PDD: Post-Disaster Dataset for Human Detection and Performance Evaluation

doi:10.1109/TIM.2023.3346508

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	PDD: Post-Disaster Dataset for Human Detection and Performance Evaluation
	Song, Haoqian 1,2,4; Song, Weiwei 4; Cheng, Long 2,4; Wei, Yue 3; Cui, Jinqiang 4
刊名	IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
	2024
卷号	73 页码:14
关键词	Detection algorithms Remote sensing YOLO Cameras Autonomous aerial vehicles Convolutional neural networks Search problems Human detection multiview image performance evaluation post-disaster ruins scene unmanned aerial vehicle (UAV) search and rescue (SAR)
ISSN号	0018-9456
DOI	10.1109/TIM.2023.3346508
通讯作者	Cheng, Long(long.cheng@ia.ac.cn)
英文摘要	Human detection is aimed at automatically labeling specific semantic objects in high-resolution images, which is a key problem in the post-disaster search and rescue (SAR) mission with unmanned aerial vehicles (UAVs). However, for large-scale and efficient search and statistics with UAVs, there is a lack of real-world image datasets of post-disaster ruins, resulting in poor performance in practical applications. In this article, we conducted a comprehensive review of current studies and created a post-disaster dataset (PDD) for human detection first, which contains the real-world post-disaster UAV images from multiple scenes, perspectives, and distances. Second, based on the YOLOv5 algorithm, we verified the validity and superiority of PDD by comparing and analyzing its detection performance with other datasets and on different training proportions. Finally, we compared and evaluated the detection performance of 11 state-of-the-art algorithms on PDD, including faster region-based convolutional neural networks (R-CNN), YOLOv5, YOLOv7, YOLOv8, improved YOLOv5 (im-YOLOv5), neural architecture search (YOLO-NAS), detection transformer (DETR), deformable DETR (DDETR), dynamic anchor box DETR (DAB-DETR), denoising DETR (DN-DETR), and DETR with improved denoising anchor box (DINO), and provided a performance analysis of different deep learning algorithms. The results demonstrate that you only look once (YOLO)-based algorithms have a better real-time statistical performance, while the DETR-based algorithms have more accurate box prediction capabilities. The PDD, codes, and models are available at https://github.com/HaoqianSong/Post-Disaster-Dataset.
资助项目	Peng Cheng Laboratory Research Project
WOS关键词	R-CNN ; REAL-TIME ; SURVEILLANCE ; DRONE
WOS研究方向	Engineering ; Instruments & Instrumentation
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:001138781000027
资助机构	Peng Cheng Laboratory Research Project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/56996]
专题	多模态人工智能系统全国重点实验室
通讯作者	Cheng, Long
作者单位	1.Chinese Acad Sci, Inst Automat, State Key Lab Multimodal Artificial Intelligence, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 3.Guangdong Lab Artificial Intelligence & Digital E, Shenzhen 518107, Peoples R China 4.Peng Cheng Lab, Shenzhen 518066, Peoples R China
推荐引用方式 GB/T 7714	Song, Haoqian,Song, Weiwei,Cheng, Long,et al. PDD: Post-Disaster Dataset for Human Detection and Performance Evaluation[J]. IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,2024,73:14.
APA	Song, Haoqian,Song, Weiwei,Cheng, Long,Wei, Yue,&Cui, Jinqiang.(2024).PDD: Post-Disaster Dataset for Human Detection and Performance Evaluation.IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT,73,14.
MLA	Song, Haoqian,et al."PDD: Post-Disaster Dataset for Human Detection and Performance Evaluation".IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 73(2024):14.