Firefighting water jet path and water landing point detection method and firefighting system

Abstract

A method and system for detecting the path and landing point of fire-fighting water jets, belonging to the field of fire automation technology, solves the problem of accurately determining the trajectory and landing point of fire-fighting water jets. This invention uses a visible light camera to capture images, analyzes the visible light images containing the fire-fighting water jet path and landing point, calculates the path trajectory equation and the coordinates of the landing point, and dynamically adjusts the pitch and horizontal rotation angles of the fire-fighting water supply device based on the coordinates of the landing point and the fire point, achieving efficient and precise fire suppression. This method effectively overcomes adverse factors in fire handling such as long distances from the fire source, variable wind conditions, and difficulty in locating the landing point, improving the fire-fighting efficiency and accuracy of fire-fighting robots, fire monitors, and other fire-fighting devices.

Description

Technical Field

[0001] This invention belongs to the field of fire protection automation technology, and relates to a method and system for detecting the path and landing point of fire water jets. Background Technology

[0002] In recent years, with the rapid development of industries such as new energy, new buildings, and petroleum refining, various new materials, processes, and products have emerged and been widely put into use, making fires increasingly complex and fire prevention and control more difficult. New firefighting devices such as fire-fighting robots and remote-controlled fire monitors have gradually emerged, which can, to some extent, replace rescue personnel entering the fire scene, locating the fire source, and spraying extinguishing agents, greatly reducing the risk of injury to personnel during fire fighting. However, currently, when fire-fighting robots and fire monitors locate the fire source and spray extinguishing agents, they still largely rely on manual handheld remote control terminals to issue commands. Fire scenes are often filled with dense smoke, making operation signals susceptible to interference. The scene is noisy and chaotic, and environmental wind can also affect the trajectory and landing point of the extinguishing agent. Therefore, the accuracy of manually judging the trajectory and landing point of the fire-fighting water spray is poor, easily leading to low fire fighting efficiency and increased accident losses. Summary of the Invention

[0003] The technical problem to be solved by this invention is how to accurately determine the trajectory and landing point of fire-fighting water jets.

[0004] The present invention solves the above-mentioned technical problems through the following technical solutions:

[0005] A method for detecting the path and landing point of fire-fighting water jets includes the following steps:

[0006] Step 1: Capture a visible light image containing information about the fire water jet path and the point of impact;

[0007] Step 2: Binarize the visible light image to determine the overall range of the highlighted connected region path, which is denoted as the first region;

[0008] Step 3: Start scanning from the top row of pixels in the first region, select the pixel with the largest gray value as the starting point of the jet, and if there are pixels with the same gray value, take the rightmost point as the starting point of the jet, and record the starting point as N1.

[0009] Step 4: In the second row of pixels directly below the starting point N1 of the jet, select multiple pixels on the left and right, and record the point with the largest gray value as S1. Calculate the sum of the absolute values ​​of the differences between each pixel and the starting point N1 in the RGB three color channels, and record the point with the smallest three color difference as S2. Take the point in the middle of the S1 and S2 pixels as the next fire water jet trajectory point, and record it as N2.

[0010] Step 5: Centered on N2, sample and determine the next jet trajectory point N3 in the third row of pixels directly below it, using the method in Step 4.

[0011] Step 6: Mark all fire water jet trajectory points on the image as in Steps 4 and 5;

[0012] Step 7: Using k rows of pixels as the step size, calculate the color feature difference of each row of trajectory points, and regard the fire water jet trajectory points that exceed the difference threshold as the landing point area of ​​the target.

[0013] Step 8: Use the straight line equation to fit all trajectory points before the water droplet area in the view plane, solve the unknown coefficients of the equation using the least squares method to obtain the fire jet trajectory equation, and use the fire jet trajectory equation to calculate the coordinates of the water droplet.

[0014] Further, the method for binarizing grayscale processing of visible light images described in step 2 is as follows: the optimal grayscale threshold Td of the visible light image is calculated using the maximum entropy criterion, and the grayscale values ​​of pixels with grayscale values ​​higher than Td are set to 1, while the grayscale values ​​of pixels with grayscale values ​​lower than Td are set to 0.

[0015] Furthermore, the calculation method for the optimal grayscale threshold Td mentioned in step 2 is as follows: Let the optimal grayscale threshold Td = s, where s takes a value between 0 and 255. The visible light image is divided into background and foreground at grayscale level s, denoted as region A and region B, respectively. The formula for calculating the sum of the information entropy of regions A and B is as follows:

[0016]

[0017]

[0018]

[0019]

[0020] I image =I(A) + I(B)

[0021] Where Ni represents the number of pixels with gray level i in the image, N represents the total number of pixels in the image, pi is the probability of a pixel with gray level i appearing, Ps represents the sum of probabilities of gray levels from 0 to s, I(A) and I(B) represent the information entropy of background region A and foreground region B, respectively. image This represents the information entropy of the entire image; solving the above equation, when I image When the maximum value is taken, the calculated gray level s is the optimal gray threshold Td mentioned in step 2.

[0022] Furthermore, the formula for calculating the sum of the absolute values ​​of the differences between each pixel and the starting jet point N1 in step 4 is as follows:

[0023] C = |R Pi -R N |+|G Pi -G N |+|B Pi -B N |

[0024] Among them, R Pi R represents the grayscale level of pixel Pi in the R color channel. N The gray level in the R channel of the previous jet trajectory point; G Pi G represents the grayscale level of the G color channel of pixel Pi. N The gray level of the G channel at the previous jet trajectory point; B Pi B represents the grayscale level of pixel Pi in the B color channel. N This represents the grayscale level in channel B of the previous jet trajectory point.

[0025] Furthermore, the formula for calculating the color feature difference of each trajectory point in step 7 is as follows:

[0026] Tc=C(N ( i+k ) -N ( i ) )+|D(N ( i+k ) )-D(N ( i ) )|

[0027] Wherein, C(N(i+k) ) -N(i ) ) is the RGB three-channel color difference between the trajectory point in row i+k and the trajectory point in row i, D(N(i ) ) is the gray value of the i-th trajectory point.

[0028] Furthermore, the color feature difference mentioned in step 7 is defined as the sum of the absolute values ​​of the gray level difference and the differences of the RGB three color channels.

[0029] Furthermore, in step 8, the trajectory points before the water droplet area are fitted using the linear equation within the view plane, and the unknown coefficients of the equation are solved using the least squares method to obtain the fire jet trajectory equation. The specific formula for calculating the coordinates of the water droplet using the fire jet trajectory equation is as follows:

[0030] Let the equation of the line be T = ax + b, and the parameters in the equation be obtained by the following formula:

[0031]

[0032]

[0033] The water droplet is located in the termination area of ​​the fire jet trajectory, and the coordinates of the water droplet satisfy the condition that the difference between the actual Y-coordinate value and the fitted result is minimized, i.e.:

[0034] mmin|Yi-ax i -b|

[0035] In the formula, xi and Yi are the coordinate values ​​of the jet trajectory points in the fire monitor's line of sight coordinate system, n is the number of trajectory points, and x and Y are the function variables of the fitted curve.

[0036] A fire protection system includes: a visible light camera, an adjustable fire monitor, and an MCU computing module;

[0037] The visible light camera is used to capture visible light images containing information about the fire water jet path and the point of impact.

[0038] The MCU computing module is set at the control computing terminal of the fire monitor and stores a computer program for the fire water jet path and landing point detection method according to any one of claims 1-6. It is used to analyze and calculate the visible light image captured by the visible light camera to obtain the fire water jet path trajectory equation and the coordinate information of the landing point.

[0039] The adjustable fire monitor is used to adjust the horizontal and vertical angles according to the fire water jet trajectory equation and the coordinates of the landing point.

[0040] Furthermore, the photos taken by the visible light camera are 8-bit color depth images.

[0041] The advantages of this invention are:

[0042] This invention uses a visible light camera to capture images, analyzes the visible light images containing the path trajectory and landing point of the fire water, calculates the path trajectory equation and the coordinates of the landing point, and dynamically adjusts the pitch and horizontal rotation angles of the fire water supply device based on the coordinates of the landing point and the fire point to achieve efficient and precise fire extinguishing. It can effectively overcome the adverse factors in fire handling such as long distance from the fire source, variable environmental wind, and difficulty in locating the landing point, and improve the fire handling efficiency and accuracy of fire fighting devices such as fire robots and fire monitors.

[0043] The fire water jet path and landing point detection algorithm provided by this invention has a fast execution speed and high calculation accuracy. It can provide fire water trajectory equations and landing point coordinates for fire-fighting devices such as fire robots and fire monitors, and assist in adjusting and accurately locating the fire source.

[0044] The method provided by this invention overcomes the shortcomings of traditional manual adjustment, such as slow speed, poor accuracy, and susceptibility to adverse factors, and can assist manual operation at the fire scene to achieve efficient and accurate operation.

[0045] The method provided by this invention has strong applicability, with no restrictions on the height of the fire monitor, the spray angle, etc. It is highly adaptable to different fire scenarios and can overcome the influence of environmental wind on the fire water path and landing point to a certain extent. Attached Figure Description

[0046] Figure 1 A flowchart of the fire water jet path and water droplet detection method of the present invention;

[0047] Figure 2 This is a visible light image of the fire-fighting water jet path according to the present invention;

[0048] Figure 3 This is a grayscale diagram of the fire water jet path of the present invention;

[0049] Figure 4 This is a schematic diagram illustrating the method for calculating the starting point and next trajectory point of the jet trajectory path in the present invention.

[0050] Figure 5 These are the trajectory points of the fire-fighting water jet in the method of this invention;

[0051] Figure 6 This is a schematic diagram of the trajectory points and landing points of the fire-fighting water jet in the method of the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0054] Example 1

[0055] like Figure 1 The diagram shows a flowchart of a method for detecting the path and landing point of fire-fighting water jets according to an embodiment of the present invention. The method of the present invention calculates the trajectory equation of the fire-fighting water jet path and the coordinate information of the landing point through image processing, which can be provided to fire-fighting robots or fire monitors to compensate and correct the angle and orientation of the water monitors, thereby achieving the goal of efficient and accurate fire extinguishing.

[0056] The specific implementation steps of the fire water jet path and water droplet detection method in this embodiment are as follows:

[0057] Step 1: Use a visible light camera to capture visible light images containing information about the path and point of impact of the fire-fighting water jets;

[0058] Step 2: Binarize the image to determine the overall range of the path, denoted as Region 1;

[0059] Step 3: Start scanning from the top row of pixels in region 1, select the pixel with the largest gray value as the starting point of the jet, and if there are pixels with the same gray value, use the rightmost point as the starting point of the jet, and record the starting point as N1.

[0060] Step 4: In the second row of pixels directly below the starting point N1 of the jet, select 10 pixels on the left and 10 pixels on the right, for a total of 20 pixels. Calculate the grayscale value of these pixels and denote the pixel with the largest grayscale value as S1. Calculate the sum of the absolute values ​​of the differences between each pixel and the RGB three color channels of the starting jet point N1 and denote the pixel with the smallest three color difference as S2. Take the point in the middle between pixels S1 and S2 as the next fire water jet trajectory point (if the middle pixel is an even number, take the pixel that is biased towards S2), and denote it as N2.

[0061] Step 5: Centered on N2, sample and determine the next jet trajectory point N3 in the third row of pixels directly below it, using the method described in Step 4.

[0062] Step 6: Mark all fire water jet trajectory points in the image as described in steps 4 and 5 above.

[0063] Step 7: Using k rows of pixels as the step size (k can be 3-8), calculate the color feature difference of each row of trajectory points, and regard the fire water jet trajectory points that exceed the difference threshold as the landing point area of ​​the target.

[0064] Step 8: Use the straight line equation to fit all trajectory points before the water droplet area in the view plane, solve the unknown coefficients of the equation using the least squares method to obtain the fire jet trajectory equation, and use the fire jet trajectory equation to calculate the coordinates of the water droplet.

[0065] The method proposed in this invention is applicable to visible light images with an 8-bit color depth. In an 8-bit color depth image, each pixel has 2 gray levels. 8 That is, 256 levels. Specifically, the grayscale value corresponding to the lowest and darkest pixel is recorded as 0, and the grayscale value of the brightest pixel is recorded as 255. Therefore, the larger the grayscale value of a pixel, the higher its brightness in terms of chromaticity. In the RGB space of an image, each pixel is composed of three color channels: R, G, and B, and the grayscale value of each color channel also ranges from 0 to 255.

[0066] The optimal grayscale threshold T d The calculation method is as follows: Let the optimal grayscale threshold T be... d =s, where s takes a value between 0 and 255. At gray level s, the visible light image is divided into background and foreground, denoted as region A and region B, respectively. The formula for calculating the sum of the information entropy of regions A and B is as follows:

[0067]

[0068]

[0069]

[0070]

[0071] I image =I(A) + I(B)

[0072] Where, N i p represents the number of pixels in the image with gray level i, N represents the total number of pixels in the entire image, and p i Let P be the probability of a pixel with gray level i appearing. S I(A) and I(B) represent the sum of probabilities of gray levels from 0 to s, respectively, and represent the information entropy of background region A and foreground region B. image This represents the information entropy of the entire image; solving the above equation, when I image When the maximum value is taken, the calculated gray level s is the optimal gray level threshold T mentioned in step 2. d .

[0073] like Figure 2 The image shown is of fire monitors spraying fire water, revealing the fire water path and impact points. Extensive experimental results show that the brightness of the fire water trajectory is significantly higher than the environmental background, even in low-light conditions, exhibiting a noticeably brighter chromaticity. This characteristic can be used to extract image information containing the complete fire water path from photographs. To enhance the distinguishability of the fire water path regions in the image, the maximum entropy criterion is used to calculate... Figure 1 The optimal grayscale threshold T d Gray values ​​higher than T d The pixel grayscale value is set to 1, and the grayscale value is lower than T. d The pixel grayscale value is set to 0.

[0074] Figure 3 Analysis of fire water path images after binarization and grayscale processing Figure 3 The highlighted connected region occupies the largest proportion in the image, and it can be determined that this connected region is the area of ​​the sprayed fire water flow, which is denoted as region 1.

[0075] To obtain information on the fire water jet landing points, it is necessary to calculate the fire water path trajectory equation. First, determine the starting point of the jet trajectory in region 1, as this starting point has the highest brightness and the greatest difference from the environmental background. The determination method is as follows: Starting from the first row of pixels at the upper left edge of region 1, scan the grayscale values ​​of the pixels in that row. Take the pixel with the highest grayscale value as the starting point of the jet. If multiple pixels have the same grayscale value, take the pixel closest to the right end, and record this starting point as N1.

[0076] Analysis of numerous visible light images of fire water flow paths reveals that the water flow trajectory not only exhibits significant brightness characteristics, but also, in terms of color space, its grayscale value is noticeably higher than that of the background area. Furthermore, the color characteristics of each trajectory point within the fire water flow area are very similar, with minimal color difference between pixels in adjacent rows within the area, and the absolute value of the three-channel color difference in the corresponding RGB color space being the smallest. Therefore, in addition to grayscale characteristics, the point with the smallest color difference between adjacent rows within region 1 can also be considered a trajectory point.

[0077] After determining the starting point N1 of the fire jet trajectory, the next trajectory point is calculated and determined in the manner described above. The schematic diagram is as follows. Figure 4 As shown. Starting from the second row of pixels in region 1, 10 pixels are symmetrically selected to the left and right of the starting point N1, labeled P1, P2...P20. The grayscale values ​​of these pixels are calculated, and the pixel with the largest grayscale value is designated as S1. The sum of the absolute values ​​of the grayscale differences between each pixel Pi and the starting jet point N1 in the RGB three color channels is calculated using the following formula:

[0078] C = |R Pi -R N |+|G Pi -G N |+|B Pi -B N |

[0079] Among them, R Pi R represents the grayscale level of pixel Pi in the R color channel. N The gray level in the R channel of the previous jet trajectory point; G Pi G represents the grayscale level of the G color channel of pixel Pi. N The gray level of the G channel at the previous jet trajectory point; B Pi B represents the grayscale level of pixel Pi in the B color channel. N This represents the grayscale level in channel B of the previous jet trajectory point.

[0080] Let S2 be the point with the smallest three-color difference C. Take the point between the pixels of S1 and S2 as the next fire water jet trajectory point (if the middle pixel is even, take the pixel biased towards S2), and denote it as N2.

[0081] Centered on N2, all jet trajectory points within region 1 are calculated row by row using the method described above. This yields the fire water path curve composed of all trajectory points, as shown below. Figure 5 As shown. Because the water disperses rapidly after falling onto the target point. Therefore... Figure 5 Not all marked fire water path trajectory points are before landing. Some trajectory points can be considered as those that scatter after colliding with the target point upon landing, and these points are not within the scope of the fire water path curve study. When fire water falls on the target point, the color and grayscale values ​​of its image pixels will change significantly. Using this characteristic, the sum of the absolute values ​​of the grayscale difference and the differences in the RGB three color channels is defined as the color feature difference between two points. With k rows of pixels as the step size, the color feature difference Tc of each row of trajectory points is calculated using the following formula:

[0082] Tc=C(N ( i+k ) -N ( i ) )+|D(N ( i+k ) )-D(N ( i ) )|

[0083] Among them, C(N) ( i+k ) -N ( i ) ) is the RGB three-channel color difference between the trajectory point in row i+k and the trajectory point in row i, D(N( i) ) is the gray value of the i-th trajectory point.

[0084] In practice, the value of k can be between 3 and 8, depending on the image size and processing speed / accuracy requirements. For ease of understanding, the following explanation will be based on k=5:

[0085] Follow the above method to Figure 5 The trajectory points are scanned starting from the first row with a step size of 5 rows. The color feature difference between the sixth row trajectory point N6 and the initial jet trajectory point N1 is calculated sequentially. The trajectory point N in the 11th row is then calculated. 11 The color feature difference between the trajectory point N6 in the 6th row and the color feature difference Tc is calculated, and so on. A difference threshold Tn is set. When the color feature difference Tc is less than the threshold Tn, it indicates that the feature difference is small and the fire jet has not terminated. When Tc exceeds the threshold Tn, it indicates that the color feature difference is significant and the fire jet has terminated. The area near the trajectory point exceeding the threshold is the water impact point. The calculated fire jet trajectory is as follows: Figure 6 As shown.

[0086] As mentioned earlier, the trajectory points of the fire-fighting water jet have been marked in the visible light image through image processing and chromaticity feature calculation. To obtain the coordinates of the landing point, the trajectory of the fire-fighting water jet must be fitted mathematically. The jet trajectory is a straight line in the viewing plane. The trajectory equation is solved by fitting all trajectory points using the least squares method. Let the equation of the straight line be Y = ax + b, and the parameters in the equation are obtained by the following formula:

[0087]

[0088]

[0089] The water droplet is located in the termination area of ​​the fire jet trajectory, and the coordinates of the water droplet satisfy the condition that the difference between the actual Y-coordinate value and the fitted result is minimized, i.e.:

[0090] min|Y i -ax i -b|

[0091] In the formula x i Y i , where are the coordinates of the jet trajectory points in the fire monitor's line-of-sight coordinate system, n is the number of trajectory points, and x and Y are the function variables of the fitted curve.

[0092] In summary, the method disclosed in this invention can calculate the trajectory of the fire jet and the coordinates of the landing point from visible light images of fire-fighting water jets. Since visible light cameras can repeatedly capture images within short intervals, the above calculation process can also be executed quickly in a computer program. Therefore, on the background monitoring platform for fire response, rescue personnel can obtain real-time fire jet trajectory equations and landing point coordinate information. Furthermore, these coordinates can, to some extent, overcome the interference of external factors such as environmental wind on the landing point. Equipment such as fire monitors can dynamically adjust the elevation and horizontal rotation angles of the fire-fighting medium supply device according to the landing point coordinates to achieve efficient and precise fire extinguishing.

[0093] This invention, based on visible light image processing technology, boasts high speed and accuracy. It provides path trajectory equations and water droplet coordinates for automatic firefighting devices such as fire robots and fire monitors, thereby compensating for and adjusting the elevation and horizontal angles of the water cannon to accurately strike the root of the fire source. This overcomes the inefficiency of traditional manual remote control and the deviation in water droplet location caused by environmental wind. By capturing images with a visible light camera and analyzing the visible light images containing the fire water path trajectory and water droplet coordinates, the invention calculates the path trajectory equation and water droplet coordinates. Based on the coordinates of the water droplet and the fire point, it dynamically adjusts the elevation and horizontal rotation angles of the fire water supply device to achieve efficient and precise firefighting. It effectively overcomes adverse factors such as long distances from the fire source, variable environmental winds, and difficulty in locating the water droplet, improving the fire handling efficiency and accuracy of fire robots, fire monitors, and other firefighting devices. The fire water jet path and water droplet detection algorithm provided by this invention has high execution speed and calculation accuracy, providing fire water trajectory equations and water droplet coordinates for fire robots, fire monitors, and other firefighting devices, assisting in precise positioning of the fire source. It overcomes the shortcomings of traditional manual adjustments, such as slow speed, poor accuracy, and susceptibility to adverse factors, and can assist personnel in achieving efficient and precise operation at the fire scene. It has strong applicability, with no restrictions on the height of the fire monitor or the spray angle, and is highly adaptable to different fire scenarios. It can also overcome the influence of environmental wind on the path and landing point of fire water to a certain extent.

[0094] Example 2

[0095] A fire protection system includes: a visible light camera, an adjustable fire monitor, and an MCU computing module;

[0096] The visible light camera is used to capture visible light images containing information about the fire water jet path and the point of impact. The shooting interval is adjustable, and multiple shots can be taken continuously in a short period of time. The captured images are 8-bit color depth images.

[0097] The MCU computing module is set at the control computing end of the fire monitor and stores the computer program of the fire water jet path and water landing point detection method of Embodiment 1. It is used to analyze and calculate the visible light image captured by the visible light camera to obtain the fire water jet path trajectory equation and the coordinate information of the water landing point.

[0098] The fire monitor adjusts its horizontal and vertical angles based on the calculated fire water trajectory equation and the coordinates of the water droplet, thus achieving precise and efficient fire extinguishing.

[0099] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the path and landing point of fire-fighting water jets, characterized in that, Includes the following steps: Step 1: Capture a visible light image containing information about the fire water jet path and the point of impact; Step 2: Binarize the visible light image to determine the overall range of the highlighted connected region path, which is denoted as the first region; The method for binarizing the gray-scale visible light image is as follows: the optimal gray-scale threshold T of the visible light image is calculated using the maximum entropy criterion d The pixel gray-scale value higher than T d is set to 1, and the pixel gray-scale value lower than T d is set to 0. The optimal gray threshold T d The calculation method is as follows: assuming that the optimal gray threshold T d =s, s is valued between 0-255, the visible light image is divided into background and foreground at the gray level s, and is respectively marked as A area and B area, and the sum of information entropy of the A area and the B area is calculated according to the following formula: Where, N i p represents the number of pixels in the image with gray level i, N represents the total number of pixels in the entire image, and p i Let P be the probability of a pixel with gray level i appearing. S Let I(A) represent the sum of probabilities of gray levels from 0 to s, and let I(B) and I(B) represent the information entropy of background region A and foreground region B, respectively. Represents the information entropy of the entire image; when When the maximum value is taken, the calculated gray level s is the optimal gray level threshold T mentioned in step 2. d ; Step 3: Start scanning from the top row of pixels in the first region, select the pixel with the largest gray value as the starting point of the jet, and if there are pixels with the same gray value, take the rightmost point as the starting point of the jet, and record the starting point as N1. Step 4: In the second row of pixels directly below the starting point N1 of the jet, select multiple pixels on the left and right, and record the point with the largest gray value as S1. Calculate the sum of the absolute values ​​of the differences between each pixel and the starting point N1 in the RGB three color channels, and record the point with the smallest three color difference as S2. Take the point in the middle of the S1 and S2 pixels as the next fire water jet trajectory point, and record it as N2. Step 5: Centered on N2, sample and determine the next jet trajectory point N3 in the third row of pixels directly below it, using the method in Step 4. Step 6: Mark all fire water jet trajectory points on the image as in Steps 4 and 5; Step 7: with k Using row pixels as the step size, calculate the color feature difference of each row trajectory point, and regard the fire water jet trajectory point where the color feature difference exceeds the difference threshold as the water landing point area reaching the target. Step 8: Use the straight line equation to fit all trajectory points before the water droplet area in the view plane, solve the unknown coefficients of the equation using the least squares method to obtain the fire jet trajectory equation, and use the fire jet trajectory equation to calculate the coordinates of the water droplet.

2. The method for detecting the fire water jet path and landing point according to claim 1, characterized in that, The formula for calculating the sum of the absolute values ​​of the differences between each pixel and the starting jet point N1 in step 4 is as follows: in, For pixel P i The grayscale levels under the R color channel, The gray level in the R channel of the previous jet trajectory point; For pixel P i The grayscale level under the G color channel, The gray level of the G channel at the previous jet trajectory point; For pixel P i The grayscale levels under the B color channel, This represents the grayscale level in channel B of the previous jet trajectory point.

3. The method for detecting the fire water jet path and landing point according to claim 2, characterized in that, The formula for calculating the color feature difference of each trajectory point in step 7 is as follows: in, It is the difference in the RGB three-channel colors between the trajectory point in the (i+k)th row and the trajectory point in the ith row. It is the gray value of the i-th trajectory point.

4. The method for detecting the fire water jet path and landing point according to claim 3, characterized in that, The color feature difference mentioned in step 7 is defined as the sum of the absolute values ​​of the gray level difference and the differences of the RGB three color channels.

5. The method for detecting the fire water jet path and landing point according to claim 4, characterized in that, In step 8, the trajectory points before the water droplet area are fitted using a straight line equation within the view plane. The unknown coefficients of the equation are solved using the least squares method to obtain the fire jet trajectory equation. The specific formula for calculating the coordinates of the water droplet using the fire jet trajectory equation is as follows: Let the equation of the line be... The parameters in the formula are obtained by the following formula: The water droplet is located in the termination area of ​​the fire jet trajectory, and the coordinates of the water droplet satisfy the condition that the difference between the actual Y-coordinate value and the fitted result is minimized, i.e.: In the formula , where are the coordinates of the jet trajectory points in the fire monitor's line-of-sight coordinate system, n is the number of trajectory points, and x and Y are the function variables of the fitted curve.

6. A fire protection system, comprising: Visible light camera, adjustable fire monitor, MCU computing module; The visible light camera is used to capture visible light images containing information about the fire water jet path and the point of impact. The MCU calculation module is characterized in that it is set at the control calculation terminal of the fire monitor and stores a computer program for the fire water jet path and water landing point detection method according to any one of claims 1-5, which is used to analyze and calculate the visible light image captured by the visible light camera, thereby obtaining the fire water jet path trajectory equation and the coordinate information of the water landing point. The adjustable fire monitor is used to adjust the horizontal and vertical angles according to the fire water jet trajectory equation and the coordinates of the landing point.

7. The fire protection system according to claim 6, characterized in that, The photos taken by the visible light camera are 8-bit color depth images.

Images