WO2022021275A1 - 一种基于机器视觉的智能化消防炮系统及工作方法 - Google Patents

一种基于机器视觉的智能化消防炮系统及工作方法 Download PDF

Info

Publication number
WO2022021275A1
WO2022021275A1 PCT/CN2020/106082 CN2020106082W WO2022021275A1 WO 2022021275 A1 WO2022021275 A1 WO 2022021275A1 CN 2020106082 W CN2020106082 W CN 2020106082W WO 2022021275 A1 WO2022021275 A1 WO 2022021275A1
Authority
WO
WIPO (PCT)
Prior art keywords
fire
fire monitor
jet trajectory
field
image
Prior art date
Application number
PCT/CN2020/106082
Other languages
English (en)
French (fr)
Inventor
李伟
朱劲松
李贝贝
刘秀梅
林达
潘禄
张帅帅
程恒煜
Original Assignee
中国矿业大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国矿业大学 filed Critical 中国矿业大学
Priority to AU2020461031A priority Critical patent/AU2020461031A1/en
Publication of WO2022021275A1 publication Critical patent/WO2022021275A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A62—LIFE-SAVING; FIRE-FIGHTING
    • A62C—FIRE-FIGHTING
    • A62C31/00—Delivery of fire-extinguishing material
    • A—HUMAN NECESSITIES
    • A62—LIFE-SAVING; FIRE-FIGHTING
    • A62C—FIRE-FIGHTING
    • A62C31/00—Delivery of fire-extinguishing material
    • A62C31/28—Accessories for delivery devices, e.g. supports
    • A—HUMAN NECESSITIES
    • A62—LIFE-SAVING; FIRE-FIGHTING
    • A62C—FIRE-FIGHTING
    • A62C37/00—Control of fire-fighting equipment
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F41—WEAPONS
    • F41F—APPARATUS FOR LAUNCHING PROJECTILES OR MISSILES FROM BARRELS, e.g. CANNONS; LAUNCHERS FOR ROCKETS OR TORPEDOES; HARPOON GUNS
    • F41F1/00—Launching apparatus for projecting projectiles or missiles from barrels, e.g. cannons; Harpoon guns
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
    • G01J5/0014—Radiation pyrometry, e.g. infrared or optical thermometry for sensing the radiation from gases, flames
    • G01J5/0018—Flames, plasma or welding
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D3/00—Control of position or direction
    • G05D3/12—Control of position or direction using feedback

Definitions

  • the invention relates to the field of fire protection, in particular to an intelligent fire monitor system and a working method based on machine vision.
  • fire monitors are also developing in the direction of intelligence.
  • Most of the active fire monitors are operated manually, which have low fire extinguishing efficiency and high requirements for professional experience.
  • Firefighters need a long time of study and practice to master the use of fire monitors, and to give full play to the role of fire monitors in firefighting operations requires high learning costs. Aiming, a lot of work.
  • the invention provides an intelligent fire monitor system and working method based on machine vision, which can intelligently track and aim the fire field during the fire extinguishing operation, quickly and accurately deliver the fire extinguishing agent to the fire area, track the change of the fire field position in real time, and continuously Track fire.
  • the present invention adopts the following technical solutions:
  • An intelligent fire monitor system based on machine vision includes: an infrared camera, a natural light camera, a mechanism bracket, a fire monitor head, a first stepping motor, a second stepping motor, a water inlet pipe, a fire monitor and a central controller.
  • the bottom of the fire monitor is connected to the water inlet pipe, and the first step motor is installed below the fire monitor, and the first step motor drives the fire monitor to do yaw motion.
  • An elbow is installed on the top of the fire monitor, a fire monitor head is installed on the elbow, a second stepper motor is installed on the fire monitor head, and the second stepper motor drives the fire monitor head to make a pitching motion.
  • a mechanism bracket is also installed on the elbow, and an infrared camera and a natural light camera are installed on the mechanism bracket.
  • the optical axis of the infrared camera and the aiming direction of the fire monitor are in the same vertical plane and parallel; the natural light camera is located at the rear and upper side of the fire monitor head.
  • the infrared camera and the natural light camera are connected to the central controller, and the central controller is connected to and controls the first stepping motor and the second stepping motor.
  • the present invention also provides a working method of an intelligent fire monitor system based on machine vision, which is suitable for an intelligent fire monitor system based on machine vision, including:
  • the infrared camera collects the infrared image of the fire field, and the natural light camera collects the near-field image of the jet trajectory of the fire nozzle and sends it to the central controller;
  • the central controller analyzes the center point of the fire field in the infrared image of the fire field, and calculates the deviation between the center point of the fire field and the geometric center point of the infrared image of the fire field in the horizontal direction.
  • the central controller controls the first step motor to adjust the horizontal position of the fire monitor until the center of the fire field The point and the geometric center point of the infrared image of the fire field are located in the same vertical direction;
  • the central controller analyzes the geometric characteristics of the near-field image of the jet trajectory, and calculates the deviation between the geometric characteristics of the near-field image of the jet trajectory and the jet trajectory model.
  • the central controller controls the second stepper motor to adjust the pitch angle of the fire monitor until the jet trajectory is close to The geometric features of the field images perfectly match the jet trajectory model.
  • the analysis method of the center point of the fire field is as follows: the central controller traverses the infrared image of the fire field to find the pixel position of the highest temperature point, and marks the position as the pixel position of the center point of the fire field.
  • the analytical method for the geometrical feature of the near-field image of the jet trajectory is as follows: the central controller adopts an image processing method to analyze the incident angle value of the jet trajectory in the near-field image of the jet trajectory.
  • the Gaussian background subtraction method is used to remove the background noise except the jet trajectory in the near-field image of the jet trajectory;
  • the near-field image of the jet trajectory is obtained and the mean value of the pixel coordinates of each column of the jet trajectory in the image is obtained; the mean value of the pixel coordinates is used to obtain the jet trajectory equation by numerical fitting method;
  • the position of the starting pixel of the trace in the image is substituted for the jet trajectory equation to obtain the incident angle value of the jet trajectory.
  • the method for calculating the deviation between the geometric features of the near-field image of the jet trajectory and the jet trajectory model is as follows: using the incident angle value of the jet trajectory and the range correlation model, the deviation between the geometric characteristics of the near-field image of the jet trajectory and the jet trajectory model is calculated. Specifically include:
  • mapping relationship between the incidence angle of the jet trajectory and the range within the effective working range of the intelligent fire monitor system based on machine vision is obtained through the calibration experiment as a correlation model;
  • the infrared camera adjusts the frame rate of image acquisition according to the change of the fire field to change the level adjustment frequency of the fire monitor, which is used to adapt to the level adjustment of the fire monitor in different angle ranges; the near-field camera adjusts the frame rate of the image acquisition to change The pitch adjustment frequency of the fire monitor is used to adjust the pitch of the fire monitor in different angle ranges.
  • the invention provides an intelligent fire monitor system based on machine vision and a working method thereof, which can realize the intelligent fire extinguishing operation of the fire monitor.
  • the automatic horizontal aiming of the fire monitor for the position of the fire field is realized by adjusting the horizontal angle of the fire monitor, and the deviation correction is continuously performed according to the change of the fire scene; by adjusting the pitch angle of the fire monitor head, the jet trajectory of the fire monitor is realized.
  • Point-to-point coverage of the fire field, and continuous deviation correction according to the change of the fire field so as to achieve fire field targeting, accurate extinguishing and real-time tracking and extinguishing.
  • the invention realizes the intelligent operation in the fire extinguishing process, improves the fire extinguishing operation efficiency, accelerates the fire extinguishing and rescue speed, and further reduces casualties and property losses.
  • Fig. 1 is the system structure schematic diagram of the present invention
  • Fig. 2 is the front view of the system structure of the present invention.
  • Fig. 3 is the system structure side view of the present invention.
  • Fig. 5 is the connection schematic diagram of the central controller
  • Fig. 6 is the working flow chart of the system of the present invention.
  • An embodiment of the present invention provides an intelligent fire monitor system based on machine vision, as shown in FIG. 1 , including: an infrared camera 1, a natural light camera 2, a mechanism bracket 3, a fire monitor head 4, a first step motor 5, The second stepper motor 6, the water inlet pipe 7, and the fire monitor 8.
  • Figure 2 Figure 3 and Figure 4 they are the front view, side view and top view of an intelligent fire monitor system based on machine vision.
  • the infrared camera 1 and the natural light camera 2 are installed above the fire monitor 8 through the mechanism bracket 3 , wherein the infrared camera 1 is arranged just above the rear side of the fire monitor head 4 , and the natural light camera 2 is arranged at the upper right of the rear side of the fire monitor head 4 .
  • the first step motor 5 is installed under the fire monitor 8 and is used to adjust the horizontal angle of the fire monitor 8.
  • the infrared camera 1, the natural light camera 2, the mechanism bracket 3, the fire monitor head 4, and the second step are installed on the fire monitor 8.
  • the motors 6 follow the fire monitor 8 for synchronous horizontal angle adjustment.
  • the second stepper motor 6 is installed in the middle of the fire monitor 8 for adjusting the pitch angle of the fire monitor head 4 .
  • the water inlet pipe 7 is installed at the bottom of the fire monitor 8 .
  • the infrared camera 1, the natural light camera 2, the first stepper motor 5, and the second stepper motor 6 are controlled by the data cable and the fire monitor center respectively.
  • the infrared camera 1 and the natural light camera 2 transmit the collected data to the central controller of the fire monitor, and the central controller of the fire monitor sends control signals to the first stepper motor 5 and the second stepper motor 6.
  • This embodiment also provides a working method of an intelligent fire monitor system based on machine vision.
  • the flowchart is shown in Figure 6, including:
  • the central controller of the fire monitor collects the infrared image of the fire field captured by the infrared camera 1 and the natural light image of the jet trajectory captured by the natural light camera 2.
  • the infrared camera 1 collects the fire scene image and sends it back to the central controller of the fire monitor.
  • the central controller traverses the infrared image of the fire field to find the pixel position of the highest temperature point, and uses this position as the pixel position of the fire field center point in the image.
  • the central controller calculates the pixel deviation value between the pixel position of the highest temperature point and the position of the geometric center point of the image in the horizontal direction.
  • the central controller converts the calculated deviation into a control signal and transmits it to the first step motor 5.
  • the first step motor 5 is controlled according to the The signal adjusts the horizontal angle of the fire monitor 8 until the pixel position of the highest temperature point of the infrared image is vertically aligned with the geometric center point of the image.
  • the frame rate of the image acquisition of the infrared camera 1 can be adjusted to change the horizontal adjustment frequency of the fire monitor 8, which is used to adapt to the horizontal adjustment of the fire monitor in different angle ranges.
  • the natural light camera 2 collects the near-field image of the jet trajectory of the fire monitor and sends it back to the central controller of the fire monitor.
  • the central controller analyzes the jet trajectory incident angle value in the near-field image, and calculates the required incident angle deviation value between the incident angle value in the current near-field image and the jet target based on the correlation model between the jet trajectory incident angle value and the range.
  • the Gaussian background subtraction method is used to remove the background noise except the jet trajectory in the near-field image of the jet trajectory;
  • the deviation between the geometric characteristics of the near-field image of the jet trajectory and the jet trajectory model is calculated, including:
  • mapping relationship between the incidence angle of the jet trajectory and the range within the effective working range of the intelligent fire monitor system based on machine vision is obtained through the calibration experiment as a correlation model;
  • the central controller converts the calculated deviation into a control signal and transmits it to the second stepper motor 6.
  • the second stepper motor 6 adjusts the pitch angle of the fire monitor head 4 according to the control signal, until the incident angle value of the jet trajectory in the near-field image reaches the target of the jet. required angle of incidence.
  • the frame rate of the image acquisition of the natural light camera 2 can be adjusted to change the pitch adjustment frequency of the fire monitor head 4, which is used to adapt to the level adjustment of the fire monitor in different angle ranges.
  • the central controller of the fire monitor After processing and analysis, the central controller of the fire monitor sends control commands to the first stepper motor 5 and the second stepper motor 6 respectively to adjust the horizontal position of the fire monitor 8 and the pitch angle of the fire monitor head 4 respectively.
  • the level adjustment of the fire monitor and the pitch adjustment of the fire monitor are carried out synchronously during the fire extinguishing process.
  • the invention provides an intelligent fire monitor system based on machine vision and a working method thereof, which can realize the intelligent fire extinguishing operation of the fire monitor.
  • the automatic horizontal aiming of the fire monitor for the position of the fire field is realized by adjusting the horizontal angle of the fire monitor, and the deviation correction is continuously performed according to the change of the fire scene; by adjusting the pitch angle of the fire monitor head, the jet trajectory of the fire monitor is realized.
  • Point-to-point coverage of the fire field, and continuous deviation correction according to the change of the fire field so as to achieve fire field targeting, accurate extinguishing and real-time tracking and extinguishing.
  • the invention realizes the intelligent operation in the fire extinguishing process, improves the fire extinguishing operation efficiency, accelerates the fire extinguishing and rescue speed, and further reduces casualties and property losses.

Landscapes

  • Emergency Management (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • General Engineering & Computer Science (AREA)
  • Plasma & Fusion (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Fire-Detection Mechanisms (AREA)
  • Operation Control Of Excavators (AREA)

Abstract

一种基于机器视觉的智能化消防炮系统及工作方法，涉及消防领域。包括视觉伺服机构以及消防炮（8）；视觉伺服机构包括红外相机（1）、自然光相机（2）以及机构支架（3）；视觉伺服机构通过机构支架安装于消防炮上方；消防炮包含消防炮头（4）、进水管（7）以及第一步进电机（5）和第二步进电机（6）；第一步进电机安装于消防炮下方调整消防炮水平角度；第二步进电机安装于消防炮中部位置调整消防炮头俯仰角度。实现在灭火作业过程中对火场的瞄准跟踪，快速、精准地将灭火水剂输送至着火区域，实时跟踪火场位置变化，持续跟踪灭火。

Description

一种基于机器视觉的智能化消防炮系统及工作方法 技术领域
本发明涉及消防领域，尤其涉及一种基于机器视觉的智能化消防炮系统及工作方法。
背景技术
伴随智能化技术的兴起，消防炮也在朝着智能化的方向发展。现役消防炮多为手动操作方式，存在灭火效率低、对于专业经验要求高。消防人员需要长时间的学习和练习，才能熟练掌握消防炮的使用，在灭火作业中充分发挥消防炮的作用，需要较高的学习成本，在灭火作业中也需要消防人员全称参与火场的识别、瞄准，工作量较大。
发明内容
本发明提供一种基于机器视觉的智能化消防炮系统及工作方法，能够在灭火作业中对火场进行智能跟踪瞄准，快速、精准地将灭火水剂输送至着火区域，实时跟踪火场位置变化，持续跟踪灭火。
为达到上述目的，本发明采用如下技术方案：
一种基于机器视觉的智能化消防炮系统，包括：红外相机、自然光相机、机构支架、消防炮头、第一步进电机、第二步进电机、进水管、消防炮及中心控制器。
消防炮底部连接进水管，消防炮的下方安装第一步进电机，第一步进电机带动消防炮做偏航运动。消防炮的顶部安装弯管，弯管上安装消防炮头，消防炮头上安装第二步进电机，第二步进电机带动消防炮头做俯仰运动。
弯管上还安装机构支架，机构支架上安装红外相机和自然光相机，红外相机的光轴和消防炮瞄准方向位于同一竖直平面，并且平行；自然光相机位于消防炮头的后上方侧边位置。
红外相机、自然光相机连接中心控制器，中心控制器连接并控制第一步进电机、第二步进电机。
本发明还提供一种基于机器视觉的智能化消防炮系统的工作方法，适用于的一种基于机器视觉的智能化消防炮系统，包括：
红外相机采集火场红外图像，自然光相机采集消防炮头的射流轨迹近场图像，并发送至中心控制器；
中心控制器解析火场红外图像中的火场中心点，并计算火场中心点与火场红外图像几何中心点在水平方向的偏差，中心控制器控制第一步进电机调整消防炮的水平位置，直至火场中心点与火场红外图像几何中心点位于同一竖直方向；
中心控制器解析射流轨迹近场图像的几何特征，并计算射流轨迹近场图像的几何特征与射流轨迹模型的偏差，中心控制器控制第二步进电机调整消防炮的俯仰角度，直至射流轨迹近场图像的几何特征完全匹配射流轨迹模型。
进一步的，火场中心点的解析方法为：中心控制器遍历火场红外图像找寻最高温度点的像素位置，将该位置标记为火场中心点的像素位置。
进一步的，射流轨迹近场图像的几何特征的解析方法为：中心控制器采用图像处理方法解析所述射流轨迹近场图像中射流轨迹的入射角度值。
具体包括：
采用高斯背景减除方法剔除射流轨迹近场图像除射流轨迹以外的背景噪声；
射流轨迹近场图像并计算获取图像中每一列为射流轨迹的像素点坐标均值；利用像素点坐标均值，采用数值拟合方法获取射流轨迹方程；
迹在图像中起始像素点的位置代入射流轨迹方程计算获取射流轨迹的入射角度值。
进一步的，射流轨迹近场图像的几何特征与射流轨迹模型的偏差计算方法为：利用射流轨迹的入射角度值和射程相关性模型，计算得到射流轨迹近场图像几何特征与射流轨迹模型的偏差。具体包括：
针对基于机器视觉的智能化消防炮系统，通过标定实验获取基于机器视觉的智能化消防炮系统有效工作范围内的射流轨迹入射角度与射程的映射关系作为相关性模型；
在基于机器视觉的智能化消防炮系统的工作过程中，通过插值计算方法，获取消防炮实际所需射程在模型中映射的射流轨迹入射角度值；
对比当前图像中射流轨迹入射角度值与插值计算结果，即为射流轨迹近场图像几何特征与射流轨迹模型的偏差。
进一步的，红外相机根据火场变化情况，调整图像采集帧率以改变消防炮水平调整频率，用于适应不同角度范围的消防炮水平调整；近场相机根据火场变化情况，调整图像采集帧率以改变消防炮俯仰调整频率，用于适应不同角度范 围的消防炮俯仰调整。
本发明的有益效果为：
本发明提供的一种基于机器视觉的智能化消防炮系统及其工作方法，可以实现消防炮的智能化灭火作业。在灭火过程中，通过对消防炮的水平角度调整实现消防炮对于火场位置的自动水平瞄准，并依据火场变化情况，不断进行偏差校正；通过对消防炮头的俯仰角度调整实现消防炮射流轨迹落点对火场的覆盖，并依据火场变化情况，不断进行偏差校正，从而实现了火场瞄准、精准扑灭以及实时跟踪灭火。本发明基于视觉伺服技术，实现在灭火过程中的智能化作业，提升灭火作业效率，加快灭火救援速度，进一步减轻了人员伤亡与财产损失。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它的附图。
图1是本发明的系统结构示意图；
图2是本发明的系统结构正视图；
图3是本发明的系统结构侧视图；
图4是本发明的系统结构俯视图；
图5是中心控制器的连接示意图；
图6是本发明系统的工作流程图；
图中：1-红外相机，2-自然光相机，3-机构支架，4-消防炮头，5-第一步进电机，6-第二步进电机，7-进水管，8-消防炮。
具体实施方式
为使本领域技术人员更好地理解本发明的技术方案，下面结合具体实施方式对本发明作进一步详细描述。
本发明实施例提供一种基于机器视觉的智能化消防炮系统，如图1所示，包括：包括红外相机1、自然光相机2、机构支架3、消防炮头4、第一步进电机5、第二步进电机6、进水管7、消防炮8。
如图2、图3、图4所示，分别是一种基于机器视觉的智能化消防炮系统 的正视图、侧视图以及俯视图。
红外相机1、自然光相机2通过机构支架3安装于消防炮8上方，其中红外相机1布置于消防炮头4后侧正上方，自然光相机2布置于消防炮头4后侧右上方。
第一步进电机5安装于消防炮8下方，用于消防炮8水平角度调整，安装于消防炮8上的红外相机1、自然光相机2、机构支架3、消防炮头4、第二步进电机6均跟随消防炮8作同步水平角度调整。第二步进电机6安装于消防炮8中部位置，用于消防炮头4俯仰角度调整。进水管7安装于消防炮8底部。
如图5所示，一种基于机器视觉的智能化消防炮系统中，红外相机1、自然光相机2、第一步进电机5、第二步进电机6分别通过数据线缆与消防炮中心控制器连接实现信号传输，红外相机1、自然光相机2传送采集的数据至消防炮中心控制器，消防炮中心控制器发出控制信号至第一步进电机5、第二步进电机6。
本实施例还提供一种基于机器视觉的智能化消防炮系统的工作方法，流程图如图6所示，包括：
1)启动灭火程序，开始工作：消防炮中心控制器采集来自红外相机1拍摄的火场红外图像以及自然光相机2拍摄的射流轨迹自然光图像。
2)消防炮水平调整：红外相机1采集到火场图像并回传至消防炮中心控制器。中心控制器遍历火场红外图像找寻最高温度点像素位置，并以该位置作为火场中心点在图像中的像素位置。
中心控制器计算最高温度点像素位置与图像几何中心点位置在水平方向的像素偏差值，中心控制器将计算偏差转换成控制信号传输至第一步进电机5，第一步进电机5依据控制信号调整消防炮8水平角度，直至红外图像最高温度点像素位置与图像几何中心点竖直对齐。上述过程根据火场变化情况，可调整红外相机1图像采集帧率以改变消防炮8水平调整频率，用于适应不同角度范围的消防炮水平调整。
3)消防炮头俯仰调整：自然光相机2采集到消防炮射流轨迹近场图像并回传至消防炮中心控制器。中心控制器解析近场图像中射流轨迹入射角度值，基于射流轨迹入射角度值与射程相关性模型，计算当前近场图像中入射角度值与喷 射目标所需的入射角度偏差值。
射流轨迹近场图像中射流轨迹的入射角度值的解析方法具体为：
采用高斯背景减除方法剔除射流轨迹近场图像除射流轨迹以外的背景噪声；
遍历射流轨迹近场图像并计算获取图像中每一列为射流轨迹的像素点坐标均值；利用像素点坐标均值，采用数值拟合方法获取射流轨迹方程；
将射流轨迹在图像中起始像素点的位置代入射流轨迹方程计算获取射流轨迹的入射角度值。
射流轨迹近场图像的几何特征与射流轨迹模型的偏差计算方法为：
利用射流轨迹的入射角度值和射程相关性模型，计算得到射流轨迹近场图像几何特征与射流轨迹模型的偏差，具体包括：
针对基于机器视觉的智能化消防炮系统，通过标定实验获取基于机器视觉的智能化消防炮系统有效工作范围内的射流轨迹入射角度与射程的映射关系作为相关性模型；
在基于机器视觉的智能化消防炮系统的工作过程中，通过插值计算方法，获取消防炮实际所需射程在模型中映射的射流轨迹入射角度值；
对比当前图像中射流轨迹入射角度值与插值计算结果，即为射流轨迹近场图像几何特征与射流轨迹模型的偏差。
中心控制器将计算偏差转换成控制信号传输至第二步进电机6，第二步进电6依据控制信号调整消防炮头4俯仰角度，直至近场图像中射流轨迹入射角度值达到喷射目标所需的入射角度。上述过程根据火场变化情况，可调整自然光相机2图像采集帧率以改变消防炮头4俯仰调整频率，用于适应不同角度范围的消防炮水平调整。
经过处理分析后，消防炮中心控制器分别发出控制指令给第一步进电机5和第二步进电机6，分别实现消防炮8的水平位置调整以及消防炮头4俯仰角度调整。消防炮水平调整以及消防炮头俯仰调整在灭火过程中同步进行。
本发明的有益效果为：
本发明提供的一种基于机器视觉的智能化消防炮系统及其工作方法，可以实现消防炮的智能化灭火作业。在灭火过程中，通过对消防炮的水平角度调整实现消防炮对于火场位置的自动水平瞄准，并依据火场变化情况，不断进行偏差校正；通 过对消防炮头的俯仰角度调整实现消防炮射流轨迹落点对火场的覆盖，并依据火场变化情况，不断进行偏差校正，从而实现了火场瞄准、精准扑灭以及实时跟踪灭火。本发明基于视觉伺服技术，实现在灭火过程中的智能化作业，提升灭火作业效率，加快灭火救援速度，进一步减轻了人员伤亡与财产损失。
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到的变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应该以权利要求的保护范围为准。

Claims (6)

  1. 一种基于机器视觉的智能化消防炮系统，包括红外相机(1)、自然光相机(2)、机构支架(3)、消防炮头(4)、第一步进电机(5)、第二步进电机(6)、进水管(7)、消防炮(8)及中心控制器；消防炮(8)底部连接进水管(7)，其特征在于，消防炮(8)的下方安装第一步进电机(5)，消防炮(8)的顶部安装弯管，所述弯管上安装消防炮头(4)，消防炮头(4)上安装第二步进电机(6)，所述弯管上还安装机构支架(3)，机构支架(3)上安装红外相机(1)和自然光相机(2)，红外相机(1)的光轴和消防炮(8)瞄准方向位于同一竖直平面，并且平行；自然光相机(2)位于消防炮头(4)的后上方侧边位置；
    红外相机(1)、自然光相机(2)连接所述中心控制器，所述中心控制器连接并控制第一步进电机(5)、第二步进电机(6)。
  2. 一种基于机器视觉的智能化消防炮系统的工作方法，适用于如权利要求1所述的一种基于机器视觉的智能化消防炮系统，其特征在于，包括：
    红外相机采集火场红外图像，自然光相机采集消防炮头的射流轨迹近场图像，并发送至中心控制器；
    中心控制器解析火场红外图像中的火场中心点，并计算火场中心点与火场红外图像几何中心点在水平方向的偏差，中心控制器控制第一步进电机调整消防炮的水平位置，直至火场中心点与火场红外图像几何中心点位于同一竖直方向；
    中心控制器解析射流轨迹近场图像的几何特征，并计算射流轨迹近场 图像的几何特征与射流轨迹模型的偏差，中心控制器控制第二步进电机调整消防炮的俯仰角度，直至射流轨迹近场图像的几何特征完全匹配射流轨迹模型。
  3. 根据权利要求2所述的一种基于机器视觉的智能化消防炮系统的工作方法，其特征在于，所述火场中心点的解析方法为：所述中心控制器遍历所述火场红外图像找寻最高温度点的像素位置，将该位置标记为火场中心点的像素位置。
  4. 根据权利要求2所述的一种基于机器视觉的智能化消防炮系统的工作方法，其特征在于，所述射流轨迹近场图像的几何特征的解析方法为：
    所述中心控制器采用图像处理方法解析所述射流轨迹近场图像中射流轨迹的入射角度值，具体为：
    采用高斯背景减除方法剔除所述射流轨迹近场图像除射流轨迹以外的背景噪声；
    遍历所述射流轨迹近场图像并计算获取图像中每一列为射流轨迹的像素点坐标均值；
    利用所述像素点坐标均值，采用数值拟合方法获取射流轨迹方程；将射流轨迹在图像中起始像素点的位置代入射流轨迹方程计算获取射流轨迹的入射角度值。
  5. 根据权利要求4所述的一种基于机器视觉的智能化消防炮系统的工作方法，其特征在于，所述射流轨迹近场图像的几何特征与射流轨迹模型的偏差计算方法为：
    利用所述射流轨迹的入射角度值和射程相关性模型，计算得到所述射流轨迹近场图像几何特征与射流轨迹模型的偏差，具体包括：
    针对所述基于机器视觉的智能化消防炮系统，通过标定实验获取所述基于机器视觉的智能化消防炮系统有效工作范围内的射流轨迹入射角度与射程的映射关系作为所述相关性模型；
    在所述基于机器视觉的智能化消防炮系统的工作过程中，通过插值计算方法，获取消防炮实际所需射程在模型中映射的射流轨迹入射角度值；
    对比当前图像中射流轨迹入射角度值与插值计算结果，即为所述射流轨迹近场图像几何特征与射流轨迹模型的偏差。
  6. 根据权利要求2所述的一种基于机器视觉的智能化消防炮系统的工作方法，其特征在于：
    所述红外相机根据火场变化情况，调整图像采集帧率以改变消防炮水平调整频率，用于适应不同角度范围的消防炮水平调整；
    所述近场相机根据火场变化情况，调整图像采集帧率以改变消防炮俯仰调整频率，用于适应不同角度范围的消防炮俯仰调整。
PCT/CN2020/106082 2020-07-27 2020-07-31 一种基于机器视觉的智能化消防炮系统及工作方法 WO2022021275A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
AU2020461031A AU2020461031A1 (en) 2020-07-27 2020-07-31 Intelligent fire monitor system based on machine vision, and working method therefor

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010729896.2 2020-07-27
CN202010729896.2A CN111840871B (zh) 2020-07-27 2020-07-27 一种基于机器视觉的智能化消防炮系统及工作方法

Publications (1)

Publication Number Publication Date
WO2022021275A1 true WO2022021275A1 (zh) 2022-02-03

Family

ID=72947143

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/106082 WO2022021275A1 (zh) 2020-07-27 2020-07-31 一种基于机器视觉的智能化消防炮系统及工作方法

Country Status (3)

Country Link
CN (1) CN111840871B (zh)
AU (1) AU2020461031A1 (zh)
WO (1) WO2022021275A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115025433A (zh) * 2022-04-26 2022-09-09 西安理工大学 高风速下的消防泡沫炮射流跟踪控制系统及其控制方法

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112604212B (zh) * 2020-12-22 2022-03-04 重庆迪马工业有限责任公司 一种可自编程、预设灭火降温系统和方法
CN113274672A (zh) * 2021-04-27 2021-08-20 中国矿业大学 一种基于机器视觉的消防炮混合控制系统及控制方法
CN114476303B (zh) * 2022-01-26 2024-08-16 南京睿实智能安全科技有限公司 一种免拆装的消防炮高压测试托盘及其自动密封对接方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012223279A (ja) * 2011-04-18 2012-11-15 Nohmi Bosai Ltd 消火設備
CN104740818A (zh) * 2015-04-17 2015-07-01 公安部天津消防研究所 一种压缩空气泡沫智能炮自动灭火方法及系统
CN105056455A (zh) * 2015-07-29 2015-11-18 成都栖林测控科技有限责任公司 一种基于图像辨识的消防炮定位方法及装置
CN106853280A (zh) * 2017-02-15 2017-06-16 南京睿实消防安全设备有限公司 一种图像型自动跟踪定位射流灭火装置
CN206526437U (zh) * 2017-02-15 2017-09-29 南京睿实消防安全设备有限公司 一种图像型自动跟踪定位射流灭火装置
CN109331376A (zh) * 2018-10-12 2019-02-15 中国矿业大学 城市主战消防车自动灭火系统及实现方法
CN109939396A (zh) * 2019-03-26 2019-06-28 应急管理部天津消防研究所 一种具有落点修正功能的自行走式大流量消防炮

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201239452Y (zh) * 2008-08-15 2009-05-20 合肥科大立安安全技术有限责任公司 一种实现消防炮雾化角控制功能的机构
CN105080023B (zh) * 2015-09-02 2018-10-16 南京航空航天大学 一种自动跟踪定位射流灭火方法
CN205586404U (zh) * 2016-05-14 2016-09-21 华东交通大学 一种车载固定式消防水炮
CN107224692A (zh) * 2017-06-26 2017-10-03 福州大学 轮式消防机器人自主瞄准灭火方法
CN208081698U (zh) * 2018-03-06 2018-11-13 中国长江三峡集团有限公司 一种图像型自动消防炮灭火装置
CN108619648B (zh) * 2018-04-27 2020-03-17 中国矿业大学 基于视觉的消防炮射流轨迹识别模拟实验装置及方法
CN109303995A (zh) * 2018-09-12 2019-02-05 东南大学 基于火源定位识别的消防机器人消防炮控制方法
CN209270695U (zh) * 2018-11-19 2019-08-20 大连汇高智能设备制造有限公司 一种精确跟踪定位射流灭火水炮

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012223279A (ja) * 2011-04-18 2012-11-15 Nohmi Bosai Ltd 消火設備
CN104740818A (zh) * 2015-04-17 2015-07-01 公安部天津消防研究所 一种压缩空气泡沫智能炮自动灭火方法及系统
CN105056455A (zh) * 2015-07-29 2015-11-18 成都栖林测控科技有限责任公司 一种基于图像辨识的消防炮定位方法及装置
CN106853280A (zh) * 2017-02-15 2017-06-16 南京睿实消防安全设备有限公司 一种图像型自动跟踪定位射流灭火装置
CN206526437U (zh) * 2017-02-15 2017-09-29 南京睿实消防安全设备有限公司 一种图像型自动跟踪定位射流灭火装置
CN109331376A (zh) * 2018-10-12 2019-02-15 中国矿业大学 城市主战消防车自动灭火系统及实现方法
CN109939396A (zh) * 2019-03-26 2019-06-28 应急管理部天津消防研究所 一种具有落点修正功能的自行走式大流量消防炮

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Master Thesis", 1 January 2012, EAST CHINA JIAOTONG UNIVERSITY, CN, article LI ZHONG: "Key Technology Research into Auto-Targeting Fire Extinguishing System of Interior Large Space", pages: 1 - 101, XP055893743 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115025433A (zh) * 2022-04-26 2022-09-09 西安理工大学 高风速下的消防泡沫炮射流跟踪控制系统及其控制方法

Also Published As

Publication number Publication date
AU2020461031A1 (en) 2023-02-02
CN111840871B (zh) 2022-08-02
CN111840871A (zh) 2020-10-30

Similar Documents

Publication Publication Date Title
WO2022021275A1 (zh) 一种基于机器视觉的智能化消防炮系统及工作方法
CN109303995A (zh) 基于火源定位识别的消防机器人消防炮控制方法
CN109719438B (zh) 一种工业焊接机器人焊缝自动跟踪方法
CN104942404B (zh) 双波长双目视觉焊缝跟踪方法及跟踪系统
CN111494853B (zh) 一种多模式视觉伺服控制消防系统及其工作方法
CN103777204B (zh) 基于光电智能感知平台目标跟踪识别的测距装置及方法
CN105056455B (zh) 一种基于图像辨识的消防炮定位方法及装置
CN105080023B (zh) 一种自动跟踪定位射流灭火方法
CN106984926A (zh) 一种焊缝跟踪系统及焊缝跟踪方法
CN106042005B (zh) 仿生眼定位追踪系统的工作方法
CN105652872B (zh) 变电站激光导航巡检机器人智能云台自动跟踪定位方法
CN102179598A (zh) 平板窄焊缝初始焊位对中和焊缝跟踪方法及装置
CN107145162B (zh) 摄像头角度调整方法、系统和虚拟现实系统
CN110864440B (zh) 一种送风方法及送风装置、空调
CN104615150B (zh) 一种基于机器视觉的自适应精准喷雾设备及方法
CN108525176A (zh) 一种机场消防车的消防炮自动跟踪精准灭火系统
WO2020073595A1 (zh) 城市主战消防车自动灭火系统及实现方法
CN115708940A (zh) 一种自主溯源消防机器人
CN113274672A (zh) 一种基于机器视觉的消防炮混合控制系统及控制方法
CN113525631A (zh) 一种基于光视觉引导的水下终端对接系统及方法
CN106238871A (zh) 一种智能龙门焊接装置
CN107168523A (zh) 一种摄像机位置调整方法和系统
CN111541846A (zh) 一种冰壶影像自动跟拍系统
WO2022217671A1 (zh) 一种具有变视角双目结构的消防炮系统及方法
CN115337581B (zh) 一种基于多目视觉消防系统的消防灭火方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20946804

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020461031

Country of ref document: AU

Date of ref document: 20200731

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20946804

Country of ref document: EP

Kind code of ref document: A1