Flood and Non-Flood Image Classification using Deep Ensemble Learning

Ellora Yasi

doi:10.1007/s11269-024-03906-9

2024-06-07

PRE

DOI：10.1007/s11269-024-03906-9

原文链接

原文求助

摘要

En 中文

Floods are one of the most frequent natural disasters, often resulting in widespread devastation. Identifying floods accurately is crucial for disaster management as it helps to locate areas requiring urgent assistance and streamline post-flood evacuation processes. Recently, deep learning models, such as Convolutional Neural Networks (CNN), have become predominant for image classification tasks, as well as flood classification problems. Deep ensemble techniques,i.e. combining several deep learning architectures, are still quite new in many fields and have not been studied extensively despite showing promising results in flood classification. In this research, we develop an ensemble deep learning framework that utilizes eight state-of-the-art CNN architectures, namely MobileNet V2, ResNet 50, VGG 16, DenseNet 201, Inception V3, EfficientNet B5, NasNet Large, and Xception. The aim is to address the gap of deep ensemble learning in flood classification and provide a more effective approach to identifying potential flooding scenarios from a wide range of visual datasets. We utilize FloodNet and flood area segmentation datasets to train, test, and validate our models. In the testing phase, our ensemble model outperforms several individual benchmark models, achieving a training accuracy of 98.9% and a test accuracy of 97.4%. Our proposed methodology will predict floods and conduct early assessments of affected areas efficiently.

Keyword:

Deep ensemble

Transfer learning

Data augmentation

Deep learning

Image classification

Convolutional neural networks (CNN)

Natural disaster management

Flood

期刊

Water Resources Management

IF:

4.7

论文数:

8.1K

被引数:

1.6W

机构

暂无机构信息

引用论文

Pre-race self-reported medical conditions and allergies in 133 641 Comrades ultramarathon (90km) runners - SAFER XXIII 赛前自我报告的医学状况和过敏史在133641名 comrades ultramarathon（90km）跑者中的分布 - SAFER XXIII

The Physician and Sportsmedicine

IF0

2021-11-29

OAAI

Stephan Brill; Martin Schwellnus; Nicola Sewry; Dina Christa Janse Van Rensburg; Audrey Jansen Van Rensburg; Marcel Jooste; Jordan Leppan; Jeremy Boulter; Ishen Seocharan; Esme Jordaan

Deep learning methods for flood mapping: a review of existing applications and future research directions

HYDROLOGY AND EARTH SYSTEM SCIENCES

IF5.8

2022-08-25

144

OAAI

Bentivoglio, Roberto; Isufi, Elvin; Jonkman, Sebastian Nicolaas; Taormina, Riccardo

A novel deep learning neural network approach for predicting flash flood susceptibility: A case study at a high frequency tropical storm area

SCIENCE OF THE TOTAL ENVIRONMENT

IF8

2020-01-01

279