Datasets
Standard Dataset
Natural Disaster Scene
- Citation Author(s):
- Submitted by:
- Yuxuan Shao
- Last updated:
- Sat, 09/21/2024 - 09:37
- DOI:
- 10.21227/dmbe-j294
- License:
71 Views- Categories:
- Keywords:
Abstract
This dataset contains a collection of images documenting various natural disasters. The images are sourced from multiple events worldwide, capturing the impact, devastation, and aftermath of these disasters. Each image is labeled with relevant metadata, such as the type of disaster, location, and date of occurrence. The dataset is designed to support research and development in fields such as disaster management, image recognition, machine learning, and humanitarian aid. It aims to facilitate the creation of predictive models and enhance response strategies for future natural disasters.（The specific data set will be provided after the paper is officially published.）
This dataset comprises images from 11 distinct natural disaster scenarios. Each disaster scenario contains over 300 images, capturing various stages of the events, such as the occurrence, damage, and aftermath. The dataset is systematically divided into a training set and a testing set, following an 80:20 split, to facilitate the development and evaluation of machine learning models. This division ensures robust model training and performance assessment. The dataset is intended for use in disaster management, image classification, and predictive modeling, aiming to enhance disaster preparedness and response strategies.
Comments
Data has been beneficial for my college research.
Starting my learning of image processing