Multi-track searching method for jet flow of fire water monitor

Abstract

The embodiment of the invention provides a multi-track searching method for jet flow of a fire water monitor, and belongs to the technical field of fire fighting. The multi-track searching method comprises the following steps: acquiring a picture about the jet flow of the fire water monitor; identifying the picture to obtain a plurality of areas about a jet flow starting point of the jet flow in the current picture; acquiring information about a plurality of jet flow starting points of jet flow in the picture; according to the information of a plurality of jet flow starting points in the current picture, obtaining a track about the jet flow of the fire water monitor; and screening tracks about the jet flow of the fire water monitor in the current picture to obtain an optimal track about the jet flow of the fire water monitor as an optimal jet flow track. According to the multi-trajectory searching method, the optimal trajectory about the jet flow can be found in the multiple trajectories.

Description

Technical Field

[0001] The present invention relates to the field of fire fighting technology, and in particular to a multi-track search method for fire fighting water cannon jets. Background Art

[0002] In real life, when extinguishing a fire, the firefighting water bubble needs to accurately find the fire source and then extinguish the fire. In the process of intelligent fire extinguishing, it is necessary to accurately grasp the jet trajectory of the firefighting water bubble, so as to accurately locate the water drop point. When searching for the jet trajectory of the firefighting water cannon in reality, there are often multiple starting points of suspected targets in one picture. When searching for the jet trajectory through the starting point of the target, multiple trajectories are often searched, which affects the search for the final jet trajectory. Therefore, it is necessary to find the final jet trajectory among multiple trajectories when the target point is unclear. Summary of the invention

[0003] The purpose of the embodiment of the present invention is to provide a multi-trajectory search method for a fire monitor jet, which can find the optimal trajectory of the jet among multiple trajectories.

[0004] In order to achieve the above-mentioned object, an embodiment of the present invention provides a multi-track search method for a fire monitor jet, and the multi-track search method comprises: obtaining a picture of the fire monitor jet;

[0005] Identify the image to obtain multiple regions of the jet starting point of the jet in the current image;

[0006] acquiring information about a plurality of jet starting points of the jet in the image;

[0007] Acquire the trajectory of the fire monitor jet according to the information of multiple jet starting points in the current image;

[0008] The trajectory of the fire-fighting water monitor jet in the current image is screened to obtain the optimal trajectory of the fire-fighting water monitor jet as the optimal jet trajectory.

[0009] Optionally, obtaining information about multiple jet starting points of the jet in the picture includes:

[0010] Acquire a plurality of regions in the image about the starting point of the jet;

[0011] Acquire a first starting point and a first trajectory angle about the first starting point in each of the regions;

[0012] Acquire, in the region, a plurality of other starting points arranged in parallel in a direction perpendicular to the first trajectory angle according to the first starting point and the first trajectory angle, wherein the trajectory angles of the plurality of other starting points are the same as the first trajectory angle;

[0013] The position information and angle information of the first starting point and other starting points in each of the areas are collected to obtain the information of the jet starting point.

[0014] Optionally, obtaining information about multiple jet starting points of the jet in the picture includes: retaining the starting point and initial trajectory angle of the best jet trajectory after screening in the previous frame of the picture as one of the jet starting points in the current frame of the picture.

[0015] Optionally, obtaining the trajectory of the fire monitor jet according to the information of multiple jet starting points in the current image includes:

[0016] Obtain all jet starting points of the fire monitor jet in each area in the current image;

[0017] Acquire the position of the jet starting point and the trajectory angle about the jet starting point;

[0018] Acquire the search range of the next trajectory point according to the position of the jet starting point and the trajectory angle;

[0019] Search the search range of the next trajectory point to search for the point with the largest grayscale in the search range as the next trajectory point, and obtain the trajectory angle of the next trajectory point to search for subsequent trajectory points until the trajectory point of the landing point is found;

[0020] The trajectory points are connected to obtain the trajectory of the fire monitor jet.

[0021] Optionally, screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes:

[0022] Obtaining the trajectory of the fire monitor jet in the current image;

[0023] Acquire the length and angle of the trajectory according to the trajectory of the fire monitor jet;

[0024] Determine whether the length and angle of the track in the current image are both within the length range and the angle range;

[0025] When the length and angle of the trajectory are not both within the length range and the angle range, determining whether the deviation of the length and angle of the trajectory from the previous frame is within a preset threshold;

[0026] When the deviations of the length and angle of the track in the current picture from the length and angle of the track in the previous frame are not both within the preset threshold, the track is discarded;

[0027] When the deviations of the length and angle of the trajectory in the current picture and the length and angle of the trajectory in the previous frame are both within a preset threshold, the trajectory is determined as the initial trajectory.

[0028] Optionally, screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes:

[0029] When the length and the angle of the trajectory are within the length range and the angle range, the trajectory is determined to be the initial trajectory.

[0030] Optionally, screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes:

[0031] Obtaining the length of the initial trajectory according to the straight-line distance between the starting point and the end point of the initial trajectory;

[0032] Obtaining an average grayscale according to the average grayscale of the pixel points where all trajectory points of the initial trajectory are located;

[0033] Performing quadratic fitting on the initial trajectory according to the least squares method, and taking the average value of the difference between the fitting curve and the original curve point by point as the fitting error;

[0034] According to formula (1), the fitness value of the initial trajectory is obtained:

[0035] f=m / (len*grey) Formula (1),

[0036] Among them, f is the fitness value corresponding to the initial trajectory, m is the fitting error, len is the length, and grey is the average grayscale;

[0037] The initial trajectory with the smallest fitness value is selected as the optimal jet trajectory.

[0038] Optionally, the obtained fitting error is squared and brought into formula (1).

[0039] Optionally, screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes:

[0040] Get the fitness value of the best jet trajectory in the current image and the quadratic coefficient of the fitting curve;

[0041] Determine whether the fitness value of the optimal jet trajectory in the current image is greater than a preset fitness value;

[0042] When the fitness value is less than the preset fitness value, determining whether the quadratic term coefficient is less than 0;

[0043] When the coefficient of the quadratic term is greater than 0, the optimal jet trajectory is determined to be the trajectory of the jet, and the end point of the optimal jet trajectory is determined to be the water-falling point of the jet;

[0044] When the fitness value is greater than the preset fitness value, the current search is invalid, and the search is performed in the next frame of the picture;

[0045] When the coefficient of the quadratic term is less than 0, the current search is invalid and the search is performed in the next frame of the picture.

[0046] Optionally, screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes:

[0047] When the fitness value and the quadratic term coefficient of the optimal jet trajectory meet the requirements, determining whether the frame number difference between the current image and the last frame trajectory saved is greater than 5 frames;

[0048] When the number of frames between the current image and the last frame of the trajectory saved is not more than 5 frames, determine whether the position difference between the end point of the best jet trajectory of the current image and the landing point of the last frame of the trajectory is greater than a preset difference;

[0049] When the position difference between the end point of the best jet trajectory of the current picture and the water-falling point of the trajectory of the previous frame is less than a preset difference, the best jet trajectory is determined to be the trajectory of the jet, and the end point of the best jet trajectory is determined to be the water-falling point of the jet;

[0050] When the number of frames between the current image and the last frame trajectory saved is greater than 5 frames, the comparison with the water-falling point of the last frame trajectory is skipped.

[0051] Through the above technical scheme, a multi-track search method for a fire-fighting water bubble jet provided by the present invention can obtain a picture of a fire-fighting water cannon, and then identify the picture, so as to obtain multiple areas of multiple suspected starting points of the jet in the current picture. Each area can contain a suspected jet starting point, and then information about multiple jet starting points of the jet can be obtained. According to the information of the jet starting point, the trajectory of the fire-fighting water cannon jet can be obtained, and the trajectory can be multiple. Among the multiple trajectories of the fire-fighting water cannon jet, the optimal trajectory of the fire-fighting water cannon can be screened as the optimal jet trajectory. The method can identify multiple trajectories of the jet in the picture, and can find the optimal jet trajectory of the fire-fighting water cannon jet.

[0052] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0054] Figure 1 is a flow chart of a multi-trajectory search method for a fire monitor jet according to an embodiment of the present invention;

[0055] Figure 2 It is a flowchart of obtaining information of a jet starting point of a multi-trajectory search method for a fire monitor jet according to an embodiment of the present invention;

[0056] Figure 3 A flowchart of a multi-trajectory search method for fire monitor jets according to an embodiment of the present invention for obtaining trajectories of fire monitor jets;

[0057] Figure 4 It is a flowchart of obtaining an initial trajectory of a multi-trajectory search method for a fire monitor jet according to an embodiment of the present invention;

[0058] Figure 5 It is a flow chart of obtaining the optimal jet trajectory of a multi-trajectory search method for a fire monitor jet according to an embodiment of the present invention;

[0059] Figure 6 It is a first flow chart of obtaining a water drop point of a multi-trajectory search method for a fire monitor jet according to an embodiment of the present invention;

[0060] Figure 7It is a second flow chart of obtaining the water drop point of a multi-trajectory search method for a fire-fighting water monitor jet according to an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0062] Figure 1 1 is a flow chart of a multi-track search method for a fire monitor jet according to an embodiment of the present invention. In the present invention, the process of performing a multi-track search for a fire monitor jet may include:

[0063] In step S1, a picture of a fire-fighting water monitor jet is obtained.

[0064] In step S2, the picture is recognized to obtain a plurality of regions of the jet starting point of the jet in the current picture.

[0065] In step S3, information about a plurality of jet starting points of the jet is acquired in the image.

[0066] In step S4, the trajectory of the fire monitor jet is obtained based on the information of multiple jet starting points in the current image.

[0067] In step S5, the trajectories of the fire monitor jet in the current image are screened to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory.

[0068] In the present invention, after obtaining a picture of the fire monitor jet, the picture can be identified, so that multiple areas of the jet starting point of the jet in the current picture can be obtained. A suspected jet starting point can be contained in each area, and then information about multiple jet starting points of the jet can be obtained in the picture. Through the information of the multiple jet starting points, multiple trajectories about the multiple jet starting points can be searched in the picture. The multiple trajectories can all be related to the fire monitor jet. After obtaining multiple trajectories about the fire monitor jet in the current picture, the multiple trajectories can be screened, so that the optimal trajectory about the fire monitor jet can be obtained as the optimal jet trajectory. The optimal jet trajectory is obtained by screening multiple trajectories in the current picture, so the optimal jet trajectory can be the trajectory closest to the real fire monitor jet.

[0069] In one embodiment of the present invention, Figure 2 As shown, the process of obtaining information about multiple jet starting points of the jet may include:

[0070] In step S6 , a plurality of regions in the image with respect to the starting point of the jet are acquired.

[0071] In step S7 , a first starting point and a first trajectory angle with respect to the first starting point are acquired in each area.

[0072] In step S8, a plurality of other starting points arranged in parallel in a direction perpendicular to the first trajectory angle are acquired in the region according to the first starting point and the first trajectory angle, and the trajectory angles of the plurality of other starting points are the same as the first trajectory angle.

[0073] In step S9, the position information and angle information of the first starting point and the remaining starting points in each area are collected to obtain the information of the jet starting point.

[0074] In the present invention, after obtaining multiple areas about the starting point of the jet in the image, the first starting point about the starting point of the jet and the first trajectory angle about the first starting point can be obtained in each area. After obtaining the first starting point and the first trajectory angle, multiple other starting points set in parallel in the vertical direction of the first trajectory angle can be obtained in the area according to the first starting point and the first trajectory angle, and the trajectory angles of the multiple other starting points are the same as the first trajectory angle. Acquiring multiple starting points in the same area can ensure that a trajectory curve that conforms to the law of this jet can be searched in each jet area. Starting points are set in multiple areas respectively, and when the area itself cannot distinguish the jet trajectory, a vector search method can be used to search for a jet trajectory that meets the requirements. After obtaining the position information and angle information of the first starting point and the remaining starting points in each area, the first starting point and the remaining starting points in each area can be combined to obtain information on the jet starting point.

[0075] In one embodiment of the present invention, in a picture of continuous frames, the deviations of the trajectories of the jets in two adjacent frames on the picture are generally not much different. Therefore, when obtaining the information of the jet starting point in the current picture, the starting point and initial trajectory angle of the best jet trajectory after screening in the previous frame can be retained as one of the jet starting points in the picture of the current frame. While ensuring the smooth progress of the search, the anti-interference ability of the multi-trajectory search method is enhanced to a certain extent.

[0076] In one embodiment of the present invention, Figure 3 As shown, the process of obtaining the trajectory of the fire monitor jet may include:

[0077] In step S10, all jet starting points of the fire monitor jet in each area in the current image are obtained.

[0078] In step S11 , the position of the jet start point and the trajectory angle with respect to the jet start point are acquired.

[0079] In step S12, the search range of the next trajectory point is obtained according to the position of the jet starting point and the trajectory angle.

[0080] In step S13, the search range of the next trajectory point is searched to search for the point with the largest grayscale in the search range as the next trajectory point, and the trajectory angle of the next trajectory point is obtained to perform subsequent trajectory point searches until the trajectory point of the landing point is found.

[0081] In step S14, the trajectory points are connected to obtain a trajectory of the fire monitor jet.

[0082] In the present invention, after obtaining all the jet starting points of the fire monitor jet in each of the multiple areas in the current picture, the trajectory angle of each jet starting point can be obtained again. Then, the search range of the next trajectory point can be obtained according to the position and trajectory angle of the jet starting point and the search can be performed. Within the search range, the point with the largest grayscale can be searched as the next trajectory point, and the trajectory angle of the trajectory point can be obtained, and subsequent trajectory searches can be performed again until the trajectory point of the landing point is searched. After obtaining the trajectory points in the current picture, the trajectory points can be connected according to the category of the jet starting point to obtain multiple trajectories about the fire monitor jet.

[0083] In one embodiment of the present invention, Figure 4 As shown, the steps of obtaining the optimal jet trajectory may include:

[0084] In step S15, the trajectory of the fire monitor jet in the current image is obtained.

[0085] In step S16, the length and angle of the trajectory are obtained according to the trajectory of the fire monitor jet.

[0086] In step S17, it is determined whether the length and angle of the track in the current image are both within the length range and the angle range.

[0087] In step S18, when the length and angle of the trajectory are not both within the length range and the angle range, it is determined whether the deviations of the length and angle of the trajectory from the previous frame are both within a preset threshold.

[0088] In step S19, when the deviations between the length and angle of the track in the current picture and the length and angle of the track in the previous frame are not both within the preset threshold, the track is discarded.

[0089] In step S20, when the deviations of the length and angle of the trajectory in the current picture from the length and angle of the trajectory in the previous frame are both within a preset threshold, the trajectory is determined as the initial trajectory.

[0090] When selecting the best jet trajectory in the trajectory in the current picture, all trajectories of the fire monitor jet in the current picture can be obtained. Then, the length and angle of the trajectory can be obtained according to the trajectory of the fire monitor jet. The length of the trajectory can be the length of the line between the starting point and the landing point of the trajectory, and the angle can be the slope of the line between the starting point and the landing point of the trajectory. After obtaining the length and angle of the trajectory, it can be determined whether the length and angle of the trajectory in the current picture are both within the length range and the angle range. When the length and angle of the trajectory are not both within the length range and the angle range, it can be determined again whether the deviation of the length and angle of the trajectory and the length and angle of the trajectory of the previous frame is within the preset threshold, that is, the correlation degree of the trajectory and the trajectory of the previous frame is determined. When the deviation of the length and angle of the trajectory in the current picture and the length and angle of the trajectory of the previous frame are not within the preset threshold, the trajectory is invalid, so the trajectory can be discarded. When the deviation between the length and angle of the trajectory in the current image and the length and angle of the trajectory in the previous frame is within a preset threshold, the trajectory can be used as one of the candidate trajectories of the optimal jet trajectory, and thus the trajectory can be selected as the initial trajectory.

[0091] In one embodiment of the present invention, Figure 4 As shown, in step S21, when the length and angle of the trajectory are both within the length range and the angle range, the trajectory is determined to be the initial trajectory.

[0092] In the present invention, when the length and the angle of the trajectory are both within the length range and the angle range, the trajectory can be used as one of the candidate trajectories of the optimal jet trajectory, and thus the trajectory can be determined as the initial trajectory.

[0093] In one embodiment of the present invention, Figure 5 As shown, the process of screening the best jet trajectory may include:

[0094] In step S22, the length of the initial trajectory is obtained according to the straight-line distance between the starting point and the end point of the initial trajectory.

[0095] In step S23, the average grayscale is obtained according to the average grayscale value of the pixel points where all the trajectory points of the initial trajectory are located.

[0096] In step S24, a quadratic term fitting is performed on the initial trajectory according to the least square method, and the average value of the point-by-point difference between the fitting curve and the original curve is taken as the fitting error.

[0097] In step S25, the fitness value of the initial trajectory is obtained according to formula (1):

[0098] f=m / (len*grey) Formula (1),

[0099] Among them, f is the fitness value corresponding to the initial trajectory, m is the fitting error, len is the length, and grey is the average grayscale.

[0100] In step S26, the initial trajectory with the smallest fitness value is selected as the optimal jet trajectory.

[0101] In the present invention, after obtaining the initial trajectory, the initial trajectory may be multiple, so the initial trajectory may be screened to obtain the best jet trajectory. The ideal jet trajectory may present parabolic characteristics and may be a smooth curve. At the same time, under normal circumstances, the area of ​​the jet trajectory is an area with a larger grayscale. Therefore, the smoother the trajectory, the longer the length, and the larger the average grayscale, the more it conforms to the ideal jet trajectory. In the screening process, the length of the initial trajectory may be obtained based on the straight-line distance between the starting point and the end point of the initial trajectory. After obtaining the length of the initial trajectory, the average grayscale may be obtained based on the average grayscale of the pixel points where all the trajectory points of the initial trajectory are located. The initial trajectory may be fitted with quadratic terms by the least squares method, and the average value of the difference between the fitting curve and the original curve point by point may be used as the fitting error. The fitness value of each initial trajectory may be obtained by formula (1). The fitness value takes into account the length of the initial trajectory, the average grayscale, and the smoothness of the fitting curve, so the initial trajectory with the smallest fitness value may be screened as the best jet trajectory.

[0102] In the present invention, when initially searching for the trajectory of the jet, the smoothness of the initial trajectory is good. In order to reduce the influence of the smoothness on the final screening result, the obtained fitting error can be squared and then substituted into formula (1) to obtain the fitness value of the initial trajectory.

[0103] In one embodiment of the present invention, Figure 6 As shown, the process of determining the landing point of the optimal jet trajectory may include:

[0104] In step S27, the fitness value of the best jet trajectory in the current image and the quadratic term coefficient of the fitting curve are obtained.

[0105] In step S28, it is determined whether the fitness value of the optimal jet trajectory in the current image is greater than a preset fitness value.

[0106] In step S29, when the fitness value is less than the preset fitness value, it is determined whether the quadratic term coefficient is less than 0.

[0107] In step S30, when the coefficient of the quadratic term is greater than 0, the optimal jet trajectory is determined as the trajectory of the jet, and the end point of the optimal jet trajectory is determined as the water-falling point of the jet.

[0108] In step S31, when the fitness value is greater than the preset fitness value, the current search is invalid and the search is performed in the next frame of the picture.

[0109] In step S32, when the coefficient of the quadratic term is less than 0, the current search is invalid and the search is performed in the next frame of the picture.

[0110] In the present invention, when obtaining the water drop point, the fitness value of the best jet trajectory in the current picture and the quadratic term coefficient of the fitting curve can be obtained. The process of selecting the best jet trajectory is also the process of gradually determining the final water drop point. After the best jet trajectory is selected, in order to ensure the accuracy of the result, a basic condition can be set, that is, the preset fitness value and the quadratic term coefficient of the fitting curve. After the fire monitor sprays the jet at high pressure, due to the influence of gravity, the jet trajectory will inevitably present a parabolic characteristic of opening downward. When the camera is not installed vertically, the jet can also present the same characteristics on the picture. When it is judged that the fitness value of the best jet trajectory after screening is greater than the preset fitness value or the quadratic term coefficient is less than 0, it can be determined that the result of this search is invalid and enter the search of the next frame. When the fitness value of the best jet trajectory is less than the preset fitness value and the quadratic term coefficient is greater than 0, it can be determined that the end point of the best jet trajectory is the water drop point of the jet, and the best jet trajectory is the trajectory of the fire monitor jet.

[0111] In one embodiment of the present invention, Figure 7 As shown, the process of screening the best jet trajectory may include:

[0112] In step S33, when the fitness value and the quadratic term coefficient of the optimal jet trajectory meet the requirements, it is determined whether the frame number difference between the current image and the last frame trajectory saved is greater than 5 frames.

[0113] In step S34, when the frame number difference between the current image and the last frame trajectory saved is not greater than 5 frames, it is determined whether the position difference between the end point of the best jet trajectory of the current image and the landing point of the last frame trajectory is greater than a preset difference.

[0114] In step S35, when the position difference between the end point of the best jet trajectory of the current picture and the water drop point of the previous frame trajectory is less than the preset difference, the best jet trajectory is determined as the trajectory of the jet, and the end point of the best jet trajectory is determined as the water drop point of the jet.

[0115] In step S36, when the number of frames between the current image and the last frame trajectory saved is greater than 5 frames, the comparison with the water-falling point of the last frame trajectory is skipped.

[0116] In the present invention, when the fitness value and the quadratic term coefficient of the optimal jet trajectory meet the requirements, it can be determined whether the frame number difference between the current picture and the previous frame trajectory is greater than 5 frames. In the continuous jet frame picture, the jet trajectory of two adjacent frames will also show a certain continuous characteristic, and the distance between their landing points will certainly not differ too much. When it is not greater than 5 frames, the previous frame trajectory can be referenced, so it can be determined whether the difference between the position of the end point of the optimal jet trajectory in the current picture and the landing point of the previous frame trajectory is greater than the preset difference. When the position difference between the end point of the optimal jet trajectory in the current picture and the water drop point of the previous frame trajectory is less than the preset difference, it can be determined that the optimal jet trajectory is the trajectory of the jet, and it can be determined that the end point of the optimal jet trajectory is the water drop point of the jet. When the position difference between the end point of the optimal jet trajectory in the current picture and the water drop point of the previous frame trajectory is greater than the preset difference, the end point may not be the water drop point, and the optimal jet trajectory may not be about the jet trajectory. This search is invalid, and the search of the next frame of the picture is entered. When the frame number difference between the current image and the last frame trajectory saved is greater than 5 frames, the last frame trajectory may not be of reference value, and the comparison with the water drop point of the last frame trajectory can be skipped, and the fitness value and quadratic term coefficient of the optimal jet trajectory can be directly judged whether they meet the requirements.

[0117] Through the above technical scheme, a multi-track search method for a fire-fighting water bubble jet provided by the present invention can obtain a picture of a fire-fighting water cannon, and then identify the picture, so as to obtain multiple areas of multiple suspected starting points of the jet in the current picture. Each area can contain a suspected jet starting point, and then information about multiple jet starting points of the jet can be obtained. According to the information of the jet starting point, the trajectory of the fire-fighting water cannon jet can be obtained, and the trajectory can be multiple. Among the multiple trajectories of the fire-fighting water cannon jet, the optimal trajectory of the fire-fighting water cannon can be screened as the optimal jet trajectory. The method can identify multiple trajectories of the jet in the picture, and can find the optimal jet trajectory of the fire-fighting water cannon jet.

[0118] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0119] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A multi-trajectory search method for fire monitor jets, characterized in that: The multi-track search method comprises: Obtaining pictures of the fire monitor jet; Identify the image to obtain multiple regions of the jet starting point of the jet in the current image; acquiring information about a plurality of jet starting points of the jet in the image; Acquire the trajectory of the fire monitor jet according to the information of multiple jet starting points in the current image; The trajectory of the fire monitor jet in the current image is screened to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory.

2. The multi-track search method according to claim 1, characterized in that: Acquiring information about a plurality of jet starting points of the jets in the image includes: Acquire a plurality of regions in the image about the starting point of the jet; Acquire a first starting point and a first trajectory angle about the first starting point in each of the regions; Acquire, in the region, a plurality of other starting points arranged in parallel in a direction perpendicular to the first trajectory angle according to the first starting point and the first trajectory angle, wherein the trajectory angles of the plurality of other starting points are the same as the first trajectory angle; The position information and angle information of the first starting point and the remaining starting points in each of the areas are collected to obtain the information of the jet starting point.

3. The multi-track search method according to claim 2, characterized in that: Acquiring information about multiple jet starting points of the jet in the picture includes: retaining the starting point and initial trajectory angle of the best jet trajectory after screening in the previous frame of the picture as one of the jet starting points in the current frame of the picture.

4. The multi-track search method according to claim 2, characterized in that: Acquiring the trajectory of the fire monitor jet according to the information of multiple jet starting points in the current picture includes: Obtain all jet starting points of the fire monitor jet in each area in the current image; Acquire the position of the jet starting point and the trajectory angle about the jet starting point; Obtaining a search range for the next trajectory point according to the position of the jet starting point and the trajectory angle; Search the search range of the next trajectory point to search for the point with the largest grayscale in the search range as the next trajectory point, and obtain the trajectory angle of the next trajectory point to search for subsequent trajectory points until the trajectory point of the landing point is found; The trajectory points are connected to obtain the trajectory of the fire monitor jet.

5. The multi-track search method according to claim 4, characterized in that: Screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes: Obtaining the trajectory of the fire monitor jet in the current image; Acquire the length and angle of the trajectory according to the trajectory of the fire monitor jet; Determine whether the length and angle of the track in the current image are both within the length range and the angle range; When the length and angle of the trajectory are not both within the length range and the angle range, determining whether the deviations of the length and angle of the trajectory from the previous frame are both within a preset threshold; When the deviations of the length and angle of the track in the current picture from the length and angle of the track in the previous frame are not both within the preset threshold, the track is discarded; When the deviations of the length and angle of the trajectory in the current picture and the length and angle of the trajectory in the previous frame are both within a preset threshold, the trajectory is determined as the initial trajectory.

6. The multi-track search method according to claim 5, characterized in that: Screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes: When the length and the angle of the trajectory are both within the length range and the angle range, the trajectory is determined to be the initial trajectory.

7. The multi-track search method according to claim 6, characterized in that: Screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes: Obtaining the length of the initial trajectory according to the straight-line distance between the starting point and the end point of the initial trajectory; Obtaining an average grayscale according to the average grayscale of the pixel points where all trajectory points of the initial trajectory are located; Performing quadratic fitting on the initial trajectory according to the least squares method, and taking the average value of the difference between the fitting curve and the original curve point by point as the fitting error; According to formula (1), the fitness value of the initial trajectory is obtained: f=m / (len*grey) Formula (1), Among them, f is the fitness value corresponding to the initial trajectory, m is the fitting error, len is the length, and grey is the average grayscale; The initial trajectory with the smallest fitness value is selected as the optimal jet trajectory.

8. The multi-track search method according to claim 7, characterized in that: The obtained fitting error is squared and brought into formula (1).

9. The multi-track search method according to claim 7, characterized in that: Screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes: Get the fitness value of the best jet trajectory in the current image and the quadratic coefficient of the fitting curve; Determine whether the fitness value of the optimal jet trajectory in the current image is greater than a preset fitness value; When the fitness value is less than the preset fitness value, determining whether the quadratic term coefficient is less than 0; When the coefficient of the quadratic term is greater than 0, the optimal jet trajectory is determined to be the trajectory of the jet, and the end point of the optimal jet trajectory is determined to be the water-falling point of the jet; When the fitness value is greater than the preset fitness value, the current search is invalid, and the search is performed in the next frame of the picture; When the coefficient of the quadratic term is less than 0, the current search is invalid and the search is performed in the next frame of the picture.

10. The multi-track search method according to claim 9, characterized in that: Screening the trajectory of the fire monitor jet in the current image to obtain the optimal trajectory of the fire monitor jet as the optimal jet trajectory includes: When the fitness value and the quadratic term coefficient of the optimal jet trajectory meet the requirements, determining whether the frame number difference between the current image and the last frame trajectory saved is greater than 5 frames; When the number of frames between the current image and the last frame of the trajectory saved is not more than 5 frames, determine whether the position difference between the end point of the best jet trajectory of the current image and the landing point of the last frame of the trajectory is greater than a preset difference; When the position difference between the end point of the best jet trajectory of the current picture and the water-falling point of the trajectory of the previous frame is less than a preset difference, the best jet trajectory is determined to be the trajectory of the jet, and the end point of the best jet trajectory is determined to be the water-falling point of the jet; When the number of frames between the current image and the last frame trajectory saved is greater than 5 frames, the comparison with the water-falling point of the last frame trajectory is skipped.

Images