Damage parameter estimation for ancient DNA

Flood mapping and damage assessment in GEE

MATLAB code to generate some results on the paper (Kitayama, & Cilsalar, 2022).

This repository houses a comprehensive deep learning-based system for the scene classification of very high-resolution satellite imagery, focusing on post-earthquake damage assessment. The primary case study involves the devastating earthquakes that occurred in Kahramanmaraş province, Türkiye, in 2023.

Building machine learning models that predict the damage level caused by an earthquake on a building using the Nepal's Earthquake dataset

Multi-Head Convolutional Networks for Post Earthquake Damage Assessment of Damaged Buildings

This repository provides the SeAn-CP method, a semi-analytical algorithm for efficiently calculating critical plane factors in fatigue analysis. Compatible with finite element analysis, it handles complex geometries and loading conditions, offering a faster alternative to traditional plane scanning with similar accuracy.

Guided Team Challenge 2021: Exposure Team Project

Post-Disaster Damage Assessment

AIM Group® Worldwide is an third party inspection and 3rd maritime professional specialists group independent acting globally in close to 100 countries.

Deployment-ready pipeline for disaster type and damage classification using before-and-after satellite imagery.

Geospatial analysis of flood impacts in Somalia, using GIS and spatial statistics.

The study uses YOLO models on satellite imagery to detect collapsed buildings in Antakya after the 2023 Türkiye earthquake, achieving good results with YOLOv7 but highlighting challenges.

AIMGroup is marine cargo ship surveyors and consultant with expertise on class and risk managing on seaworthiness and tecnical condition.

KML File and script to generate the file to show the official damage assessment from CAL FIRE for the 2025 Palisades Fire for use with Google Earth Pro, possibly other GIS software too.

Evaluating Mayotte disaster damages: application of transfer learning on a model trained by RescueNet on data collected from the French island of Mayotte after Cyclone Chido and evaluation of the generalizability of this model.