主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2024, Vol. 43 ›› Issue (2): 183-188.

• • 上一篇    下一篇

基于FireNet的古建筑火灾检测方法研究及改进

陈庆典1, 钟晨2, 刘慧1,3, 王晓辉1   

  1. （1. 北京建筑大学 电气与信息工程学院，北京 102616；2. 应急管理部沈阳消防研究所，辽宁 沈阳 110034；3. 建筑大数据智能处理方法研究北京市重点实验室，北京 102616）
  • 出版日期:2024-02-15 发布日期:2024-02-15
  • 作者简介:陈庆典（1999- ），女，北京建筑大学电气与信息工程学院控制科学与工程专业硕士研究生，主要从事深度学习和火灾检测方面的研究，北京市大兴区黄村镇永源路15号，102616。
  • 基金资助:
    国家重点研发计划课题（2020YFC1522804）

Research and improvement of fire detection method for historical buildings based on FireNet

Chen Qingdian1, Zhong Chen2, Liu Hui1,3, Wang Xiaohui1   

  1. (1. School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing 102616, China;2. Shenyang Fire Science and Technology Research Institute of EME, Liaoning Shenyang 110034, China; 3. Beijing Key Laboratory of Intelligent Processing for Building Big Data, Beijing University of Civil Engineering and Architecture, Beijing 102616, China)
  • Online:2024-02-15 Published:2024-02-15

摘要： 针对古建筑火灾检测需要快速、准确及实时的需求，建立了一个专门用于古建筑火灾检测的数据集，用于古建筑火灾检测的深度学习研究。利用CBAM注意力机制模块，结合多尺度特征融合，对FireNet网络进行改进，提出适用于古建筑火灾检测的轻量级FireNet-AMF网络，在FireNet数据集和本文构建的古建筑火灾检测数据集上验证了FireNet-AMF网络的火灾检测能力。与改进前的网络相比，FireNet-AMF网络在FireNet数据集上对火灾识别的准确率达到了95.08%，与原网络相比提高了1.17%，在本文构建的古建筑火灾检测数据集上的准确率达到了95.62%，比原网络提高了1.62%。该网络在保证轻量级的同时也保证了在古建筑火灾检测中较高的检测精度。

关键词: 古建筑, 火灾检测, 图像分类, FireNet, 注意力机制, 多尺度特征融合

Abstract: In response to the need for fast, accurate, and real-time fire detection of historical buildings, this paper builds a dataset specifically for historical building fire detection, which is used for deep learning in historical building fire detection for the first time. By fusing the CBAM attention mechanism and combining it with multi-scale feature fusion, we improve and propose the FireNet-AMF network based on the FireNet network. The fire detection capability of the FireNet-AMF network is verified on the FireNet dataset and the historical building fire detection dataset. The FireNet-AMF network achieves an accuracy of 95.08% for fire detection with the FireNet dataset, an improvement of 1.17% compared to the FireNet network, and an accuracy of 95.62% for experiments on the historical building fire detection dataset we built, which is an improvement of 1.62% compared to the FireNet network. The network ensures a light weight while guaranteeing a high level of historical building fire detection accuracy.

Key words: historical building, fire detection, image classification, FireNet, attention mechanism, multi-scale feature fusion