Evaluating Building Damage Classification Accuracy: A Benchmarking Study of UNet

Title

Evaluating Building Damage Classification Accuracy: A Benchmarking Study of UNet

Creator

Ikkara, Sajitha; Sambandam, Rakoth Kandan; John, Saju P.

Description

Building damage classification must be done accurately and quickly in order to support disaster response and recovery activities. Deep learning models, particularly U-Net, have demonstrated strong potential in automating damage assessment from satellite and aerial imagery. This study benchmarks the accuracy of U-Net in classifying building damage across multiple datasets, evaluating its performance against ground truth labels. Key factors such as data preprocessing, augmentation techniques, and model variations are analyzed to determine their impact on classification accuracy. The results provide insights into the strengths and limitations of variations in U-Net for damage assessment, highlighting areas for improvement and future research directions 2025 IEEE.

Source

2025 IEEE International Conference on Next-Gen Technologies of Artificial Intelligence and Geoscience Remote Sensing, EarthSense 2025;

Date

01-01-2025

Publisher

Institute of Electrical and Electronics Engineers Inc.

Subject

augmentation; classification; damage assessment; Deep learning; satellite images

Coverage

Ikkara S., Christ University, Dept. of Computer Science and Engineering, Karnataka, Bangalore, India; Sambandam R.K., Jyothi Engineering College, Dept. of Computer Science and Engineering, Kerala, Thrissur, India; John S.P., Christ University, Dept. of Computer Science and Engineering, Karnataka, Bangalore, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISBN: 979-833150288-1;

Format

online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1109/EarthSense66084.2025.11297371

https://www.scopus.com/pages/publications/105032199046?origin=resultslist