An Ensemble Model for YOLO Algorithm for Object Detection in Low Visibility in Disaster Management: A Survey Paper
Year: 2025
Type: article
Abstract: : Object detection is very important in crisis management since it helps with speedy assessment, finding victims, and distributing resources. Natural catastrophes often make it hard to see because of smoke, dust, fog, heavy rain, or debris. This makes typical object detection algorithms much less efficient. The You Only Look Once (YOLO) family of algorithms is a good choice for these kinds of circumstances because they can process data in real time. This survey paper goes into great detail about the problems of detecting objects in low visibility during disasters. It also looks into the idea of using ensemble models based on YOLO algorithms to get past these problems. We look into several ensemble methods, picture enhancing methods, and reliable feature extraction methods that can be used with YOLO to make it more accurate and reliable in difficult visual settings. To move the subject of object detection for disaster management in low-visibility circumstances forward, the study also looks at current research trends, datasets, assessment criteria, and critical future research topics.
Authors: ankur mishra, ankur khare
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4413435904