Author Affiliation

Department of Electrical and Computer Engineering, Old Dominion University

Location

Virginia Modeling, Analysis and Simulation Center, Room 1201

Conference Title

Modeling, Simulation and Visualization Student Capstone Conference 2023

Conference Track

General Sciences & Engineering

Document Type

Paper

Abstract

Satellite image analysis of natural disasters is critical for effective emergency response, relief planning, and disaster prevention. Semantic segmentation is believed to be on of the best techniques to capture pixelwise information in computer vision. In this work we will be using a U-Net architecture to do a three class semantic segmentation for the Xview2 dataset to capture the level of damage caused by different natural disaster which is beyond the visual scope of human eyes.

Keywords:

Natural disaster, Damage assessment, Semantic segmentation, U-Net

Start Date

4-20-2023

End Date

4-20-2023

Share

COinS
 
Apr 20th, 12:00 AM Apr 20th, 12:00 AM

U-Net Based Multiclass Semantic Segmentation for Natural Disaster Based Satellite Imagery

Virginia Modeling, Analysis and Simulation Center, Room 1201

Satellite image analysis of natural disasters is critical for effective emergency response, relief planning, and disaster prevention. Semantic segmentation is believed to be on of the best techniques to capture pixelwise information in computer vision. In this work we will be using a U-Net architecture to do a three class semantic segmentation for the Xview2 dataset to capture the level of damage caused by different natural disaster which is beyond the visual scope of human eyes.