DOIONLINE 
DOIONLINE Publish In |
International Journal of Advances in Electronics and Computer Science-IJAECS |
 Journal Home Volume Issue |
||||||||
Issue |
Volume-8,Issue-10 ( Oct, 2021 ) | |||||||||
Paper Title |
A Variety of Deep Learning Models to Classify Disaster Scene Videos | |||||||||
Author Name |
Haili Wang, Yuan Li, Bill Buckles | |||||||||
Affilition |
Department of Computer Science and Engineering University of North Texas, USA | |||||||||
Pages |
7-11 | |||||||||
Abstract |
The increasing frequency and severity of natural disasters has become more and more prevalent today. The fifth assessment report of the Intergovernmental Panel on Climate Change (IPCC, 2014) predicts that as global warming continues in the coming decades, its contribution to the increase in natural disaster losses will become more prominent. However, through rapid and accurate analysis of disaster scenarios, there is still an opportunity to significantly reduce catastrophic losses caused by extreme events. From a video, we extract key frames and identify embedded objects (using YOLOv3). The densely labeled images are given a global label using various VGG and ResNet tools. Classical quality measures (accuracy, precision, and recall) will guide subsequent development directions. Keywords - Disaster Management, Deep Learning, Object Detection, Scene Classification | |||||||||
| View Paper | ||||||||||
