| 李宁,刘青,熊俊,董力文.基于无人机的火灾检测系统设计与实现[J].火灾科学,2022,31(1):46-51. |
| 基于无人机的火灾检测系统设计与实现 |
| Design and implementation of fire detection system based on UAV |
| |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI：10.3969/j.issn.1004-5309.2022.01.06 |
| 基金项目: |
|
| 中文关键词: UAV 火灾检测 改进 YOLOv3 |
| 英文关键词:UAV Fire detection Improved-YOLOv3 |
| 摘要点击次数: 492 |
| 全文下载次数: 5151 |
| 中文摘要: |
| 针对传统森林火灾检测手段响应速度慢、效率低、误报率高等问题,设计了无人机搭载的由云台和相机组成的图像采集平台,通过火灾智能识别技术,实时识别监测火灾的发生,并达到了自动抵近侦察及实时态势感知的效果。在火灾智能识别算法方面,提出了improved-YOLOv3算法,通过在特征交互阶段增加yolo层,加强了网络对特征的融合度,从而增加了网络的检测能力。通过与性能相似的网络进行对比测试,验证了改进算法的有效性。测试结果表明,提出的算法检测准确率高、漏检率低、推理速度快,能够适用于实际火灾现场监测。 |
| 英文摘要: |
| Given the problems of slow response speed, low efficiency and high false alarm rate of traditional forest fire detection methods, this paper designs and realizes the image acquisition platform based on the pan-tilt and camera carried by UAV. The fire intelligent recognition technology can identify and monitor the occurrence of fire in real time and realize the functions of automatic approach reconnaissance and situation awareness. In the intelligent fire identification algorithm, the improved Yolo layer is added in the feature interaction stage, and the fusion degree of the network to the feature is strengthened. Thus the detection ability of the network is increased. The effectiveness of the proposed algorithm is verified by comparing it with a network with similar performance. The test results show that the algorithm proposed in this paper has high detection accuracy, low leakage rate and fast reasoning speed, which can be applied to the actual fire monitoring site. |
| 关闭 |
|
|
|
