Ai-based Damage Detection Of Buiding Using Drone Imagery

Abstract

Unmanned Aerial Vehicles (UAVs), Commonly Known As Drones, Equipped With High Resolution Cameras Generate Vast Amounts Of Structural Imagery Useful For Assessing Building Damage After Earthquakes, Storms, Or Explosions. This Paper Presents A Comprehensive Framework For Automated Damage Detection Using Deep Learning Models Trained On Drone Imagery. We Combine Computer Vision Pipelines With Convolutional Neural Networks (CNNs) For Accurate Pixel Level And Object Level Damage Classification. Experimental Results On Benchmark Datasets Show High Accuracy, Demonstrating The Effectiveness Of AI Based Methods For Rapid Post Disaster Building Evaluation.

Keywords