yang, Shujuan Huang, Bozhi Xiao, published by Xi’an Technological University
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.