阅读说明：本技术 一种基于火焰图像动态识别的消防无人艇水枪控制装置 (Fire control unmanned ship squirt controlling means based on flame image dynamic identification ) 是由 徐小强 张瑞琦 冒燕 于 2019-09-20 设计创作，主要内容包括：本发明涉及一种基于火焰图像动态识别的消防无人艇水枪控制装置,通过双目摄像头采集火焰视频数据以及无人艇与燃烧物之间距离,通过风速检测仪采集风速、风向信息,通过对视频数据以及风速、风向信息的处理产生消防水枪的倾斜角控制信号。通过本发明,能够实现增强单条母船出勤时的灭火能力,提高消防工作效率。(The invention relates to a fire-fighting unmanned ship water gun control device based on flame image dynamic identification. By the aid of the method and the system, fire extinguishing capacity of a single mother ship during attendance can be enhanced, and fire fighting efficiency is improved.)

1. A fire-fighting unmanned ship water gun control device based on flame image dynamic recognition is characterized by comprising a ship body (4), wherein a wind speed and direction sensor (1), a binocular camera (2), a water gun module (3), an unmanned ship control system, a pusher (8) and a temperature sensor (9) are mounted on the ship body (4); the unmanned ship control system comprises a main control center (6), a water gun control module (5) and a communication module (7); the wind speed and direction sensor (1) and the binocular camera (2) are arranged at the top of the unmanned ship and used for measuring wind speed and direction and acquiring flame images; the water gun module (3) is installed in front of the binocular camera (2), the height of the top end of the water gun module (3) away from the ship surface of the ship body (4) is half of that of the binocular camera (2), the water gun control module (5), the main control center (6) and the communication module (7) are installed inside a cabin of the ship body (4), and the temperature sensor (9) is installed at the ship head of the ship body (4);

the flame image information that distance information and binocular camera (2) left side camera gathered between unmanned ship and the burning thing that obtain through binocular camera (2), adjust the unmanned ship of surface of water fire control to best operating position, combine the wind direction wind speed information that wind speed and direction sensor (1) gathered and the flame image information that binocular camera (2) left side camera gathered to optimize the inclination of squirt support, reach quick fire extinguishing.

2. A fire fighting unmanned surface water gun control device based on dynamic identification of flame images as claimed in claim 1, wherein the angle of inclination of the high pressure water gun is changed by changing the angle of the high pressure water gun support.

3. The fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 1, wherein after the unmanned ship is launched, flame image information is collected through a left side camera of the binocular camera (2), if the flame image information is not detected, the unmanned ship is driven to rotate in situ, and the in-situ rotation is stopped until the flame image information is collected; the distance between the unmanned ship and the comburent obtained by combining the binocular camera (2) is used for adjusting the fire extinguishing distance of the unmanned ship.

4. The fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 3, wherein if the distance between the unmanned ship and the comburent obtained by the binocular camera (2) is greater than 10 m, the unmanned ship is driven to move forwards until the distance between the unmanned ship and the comburent is less than or equal to 10 m, whether the flame image height collected by the left camera in the binocular camera (2) is full screen is judged, if so, the unmanned ship stops moving forwards, and the high-pressure water gun support angle is adjusted by combining real-time wind speed and wind direction to extinguish fire; otherwise, driving the unmanned boat to move forwards until the screen is full or the distance between the unmanned boat and the burning objects is less than or equal to 5 meters, and adjusting the angle of the high-pressure water gun support by combining the real-time wind speed and the wind direction to extinguish the fire; if the distance between the unmanned boat and the comburent is less than 10 meters and more than 5 meters, the flame image height collected by the left camera in the binocular camera (2) is judged to be full, the high-pressure water gun support angle is adjusted by combining the real-time wind speed and the wind direction to start fire extinguishing, otherwise, the unmanned boat is driven to move forwards until the flame image height is full or the distance between the unmanned boat and the comburent is less than or equal to 5 meters, and the high-pressure water gun support angle is adjusted by combining the real-time wind speed and the wind direction to start fire extinguishing; if the distance between the unmanned ship and the comburent is less than or equal to 5 meters, whether the flame image height collected by the left side camera in the binocular camera (2) is full screen or not is combined, if the flame image height is not full screen, the angle of the high-pressure water gun support is adjusted in combination with real-time wind speed to extinguish the fire, otherwise, the unmanned ship is driven to drive backwards until the flame image height collected by the left side camera (2) in the binocular camera is not full screen or the distance between the unmanned ship and the comburent is detected to be more than or equal to 10 meters, and then the angle of the high-pressure water gun support is adjusted in combination with real-time wind speed and wind direction to extinguish the.

5. A fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 3, characterized in that: binocular camera (2) will detect the image transmission and handle for TMS320DM642 chip, wherein with in binocular camera (2) the image that the left side camera was gathered behind the flame image processing flow, combine the flame image information of gathering and through the binocular camera obtain with the combustion object between the distance, produce the angle of inclination control signal of fire-fighting lance, carry out the high-efficient fire extinguishing.

6. The fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 5, wherein the image acquisition, image preprocessing, flame image segmentation, flame image feature extraction and flame identification are included, the image acquisition is a digitization process of images, the image preprocessing is a first step of processing flame images, and is an operation of performing wavelet denoising on the obtained images to ensure that high-quality flame features are extracted, the image segmentation is to divide the preprocessed images into suspected flame regions and non-flame regions through a certain criterion, image feature extraction on the non-flame regions is not required in the subsequent processing process, the processing workload is saved, and finally flame identification is performed; the image segmentation adopts an algorithm for image segmentation by using a YCbCr color space, and a conversion formula from RGB to YCbCr space is shown as a formula (1):

the value of the Y channel of the flame region is larger than the average Y channel of the whole image_meanvalueThe Cb channel value of the large flame area is larger than the Cb channel mean Cb of the whole image_meanvalueSmall, the value of the Cr channel in the flame region is larger than the average Cr channel of the whole image_meanvalueLarge, i.e., as described in equation (2);

wherein, Y_meanvalue,Cr_meanvalue，Cb_meanvalueRespectively representing the brightness information of the image, the mean value of the red color difference and the mean value of the blue color difference, and is obtained by a formula (3);

7. a fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 3, wherein in the process of identifying flames, the extraction of the characteristics of the flames comprises the extraction of static characteristics and dynamic characteristics of the flames.

8. The fire fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 7, wherein static features of flames are extracted by using a single frame image of flames, the static features comprise regional structural features of flames and structural features of flames, the flames have specific shapes and structures from the structural aspect of flames, and the regions of flames are extracted by using rectangular similarity and circular similarity according to the structural features of flames.

9. A fire-fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 7 is characterized in that the random motion characteristics of flames are extracted, the random motion characteristics of flames are reflected in the random motion of flames, and the flames are in a flickering state through distinguishing the changes of images between frames.

10. A fire fighting unmanned ship water gun control device based on flame image dynamic identification as claimed in claim 1, characterized in that after the image of the flame is detected, the inclination angle of the fire fighting water gun is adjusted to achieve the best fire fighting effect by comprehensively calculating the height of the flame, the distance of the flame from the unmanned ship and the collected wind speed.

Technical Field

The invention relates to the technical field of ship fire fighting, in particular to a fire fighting unmanned ship water gun control device based on flame image dynamic identification.

Background

An unmanned surface vessel, also called a surface robot, is an unmanned surface naval vessel. At present, high-risk manual operation areas exist on a plurality of water surfaces, and the intelligent water surface unmanned boat can be used for replacing human beings to finish water surface operation with high operation difficulty and danger.

The existing water surface fire boat is mostly a manual fire extinguishing device. When a serious fire occurs, a plurality of fire boats need to be driven, and the safety of fire extinguishers can be threatened.

Disclosure of Invention

The technical problem to be solved by the present invention is to provide a fire fighting unmanned ship water gun control device based on flame image dynamic identification, aiming at the above defects in the prior art.

The technical scheme adopted by the invention for solving the technical problems is as follows: a fire-fighting unmanned ship water gun control device based on flame image dynamic identification is constructed, and comprises a ship body, wherein a wind speed and direction sensor, a binocular camera, a water gun module, an unmanned ship control system, a pusher and a temperature sensor are mounted on the ship body; the unmanned ship control system comprises a master control center, a water gun control module and a communication module; the wind speed and direction sensor and the binocular camera are arranged at the top of the unmanned ship, and are used for measuring wind speed and direction and acquiring flame images; the water gun module is arranged in front of the binocular camera, the height from the top end of the water gun module to the ship surface of the ship body is half of that of the binocular camera, the water gun control module, the master control center and the communication module are arranged in a cabin of the ship body, and the temperature sensor is arranged at the ship head of the ship body;

the unmanned ship of surface of water fire control adjusts the flame image information that the unmanned ship of the binocular camera obtained and gather through the distance information between unmanned ship and the burning thing that the binocular camera was gathered to the best operating position, combines the wind direction wind speed information optimization squirt support that wind speed and direction sensor gathered inclination, reaches and puts out a fire fast.

In the above-mentioned scheme, the high-pressure squirt angle of inclination changes through changing high-pressure squirt support angle.

In the scheme, after the unmanned ship is launched, flame image information is collected through a left camera of a binocular camera, if the flame image information is not detected, the unmanned ship is driven to rotate in situ until the flame image information is collected, and the in-situ rotation is stopped; the distance that unmanned ship put out a fire begins to adjust through the distance between unmanned ship that the binocular camera reachd and the burning thing.

In the scheme, if the distance between the unmanned boat and the comburent obtained through the binocular camera is larger than 10 meters, the unmanned boat is driven to move forwards until the distance between the unmanned boat and the comburent is smaller than or equal to 10 meters, whether the flame image height collected by the left camera in the binocular camera is full or not is judged, if so, the unmanned boat stops moving forwards, and the angle of the high-pressure water gun support is adjusted by combining real-time wind speed to extinguish a fire; otherwise, driving the unmanned boat to move forwards until the screen is full or the distance between the unmanned boat and the burning objects is less than or equal to 5 meters, and adjusting the angle of the high-pressure water gun support by combining the real-time wind speed to extinguish the fire; if the distance between the unmanned boat and the combustion object is less than 10 meters and more than 5 meters, judging that the flame image acquired by the left camera in the binocular cameras is full of screen height, adjusting the angle of the high-pressure water gun support by combining real-time wind speed to start fire extinguishing, otherwise driving the unmanned boat to move forwards until the flame image is full of screen height or the distance between the unmanned boat and the combustion object is less than or equal to 5 meters, and adjusting the angle of the high-pressure water gun support by combining real-time wind speed to start fire extinguishing; if the distance between unmanned ship and the comburent is less than or equal to 5 meters, whether the flame image height that the left side camera was gathered in the combination binocular camera is full screen, if the flame image height does not have full screen, then combine real-time wind speed adjustment high pressure squirt support angle to begin to put out a fire, otherwise then order about unmanned ship and go backward until the flame image height that the left side camera was gathered in the binocular camera does not have full screen or detect the distance between unmanned ship and the comburent more than or equal to 10 meters, then combine real-time wind speed adjustment high pressure squirt support angle to begin to put out a fire.

Among the above-mentioned scheme, the binocular camera will detect the image transmission and handle for the TMS320DM642 chip, wherein with in the binocular camera the image that the left side camera was gathered behind the flame image processing flow, combine the flame image information of gathering and through the binocular camera obtain with the combustion object between the distance, produce the angle of inclination control signal of fire-fighting lance, carry out the high-efficient fire extinguishing.

In the above scheme, the flame image processing flow includes: the method comprises the steps of image acquisition, image preprocessing, flame image segmentation, flame image feature extraction and flame identification, wherein the image acquisition is the digitization process of an image, the image preprocessing is the first step of processing a flame image and is the operation of performing wavelet denoising on the obtained image to ensure that high-quality flame features are extracted, the image segmentation is to divide the preprocessed image into a suspected flame region and a non-flame region through a certain criterion, the image feature extraction is not required to be performed on the non-flame region in the subsequent processing process, the processing workload is saved, and finally the flame identification is performed; the image segmentation adopts an algorithm for carrying out image segmentation by using a YCbCr color space, and a conversion formula from RGB to YCbCr space is shown as a formula (1);

the value of the Y channel of the flame region is larger than the average Y channel of the whole image_meanvalueThe Cb channel value of the large flame area is larger than the Cb channel mean Cb of the whole image_meanvalueSmall, the value of the Cr channel in the flame region is larger than the average Cr channel of the whole image_meanvalueLarge, i.e., as described in equation (2);

wherein, Y_meanvalue,Cr_meanvalue，Cb_meanvalueRespectively representing the brightness information of the image, the mean value of the red color difference and the mean value of the blue color difference, and is obtained by a formula (3);

in the scheme, in the process of identifying the flame, the characteristic extraction of the flame comprises the extraction of static characteristics and dynamic characteristics of the flame.

In the scheme, the static features of the flame are extracted by using a single frame image of the flame, the static features comprise the regional structural features of the flame and the structural features of the flame, the flame has a specific shape and structure from the structural aspect of the flame, and the region of the flame is extracted by using the rectangular similarity and the circular similarity according to the structural features of the flame.

In the scheme, the random motion characteristics of the flame are extracted and are reflected in the random motion of the flame, and the flame is judged to be in a flickering state through the change of the inter-frame images.

According to the scheme, when the unmanned ship detects that flames are extinguished, the unmanned ship enters the original place to stand by again and waits for the next instruction of the mother ship.

In the above scheme, after the image of flame is detected, the optimal fire extinguishing effect is achieved by comprehensively calculating the height of the flame, the distance from the flame to the unmanned ship and the collected wind speed and adjusting the inclination angle of the fire-fighting lance.

According to the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the flame video data and the distance between an unmanned ship and a burning object are collected through the binocular camera, the wind speed and wind direction information is collected through the wind speed detector, and the inclination angle control signal of the fire-fighting water gun is generated through processing the video data and the wind speed and wind direction information. By the aid of the method and the system, fire extinguishing capacity of a single mother ship during attendance can be enhanced, and fire fighting efficiency is improved.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

fig. 1 is a schematic structural diagram of a fire fighting unmanned ship water gun control device based on flame image dynamic identification provided by the invention;

FIG. 2 is a schematic flow chart of flame identification of a fire-fighting unmanned ship water gun control device based on flame image dynamic identification provided by the invention;

fig. 3 is a schematic diagram of a fire fighting high-pressure water gun control fire extinguishing process of the fire fighting unmanned ship water gun control device based on flame image dynamic recognition.

Detailed Description

For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

As shown in fig. 1, the invention provides a fire-fighting unmanned ship water gun control device based on flame image dynamic recognition, which comprises a ship body 4, wherein a wind speed and direction sensor 1, a binocular camera 2, a water gun module 3, an unmanned ship control system, a pusher 8 and a temperature sensor 9 are arranged on the ship body 4; the unmanned ship control system comprises a main control center 6, a water gun control module 5 and a communication module 7; the wind speed and direction sensor 1 and the binocular camera 2 are arranged at the top of the unmanned boat and used for measuring wind speed and direction and acquiring flame images; the water gun module 3 is installed in front of the binocular camera 2, the height from the top end of the water gun module 3 to the ship surface of the ship body 4 is half of that of the binocular camera 2, the water gun control module 5, the main control center 6 and the communication module 7 are installed inside the cabin of the ship body 4, and the temperature sensor 9 is installed at the ship head of the ship body 4;

the unmanned ship of surface of water fire control adjusts the flame image information that distance information and the 2 left cameras of binocular camera gathered between unmanned ship and the burning thing that obtain through binocular camera 2, to the best operating position, combines the inclination of wind direction wind speed information optimization squirt support that wind speed and direction sensor 1 gathered, reaches and puts out a fire fast.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the inclination angle of the high-pressure water gun is changed by changing the angle of the high-pressure water gun support.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, after an unmanned ship is thrown in, flame image information is collected through the left side camera of the binocular camera 2, if the flame image information is not detected, the unmanned ship is driven to rotate in situ until the flame image information is collected, and the in-situ rotation is stopped; the distance between the unmanned ship and the comburent obtained by combining the binocular camera 2 is used for adjusting the fire extinguishing distance of the unmanned ship.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, if the distance between an unmanned ship and a burning object obtained through the binocular camera 2 is larger than 10 meters, the unmanned ship is driven to move forwards until the distance between the unmanned ship and the burning object is smaller than or equal to 10 meters, whether the height of a flame image acquired by the left camera in the binocular camera 2 is full screen or not is judged, if the screen is full screen, the unmanned ship stops moving forwards, and the angle of a high-pressure water gun support is adjusted by combining real-time wind speed to start fire extinguishing; otherwise, driving the unmanned boat to move forwards until the screen is full or the distance between the unmanned boat and the burning objects is less than or equal to 5 meters, and adjusting the angle of the high-pressure water gun support by combining the real-time wind speed to extinguish the fire; if the distance between the unmanned boat and the burning objects is less than 10 meters and more than 5 meters, the flame image height collected by the left camera in the binocular camera (2) is judged to be full, the angle of the high-pressure water gun support is adjusted by combining the real-time wind speed to start fire extinguishing, otherwise, the unmanned boat is driven to move forwards until the flame image height is full or the distance between the unmanned boat and the burning objects is less than or equal to 5 meters, and the angle of the high-pressure water gun support is adjusted by combining the real-time wind speed to start fire extinguishing; if the distance between unmanned ship and the comburent is less than or equal to 5 meters, whether the flame image height that the left side camera gathered in the combination binocular camera 2 is full screen or not, if the flame image height is not full screen, then combine real-time wind speed adjustment high pressure squirt support angle to begin to put out a fire, otherwise then order about unmanned ship and go backward until the flame image height that left side camera 2 gathered in the binocular camera does not have full screen or detect the distance between unmanned ship and the comburent more than or equal to 10 meters, then combine real-time wind speed adjustment high pressure squirt support angle to begin to put out a fire.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the binocular camera 2 transmits detected images to the TMS320DM642 chip for processing, wherein after image preprocessing, flame image segmentation, flame image feature extraction and flame identification are carried out on images collected by the left camera in the binocular camera 2, an inclination angle control signal of a fire-fighting water gun is generated by combining collected flame image information and the distance between the collected flame image information and a combustion object obtained through the binocular camera, and efficient fire extinguishing is carried out. As shown in fig. 2.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the flame image processing flow comprises the following steps: the method comprises the steps of image acquisition, image preprocessing, flame image segmentation, flame image feature extraction and flame identification, wherein the image acquisition is the digitization process of an image, the image preprocessing is the first step of processing a flame image and is the operation of performing wavelet denoising on the obtained image to ensure that high-quality flame features are extracted, the image segmentation is to divide the preprocessed image into a suspected flame region and a non-flame region through a certain criterion, the image feature extraction is not required to be performed on the non-flame region in the subsequent processing process, the processing workload is saved, and finally the flame identification is performed; the image segmentation adopts an algorithm for carrying out image segmentation by using a YCbCr color space, and a conversion formula from RGB to YCbCr space is shown as a formula (1);

the value of the Y channel of the flame region is larger than the average Y channel of the whole image_meanvalueThe Cb channel value of the large flame area is larger than the Cb channel mean Cb of the whole image_meanvalueSmall, the value of the Cr channel in the flame region is larger than the average Cr channel of the whole image_meanvalueLarge, i.e., as described in equation 2;

wherein, Y_meanvalue,Cr_meanvalue，Cb_meanvalueRespectively representing the brightness information of the image, the mean value of the red color difference and the mean value of the blue color difference, and obtaining the average values through a formula 3;

in the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the extraction of the flame characteristics comprises the extraction of static characteristics and dynamic characteristics of the flame in the process of identifying the flame; wherein the static features of the flame are extracted using a single frame image of the flame.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the static characteristics of flame comprise: the method comprises the following steps of (1) extracting the flame region by utilizing the rectangular similarity and the circular similarity according to the structural characteristics of the flame, wherein the flame has a specific shape and structure from the structural point of view of the flame; extracting the random motion characteristics of the flame from the dynamic characteristics of the flame; the random motion characteristic of the flame is reflected in the random motion of the flame, and the flame is judged to be in a flickering state through the change of the images between frames.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, when the unmanned ship detects that flames are extinguished, the unmanned ship enters the original place to stand by again and waits for the next instruction of a mother ship.

In the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, after the image of flame is detected, the inclination angle of the fire-fighting water gun is adjusted to achieve the optimal fire-fighting effect by comprehensively calculating the height of the flame, the distance between the flame and the unmanned ship and the collected wind speed. As shown in fig. 3.

When the fire-fighting high-pressure water gun works, the falling point of water flow of the fire-fighting high-pressure water gun is influenced by water flow, wind speed and the inclination angle of the fire-fighting high-pressure water gun support. The height and the flow of the high-pressure water gun in the unmanned fire fighting boat are preset fixed values, and after the image of flame is detected, the optimal fire extinguishing effect is achieved by comprehensively calculating the height of the flame, the distance from the flame to the unmanned fire fighting boat and the collected wind speed to change the inclination angle of the fire fighting water gun. The inclination angle of the high-pressure water gun is changed by acquiring flame images through the flame images, so that a closed-loop system for extinguishing fire is formed. The inclination angle of the high-pressure water gun support changes the position of the node 1 by giving an electric signal through the control center, and then the angle of the high-pressure water gun support is changed. Wherein the water with the mass of 1 unit is taken as a standardCalculating, wherein k and b are constants, and the detected horizontal wind speed is V_xWind speed in vertical direction V_yAnd when the horizontal wind and the bow are detected to be in the same direction, the wind speed is recorded as a positive number, otherwise, the wind speed is recorded as a negative number, and when the vertical wind is detected to be downward, the wind speed is recorded as a positive number, otherwise, the wind speed is recorded as a negative number. The speed of water sprayed by the fire-fighting lance is V₀The distance between the spray orifice of the fire-fighting lance and the horizontal plane is y, the distance between the fire-fighting lance and the fire source is x when the fire-fighting lance is detected, g is the gravity acceleration, and if the time shared by the water from the water spray to the ground is t, the water sprayed by the fire-fighting lance can be just fallen on the fire source when h is the value according to the formulas (9) and (10).

According to the fire-fighting unmanned ship water gun control device based on flame image dynamic identification, the flame video data and the distance between an unmanned ship and a burning object are collected through the binocular camera, the wind speed and wind direction information is collected through the wind speed detector, and the inclination angle control signal of the fire-fighting water gun is generated through processing the video data and the wind speed and wind direction information. By the aid of the method and the system, fire extinguishing capacity of a single mother ship during attendance can be enhanced, and fire fighting efficiency is improved.

While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.