基于红外与可见光图像火灾定位研究  

Research on Fire Location Based on Infrared and Visible Images

作　　者：段琳锋 侯新国[1] DUAN Linfeng;HOU Xinguo(School of Electrical Engineering,Naval University of Engineering,Wuhan 430033)

机构地区：[1]海军工程大学电气工程学院,武汉430033

出　　处：《舰船电子工程》2022年第3期183-187,共5页Ship Electronic Engineering

基　　金：国家自然科学基金项目(编号:41774021)资助。

摘　　要：火灾定位是火灾探测到火灾自动扑救的重要环节,在红外与可见光图像火灾定位中,常规的可见光摄像机标定板无法直接用于红外摄像机标定。针对红外与可见光摄像机联合标定问题,利用不同材料的受热差异,该文设计了可同时标定红外摄像机与可见光摄像机的标定板,并利用Matlab中的摄像机模块对红外摄像机和可见光摄像机进行了标定。标定结果重投影误差为0.92像素,达到亚像素级别,为红外摄像机与可见光摄像机的标定提供了简单有效的标定方法。最后,采集不同位置的火灾视频,并进行了火灾定位实验,得出实验结果相对误差在7%以内,并对误差原因进行了分析。Fire location is an important part of fire detection and automatic fire fighting.In infrared and visible image fire location,the conventional visible camera calibration board can not be directly used for infrared camera calibration.Aiming at the problem of joint calibration of infrared and visible cameras,the heating difference of different materials is used,a calibration board which can calibrate infrared and visible cameras at the same time is designed,and the infrared and visible cameras are calibrated by using the camera module in Matlab.The re projection error of the calibration result is 0.92 pixel,reaching the sub-pixel level,which provides a simple and effective calibration method for the calibration of infrared camera and visible camera.Finally,fire videos from different locations are collected,and fire location experiments are carried out.The relative error of the experimental results is less than 7%,and the causes of the error are analyzed.

关 键 词：红外摄像机 可见光摄像机 标定板 火灾定位 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置] TU998.13[自动化与计算机技术—控制科学与工程]