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 ›› 2021, Vol. 40 ›› Issue (9): 1276-1280.

• • 上一篇    下一篇

基于红外图像的高温点定位跟踪算法研究

黄琼1，司颖1，王浩宇1，丁兆云2   

  1. 1. 中国消防救援学院，北京102202；2. 国防科技大学，湖南长沙410073
  • 出版日期:2021-09-15 发布日期:2021-09-15
  • 作者简介:黄琼(1972-)，女，江西丰城人，中国消防救援学院基础部信息技术教研室讲师，硕士，主要从事数据挖掘与机器学习、安全工程方面的研究，北京市昌平区南雁路4 号，102202。
  • 基金资助:
    国家自然科学基金资助项目(91846301)

Research of high temperature point location and tracking algorithm based on infrared image

HUANG Qiong1， SI Ying1，WANG Hao-yu1， DING Zhao-yun2   

  1. 1. China Fire and Rescue Institute，Beijing 102202, China;2. National University of Defense Technology, Hunan Changsha 410073, China
  • Online:2021-09-15 Published:2021-09-15

摘要： 提出一种基于红外图像的高温点定位跟踪算法，并将该算法实现于基于FPGA和红外探测器技术的智能消防头盔中，通过红外探测器收集火场温度图像信息，实现火场高温点的定位与跟踪，并通过所获取的高温点位置进行信息共享，以有效提高协同灭火战斗能力，及时高效地扑灭火灾。实验结果表明，基于该算法实现的智能消防头盔能够达到较为良好的高温点定位效果，且在搜寻过程中具有较小的目标偏移距离和较短的目标搜索时间，高温点位置判断的准确率相对较高，能够为灭火指挥与救援战斗提供有效而可信的辅助信息。

关键词: 高温点定位, 跟踪, FPGA, 红外探测器

Abstract: A positioning and tracking algorithm based on infrared image is proposed, and the algorithm is implemented in the intelligent fire helmet based on FPGA and infrared detector technology. The infrared detector collects the information of fire temperature image, realizes the location and tracking of fire high temperature points, and shares the information through the obtained high temperature points, so as to effectively improve the cooperative fire fighting ability and make it more timely and efficient to put out the fire in a good way. The experimental results show that the intelligent fire helmet can achieve better positioning effect of high temperature points, and has smaller target offset distance and shorter target search time in the search process. The accuracy of high temperature point position judgment is relatively high, which can provide effective and reliable auxiliary information for fire fighting command and rescue combat.