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 ›› 2023, Vol. 42 ›› Issue (12): 1709-1713.

• • 上一篇    下一篇

飞机货舱火焰可见和红外多特征融合图像探测算法研究

邓力1,2, 谢爽爽1, 贺元骅1,2, 刘全义1,2   

  1. （1. 宁波市消防救援支队，浙江 宁波 315000；2. 霍尼韦尔智能建筑科技集团，广东 深圳 518066）
  • 出版日期:2023-12-15 发布日期:2023-12-15
  • 作者简介:邓 力（1986- ），男，中国民用航空飞行学院民航安全工程学院副教授，工学博士，硕士生导师，主要从事民航安全工程技术、机载火灾探测技术、信息处理及嵌入式开发等方面研究，四川省广汉市南昌路四段46号，618307。
  • 基金资助:
    民机火灾科学与安全工程四川省重点实验室项目（MZ2022JB01）；德阳市科技局重点研发项目（2021SZ001）

Research on flame visible and infrared mult-feature fusion image detection algorithm in aircraft cargo compartment

Deng Li1,2, Xie Shuangshuang1, He Yuanhua1,2, Liu Quanyi1,2   

  1. (1.College of Civil Aviation Safety Engineering, Civil Aviation Flight University of China, Sichuan Guanghan 618307, China;2. Civil Aircraft Fire Science and Safty Engineering Key Laboratory of Sichuan Province, Sichuan Guanghan 618307, China)
  • Online:2023-12-15 Published:2023-12-15

摘要： 以环己烷和航空煤油作为试验燃料，在模拟飞机货舱内开展油池火灾燃烧探测试验，结合图像相关性原理开展燃烧图像识别及火焰振荡频率检测，并以此作为是否发生火灾的判断标准。结果表明：在低于60 kPa情况下，火焰温度变化明显，且抖动严重；环己烷与航空煤油在等间距燃烧时间序列下，其平均火焰振荡频率大于3 Hz，与理论大小一致并可作为判断火灾发生的依据；通过YOLO模型检测环己烷以及航空煤油燃烧红外图像，判断是否有火焰产生，精确率高达0.965。试验结果表明，本文方法对火焰检测具有良好的效果，可为低压环境下可燃物燃烧火行为分析与火灾探测预警关键技术提供理论及数据支持。

关键词: 飞机货舱, 火灾探测, 图像相关性, 火焰振荡频率, 红外图像

Abstract: Cyclohexane and aviation kerosene are used as experimental fuels to carry out pool fire detection experiments in the aircraft cargo model. Combined with the principle of image correlation, combustion image recognition and flame oscillation frequency detection are carried out as the criteria for judging whether a fire occurs. The results indicate that the flame temperature varies significantly when the temperature is below 60 kPa and the jitter is severe; Under equidistant combustion time series, the mean flame oscillation frequency of cyclohexane and aviation kerosene is above 3 Hz, which is in agreement with the theoretical value and can be used as the basis for judging the occurrence of fire. The YOLO model is used to detect the infrared images of cyclo—hexane and aviation kerosene combustion to determine whether there is flame generation, and the accuracy is as high as 0.965. The experimental results show that the proposed method has a good effect on flame detection, which can be used in theory and data to analyze the fire behavior of combustion and the key techniques of fire detection and early warning in low—pressure environment.

Key words: aircraft cargo;fire detection, imagery correlation, flame oscillation frequency, infrared image