A detection and management method, device and system for hidden dangers in power transmission line construction

Abstract

The present invention discloses a detection management method, device and system for hidden dangers in the construction of power transmission lines. The detection management device includes an acquisition calculation unit, a mechanical identification unit and a detection judgment unit. The detection management system includes a detection management module, a data storage module and a monocular camera. By pre-calculating and defining the height limit warning line, a construction machinery detection model with an SSD-MobileNet neural network structure is used to detect the line construction image to be detected and managed. The detection management method, device and system improve the efficiency of monitoring and reduce the pressure of data transmission; further, the detection management method, device and system for hidden dangers in the construction of power transmission lines provided by the present invention can flexibly determine whether remote early warning is needed according to the hidden danger detection results, thereby further reducing the pressure of data transmission.

Description

Technical Field

[0001] The present invention relates to the field of detection and management of hidden dangers in power transmission line construction, and relates to a detection and management method, device and system for hidden dangers in power transmission line construction. Background Art

[0002] Due to the wide and dense distribution of transmission lines, a large number of roads, ports, building construction sites and transmission line channels overlap. There are a large number of highly destructive, large working radius and ultra-high machines such as cranes, tower cranes and excavators on the construction site. In order to prevent these potential danger equipment from damaging the transmission system, it is necessary to continuously monitor these potential danger targets in real time.

[0003] In the prior art, since image and video equipment has the advantages of low cost and the ability to directly reflect on-site conditions, it is usually widely used in power grid monitoring systems to manually detect, identify and warn of construction hazards.

[0004] However, due to the large number of monitoring terminal devices, operation and maintenance personnel are unable to track the status of all lines used at the same time, resulting in low monitoring efficiency; at the same time, the amount of data carried by video images is large and mostly duplicate data, which makes the data transmission load heavy.

[0005] Therefore, there is a need for a method, device and system for detecting and managing hidden dangers in power transmission line construction. Summary of the invention

[0006] In view of the above-mentioned existing technical problems, the purpose of the present invention is to provide a detection and management method, device and system for hidden dangers in transmission line construction, so as to improve the efficiency of monitoring and reduce the pressure of data transmission.

[0007] The present invention provides a detection and management method for hidden dangers in power transmission line construction, the detection and management method comprising: calibrating an internal parameter matrix of a monocular camera according to a Zhang Zhengyou calibration method, and calculating a height limit warning line according to the internal parameter matrix; acquiring a line construction image to be detected and managed, and inputting the line construction image into a preset construction machinery detection model to obtain a detection result; the construction machinery detection model has an SSD-Mobile Net neural network structure; and acquiring a hidden danger detection result according to the detection result and the height limit warning line.

[0008] In one embodiment, before calibrating the intrinsic parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method and calculating the height limit warning line according to the intrinsic parameter matrix, the detection management method also includes: obtaining multiple historical construction hazard images, and marking the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, and obtaining a first historical construction hazard image; scaling each first historical construction hazard image to a preset size, and obtaining a second historical construction hazard image; inputting the second historical construction hazard image into the SSD-Mobile Net network for training to obtain a construction machinery detection model.

[0009] In one embodiment, the intrinsic parameter matrix of the monocular camera is calibrated according to the Zhang Zhengyou calibration method, and the height limit warning line is calculated according to the intrinsic parameter matrix, specifically including: determining the intrinsic parameter matrix of the camera through the Zhang Zhengyou calibration method and a preset four-parameter model; calculating and obtaining the height limit warning line according to the intrinsic parameter matrix and a preset height limit warning surface standard.

[0010] In one embodiment, the detection management method further includes: judging whether a remote warning is required based on the hidden danger detection result; if necessary, sending warning data and storing the warning data in a historical warning data group.

[0011] In one embodiment, based on the hidden danger detection result, it is determined whether a remote warning is required, specifically including: determining whether there is an ultra-high hidden danger in the hidden danger detection result; if there is an ultra-high hidden danger, determining whether there is warning data in the historical warning data group; if there is no warning data in the historical warning data group, a remote warning is required; if there is warning data in the historical warning data group, calculating the time interval between a first detection time corresponding to the hidden danger detection result and a second detection time corresponding to the warning data closest to the first detection time, and determining whether the time interval is greater than a preset interval threshold; if the time interval is greater than the preset interval threshold, a remote warning is required; if the time interval is not greater than the preset interval threshold, determining whether the first number of ultra-high hidden dangers in the hidden danger detection result is greater than the second number of ultra-high hidden dangers in the warning data corresponding to most second detection times; if the first number is greater than the second number, a remote warning is required.

[0012] The present invention also provides a detection and management device for hidden dangers in construction of power transmission lines, the detection and management device comprising a calibration calculation unit, a mechanical identification unit and a detection and judgment unit, wherein the calibration calculation unit is used to calibrate the internal parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method, and calculate the height limit warning line according to the internal parameter matrix; the mechanical identification unit is used to obtain the line construction image to be detected and managed, and input the line construction image into a preset construction machinery detection model to obtain the detection result; the construction machinery detection model has an SSD-Mobile Net neural network structure; the detection and judgment unit is used to obtain the hidden danger detection result according to the detection result and the height limit warning line.

[0013] In one embodiment, the detection management device also includes a model training unit, which is used to: obtain multiple historical construction hazard images, and mark the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, so as to obtain a first historical construction hazard image; scale each first historical construction hazard image to a preset size, so as to obtain a second historical construction hazard image; input the second historical construction hazard image into the SSD-Mobile Net network for training, so as to obtain a construction machinery detection model.

[0014] In one embodiment, the detection management device also includes a remote warning unit, which is used to: determine whether a remote warning is needed based on the hidden danger detection results; if necessary, send warning data and store the warning data in a historical warning data group.

[0015] The present invention also provides a detection and management system for hidden dangers in construction of power transmission lines, the detection and management system comprising a detection and management module, a data storage module and a monocular camera, the detection and management module, the data storage module and the monocular camera are communicatively connected, the data storage module is used to store all data, the detection and management module is used to execute the detection and management method for hidden dangers in construction of power transmission lines as described above, and the monocular camera is used to collect and send line construction images to be detected and managed to the detection and management module.

[0016] In one embodiment, the detection management system further includes a user interaction module, and the user interaction module is used to receive the early warning data sent by the detection management module and send the early warning data to the user.

[0017] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0018] The present invention provides a detection and management method, device and system for hidden dangers in power transmission line construction. By pre-calculating and defining a height limit warning line, a construction machinery detection model with an SSD-Mobile Net neural network structure is used to detect the line construction image to be detected and managed. The detection and management method, device and system improve the efficiency of monitoring and reduce the pressure of data transmission.

[0019] Furthermore, the method, device and system for detecting and managing hidden dangers in power transmission line construction provided by the present invention can flexibly determine whether remote warning is needed based on the hidden danger detection results, thereby further reducing the pressure of data transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described below in conjunction with the accompanying drawings, wherein:

[0021] Figure 1 A flow chart showing an embodiment of a method for detecting and managing hidden dangers in power transmission line construction according to the present invention;

[0022] Figure 2 An embodiment of a spatial relationship diagram between a height-limited warning surface and a camera is shown;

[0023] Figure 3 A flow chart showing another embodiment of a method for detecting and managing hidden dangers in power transmission line construction according to the present invention;

[0024] Figure 4 A structural diagram showing an embodiment of a device for detecting and managing hidden dangers in power transmission line construction according to the present invention;

[0025] Figure 5 A structural diagram of an embodiment of a detection and management system for hidden dangers in power transmission line construction according to the present invention is shown. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Specific embodiment 1

[0028] The embodiment of the present invention firstly describes a method for detecting and managing hidden dangers in construction of power transmission lines. Figure 1 A flow chart of an embodiment of a method for detecting and managing hidden dangers in power transmission line construction according to the present invention is shown.

[0029] like Figure 1 As shown, the detection management method includes the following steps:

[0030] S1: According to Zhang Zhengyou calibration method, calibrate the intrinsic parameter matrix of the monocular camera, and calculate the height limit warning line according to the intrinsic parameter matrix.

[0031] Since the identification of construction hazards involves the detection of the height of construction machinery in the image, and different cameras have different intrinsic parameters, after obtaining the line construction image to be detected and managed, it is necessary to first calibrate the intrinsic parameters of the monocular camera in order to calculate the height limit warning line.

[0032] Figure 2 An embodiment of a spatial relationship diagram between a height-limited warning surface and a camera is shown.

[0033] Specifically, the Zhang Zhengyou calibration method is used to determine the camera's intrinsic parameter matrix K. The intrinsic parameter matrix adopts a four-parameter model:

[0034]

[0035] Among them, f x , f y are the equivalent focal lengths of the camera in the x and y directions, respectively, (u 0 , v 0 ) is the principal point coordinate of the camera optical axis on the image. Then, according to the characteristics of camera geometric imaging, the image of the spatial plane passing through the optical axis is a straight line. The spatial plane passing through the camera optical axis and perpendicular to the imaging plane is used as the height limit warning surface, and the image of this plane is used as the height limit warning line. According to the obtained principal point coordinate (u 0 , v 0 ), calculate the warning line on the image: y = v 0 .

[0036] In one embodiment, the intrinsic parameter matrix of the monocular camera is calibrated according to the Zhang Zhengyou calibration method, and the height limit warning line is calculated according to the intrinsic parameter matrix, specifically including: determining the intrinsic parameter matrix of the camera through the Zhang Zhengyou calibration method and a preset four-parameter model; calculating and obtaining the height limit warning line according to the intrinsic parameter matrix and a preset height limit warning surface standard.

[0037] After calculating the height limit line, the calibrated camera should be installed on the tower. The optical axis of the camera should be lower than the lowest point of the conductor and as parallel to the conductor as possible, so that the spatial relationship between the height limit warning surface standard and the camera is as follows: Figure 2 shown.

[0038] S2: Acquire the line construction image to be inspected and managed, and input the line construction image into a preset construction machinery inspection model to obtain the inspection result.

[0039] The construction machinery detection model has an SSD-Mobile Net neural network structure. The detection result includes the type of construction machinery and the location of the construction machinery. The location of the construction machinery includes the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machinery in the image.

[0040] The line construction image captured by the camera is scaled to the preset model input size: 300×300. The scaled line construction image is input into the preset construction machinery detection model to detect the normalized position and confidence of the construction machinery on the scaled image:

[0041] (x 1 ,y 1 ,x 2 ,y 2 ,s);

[0042] Among them, (x 1 ,y 1 ) is the normalized image coordinate of the upper left point of the rectangle, (x 2 ,y 2 ) is the normalized image coordinate of the lower right point of the rectangle, s is the confidence of the detection result, x 1 ,y 1 ,x 2 ,y 2 , the value range of s is [0, 1].

[0043] Then, according to the preset confidence threshold T s , filter out all the values ​​whose threshold is greater than T s The detected normalized coordinates are converted into the pixel coordinates of the image:

[0044]

[0045] Where W and H are the width and height of the image captured by the camera, respectively. 1 ,y′ 1 )、(x′ 2 ,y′ 2 ) are the image coordinates of the upper left point and lower right point of the circumscribed rectangle of the construction machinery.

[0046] S3: Obtain hidden danger detection results based on the detection results and the height limit warning line.

[0047] After obtaining the detection results, the construction machinery can be classified into overheight hidden dangers and non-overheight hidden dangers according to the image coordinates of the upper left point and lower right point of the circumscribed rectangle of the construction machinery in the image in the detection results and the height limit warning line calculated in the above steps. Specifically, the highest point y′ of the construction machinery in the image is calculated. 1The image y-axis distance D from the height limit warning line is used to determine the type of the construction machine according to the distance D. In one embodiment, when the distance D is greater than or equal to 0, the corresponding construction machine is an overheight hazard; when the distance D is less than 0, the corresponding construction machine is a non-overheight hazard.

[0048] In one embodiment, the distance D is calculated as follows:

[0049] D = y′ 1 -v 0 ;

[0050] In the formula, y′ 1 is the ordinate value of the upper left point (or upper right point) of the circumscribed rectangle of the construction machinery, v 0 The vertical coordinate value of the main point coordinate.

[0051] The embodiment of the present invention describes a detection and management method for hidden dangers in the construction of power transmission lines. By pre-calculating and defining the height limit warning line, a construction machinery detection model with an SSD-Mobile Net neural network structure is used to detect the line construction image to be detected and managed. The detection and management method improves the efficiency of monitoring and reduces the pressure of data transmission. Specific embodiment 2

[0053] Furthermore, an embodiment of the present invention also describes a method for detecting and managing hidden dangers in construction of power transmission lines. Figure 3 A flow chart of another embodiment of a method for detecting and managing hidden dangers in power transmission line construction according to the present invention is shown.

[0054] like Figure 3 As shown, the detection management method includes the following steps:

[0055] A1: Acquire multiple historical construction hazard images, and mark the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, and obtain a first historical construction hazard image accordingly.

[0056] The bounding rectangle uses a normalized four-dimensional vector (x 1 ,y 1 ,x 2 ,y 2 ) expression, where (x 1 ,y 1 ) is the normalized image coordinate of the upper left point of the rectangle, (x 2 ,y 2 ) is the normalized image coordinate of the lower right point of the rectangle.

[0057] A2: scaling each first historical construction hazard image to a preset size to obtain a corresponding second historical construction hazard image.

[0058] In one embodiment, the preset size is 300×300.

[0059] A3: Input the second historical construction hazard image into the SSD-Mobile Net network for training to obtain a construction machinery detection model.

[0060] A4: According to Zhang Zhengyou's calibration method, calibrate the intrinsic parameter matrix of the monocular camera, and calculate the height limit warning line based on the intrinsic parameter matrix.

[0061] Since the identification of construction hazards involves the detection of the height of construction machinery in the image, and different cameras have different intrinsic parameters, the intrinsic parameters of the monocular camera must be calibrated before performing construction machinery detection on the construction image, so as to calculate the height limit warning line.

[0062] Figure 2 An embodiment of a spatial relationship diagram between a height-limited warning surface and a camera is shown.

[0063] Specifically, the Zhang Zhengyou calibration method is used to determine the camera's intrinsic parameter matrix K. The intrinsic parameter matrix adopts a four-parameter model:

[0064]

[0065] Among them, f x , f y are the equivalent focal lengths of the camera in the x and y directions, respectively, (u 0 , v 0 ) is the principal point coordinate of the camera optical axis on the image. Then, according to the characteristics of camera geometric imaging, the spatial plane through the optical axis is a straight line. The spatial plane through the camera optical axis and perpendicular to the imaging plane is used as the height limit warning surface, and the image of the plane is used as the height limit warning line. According to the obtained principal point coordinate (u 0 , v 0 ), calculate the warning line on the image: y = v 0 .

[0066] In one embodiment, the method of calibrating the intrinsic parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method and calculating the height limit warning line according to the intrinsic parameter matrix specifically includes: determining the intrinsic parameter matrix of the camera through the Zhang Zhengyou calibration method and a preset four-parameter model; calculating the height limit warning line according to the intrinsic parameter matrix and a preset height limit warning surface standard.

[0067] After calculating the height limit line, the calibrated camera should be installed on the tower. The optical axis of the camera should be lower than the lowest point of the conductor and as parallel to the conductor as possible, so that the spatial relationship between the height limit warning surface standard and the camera is as follows: Figure 2 shown.

[0068] A5: Obtain the line construction image to be inspected and managed, and input the line construction image into a preset construction machinery inspection model to obtain the inspection result.

[0069] The construction machinery detection model has an SSD-Mobile Net neural network structure. The detection result includes the type of construction machinery and the location of the construction machinery. The location of the construction machinery includes the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machinery in the image.

[0070] The line construction image captured by the camera is scaled to the preset model input size: 300×300. The scaled line construction image is input into the preset construction machinery detection model to detect the normalized position and confidence of the construction machinery on the scaled image:

[0071] (x 1 ,y 1 ,x 2 ,y 2 ,s);

[0072] Among them, (x 1 ,y 1 ) is the normalized image coordinate of the upper left point of the rectangle, (x 2 ,y 2 ) is the normalized image coordinate of the lower right point of the rectangle, s is the confidence of the detection result, x 1 ,y 1 ,x 2 ,y 2 , the value range of s is [0, 1].

[0073] Then, according to the preset confidence threshold T s , filter out all the values ​​whose threshold is greater than T s The detection result. The detected normalized coordinates are converted into the pixel coordinates of the image:

[0074]

[0075] Where W and H are the width and height of the image captured by the camera, respectively. 1 ,y′ 1 )、(x′ 2 ,y′ 2 ) are the image coordinates of the upper left point and lower right point of the circumscribed rectangle of the construction machinery.

[0076] A6: Obtain hidden danger detection results based on the detection results and the height limit warning line.

[0077] After obtaining the detection results, the construction machinery can be classified into overheight hidden dangers and non-overheight hidden dangers according to the image coordinates of the upper left point and lower right point of the circumscribed rectangle of the construction machinery in the image in the detection results and the height limit warning line calculated in the above steps. Specifically, the highest point y of the construction machinery in the image is calculated. 1 ' is the image y-axis distance D from the height limit warning line, thereby determining the type of the construction machine according to the distance D. In one embodiment, when the distance D is greater than or equal to 0, the corresponding construction machine is an overheight hazard; when the distance D is less than 0, the corresponding construction machine is a non-overheight hazard.

[0078] In one embodiment, the distance D is calculated as follows:

[0079] D = y′ 1 -v 0 ;

[0080] In the formula, y′ 1 is the ordinate value of the upper left point (or upper right point) of the circumscribed rectangle of the construction machinery, v 0 The ordinate value of the main point coordinates.

[0081] After obtaining the hidden danger detection results, it can be determined whether a remote warning is needed based on the hidden danger detection results so that the user can take timely measures against the corresponding hidden dangers.

[0082] In one embodiment, the detection management method further includes: judging whether a remote warning is required based on the hidden danger detection result; if necessary, sending warning data and storing the warning data in a historical warning data group.

[0083] Specifically, the judgment method is: if the current warning information is the first warning, a remote warning is issued; otherwise, the time interval Δt from the last report is calculated, and if Δt>T time , a remote warning is issued; otherwise, the number of extremely high hidden dangers N in the current alarm data is calculated c and the number of medium and high hidden dangers N in the previous warning data l For comparison, if Then a remote warning is carried out. Among them, the warning time threshold T time Pre-set.

[0084] In one embodiment, based on the hidden danger detection result, it is determined whether a remote warning is required, specifically including: determining whether there is an ultra-high hidden danger in the hidden danger detection result; if there is an ultra-high hidden danger, determining whether there is warning data in the historical warning data group; if there is no warning data in the historical warning data group, a remote warning is required; if there is warning data in the historical warning data group, calculating the time interval between a first detection time corresponding to the hidden danger detection result and a second detection time corresponding to the warning data closest to the first detection time, and determining whether the time interval is greater than a preset interval threshold; if the time interval is greater than the preset interval threshold, a remote warning is required; if the time interval is not greater than the preset interval threshold, determining whether the first number of ultra-high hidden dangers in the hidden danger detection result is greater than the second number of ultra-high hidden dangers in the warning data corresponding to most second detection times; if the first number is greater than the second number, a remote warning is required.

[0085] An embodiment of the present invention describes a method for detecting and managing hidden dangers in the construction of power transmission lines. By pre-calculating and defining a height limit warning line, a construction machinery detection model with an SSD-Mobile Net neural network structure is used to detect the line construction image to be detected and managed. The detection and management method improves the efficiency of monitoring and reduces the pressure on data transmission. Furthermore, the method for detecting and managing hidden dangers in the construction of power transmission lines described in an embodiment of the present invention further reduces the pressure on data transmission by flexibly judging whether remote early warning is needed based on the hidden danger detection results. Specific embodiment three

[0087] In addition to the above method, the embodiment of the present invention also describes a detection and management device for hidden dangers in transmission line construction. Figure 4 A structural diagram of an embodiment of a device for detecting and managing hidden dangers in construction of a power transmission line according to the present invention is shown.

[0088] like Figure 4 As shown, the detection management device includes a calibration calculation unit 11, a mechanical identification unit 12 and a detection judgment unit 13.

[0089] The calibration calculation unit 11 is used to calibrate the intrinsic parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method, and calculate the height limit warning line according to the intrinsic parameter matrix.

[0090] The machine identification unit 12 is used to obtain the line construction image to be detected and managed, and input the line construction image into a preset construction machine detection model to obtain a detection result. The construction machine detection model has an SSD-MobileNet neural network structure.

[0091] The detection and judgment unit 13 is used to obtain hidden danger detection results according to the detection results and the height limit warning line.

[0092] When it is necessary to detect and manage hidden dangers in the construction of power transmission lines, first, the intrinsic parameter matrix of the monocular camera is calibrated according to the Zhang Zhengyou calibration method by obtaining the calculation unit 11, and the height limit warning line is calculated according to the intrinsic parameter matrix; then, the line construction image to be detected and managed is obtained through the mechanical recognition unit 12, and the line construction image is input into a preset construction machinery detection model to obtain the detection result; finally, the hidden danger detection result is obtained according to the detection result and the height limit warning line by the detection judgment unit 13.

[0093] In one embodiment, the detection management device also includes a model training unit, which is used to: obtain multiple historical construction hazard images, and mark the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, so as to obtain a first historical construction hazard image; scale each first historical construction hazard image to a preset size to obtain a second historical construction hazard image; input the second historical construction hazard image into the SSD-Mobile Net network for training, so as to obtain a construction machinery detection model.

[0094] In one embodiment, the detection management device also includes a remote warning unit, which is used to: determine whether a remote warning is needed based on the hidden danger detection results; if necessary, send warning data and store the warning data in a historical warning data group.

[0095] An embodiment of the present invention describes a detection and management device for hidden dangers in the construction of power transmission lines. By pre-calculating and defining a height limit warning line, a construction machinery detection model with an SSD-Mobile Net neural network structure is used to detect the line construction image to be detected and managed. The detection and management device improves the efficiency of monitoring and reduces the pressure on data transmission. Furthermore, the detection and management device for hidden dangers in the construction of power transmission lines described in an embodiment of the present invention can flexibly determine whether remote early warning is needed based on the hidden danger detection results, thereby further reducing the pressure on data transmission. Specific embodiment 4

[0097] In addition to the above method and device, the present invention also describes a detection and management system for hidden dangers in power transmission line construction. Figure 5 A structural diagram of an embodiment of a detection and management system for hidden dangers in power transmission line construction according to the present invention is shown.

[0098] like Figure 5As shown, the detection management system includes a detection management module 1, a data storage module 2 and a monocular camera 3. The detection management module 1, the data storage module 2 and the monocular camera 3 are communicatively connected with each other. The data storage module 2 is used to store all data. The detection management module 1 is used to execute the detection management method for hidden dangers in the construction of power transmission lines as described above. The monocular camera 3 is used to collect and send line construction images to be detected and managed to the detection management module.

[0099] In one embodiment, the detection management system further includes a user interaction module, and the user interaction module is used to receive the early warning data sent by the detection management module and send the early warning data to the user.

[0100] In one embodiment, the user interaction module is a remote monitoring background.

[0101] Compared with the existing domestic technical solutions, the technical solution of the present invention has the following obvious advantages:

[0102] 1. Use SSD-Mobile Net network to identify potential danger targets in real time, and the recognition rate can reach 30 frames per second on the embedded processor.

[0103] 2. Use the main point of the monocular camera to set the high warning line of hidden dangers to determine whether the hidden danger target is high. Only the intrinsic parameters of the camera are used, and the structured information of the scene is not required. It is flexible and has a wide range of use.

[0104] 3. Real-time statistics on hidden danger distribution, automatically report alarm information according to changes in hidden danger data distribution of adjacent monitoring frames, effectively reducing the operating pressure of the server and the workload of operation and maintenance personnel.

[0105] The embodiment of the present invention describes a detection and management system for hidden dangers in the construction of power transmission lines. By pre-calculating and defining height limit warning lines, a construction machinery detection model with an SSD-Mobile Net neural network structure is used to detect line construction images to be detected and managed. The detection and management system improves the efficiency of monitoring and reduces the pressure on data transmission. Furthermore, the detection and management system for hidden dangers in the construction of power transmission lines described in the embodiment of the present invention can flexibly determine whether remote early warning is needed based on the hidden danger detection results, thereby further reducing the pressure on data transmission.

[0106] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting and managing hidden dangers in power transmission line construction. It is characterized in that The detection management method comprises: According to Zhang Zhengyou's calibration method, the intrinsic parameter matrix of the monocular camera is calibrated, and the height limit warning line is calculated according to the intrinsic parameter matrix; Acquire a line construction image to be detected and managed, and input the line construction image into a preset construction machinery detection model to obtain a detection result; the construction machinery detection model has an SSD-Mobile Net neural network structure; the detection result includes the type of construction machinery and the position of the construction machinery; the construction machinery position includes the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machinery in the image; Obtaining hidden danger detection results according to the detection results and the height limit warning line includes: Obtaining the ordinate of the highest point of the construction machinery in the image according to the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machinery in the image in the detection result; Calculate the distance between the ordinate value of the highest point and the ordinate value of the image principal point coordinate of the height limit warning line. If the distance is greater than or equal to 0, the corresponding construction machinery is determined to be an overheight hidden danger. Otherwise, the corresponding construction machinery is determined to be a non-overheight hidden danger. The ordinate value of the image principal point coordinate is the principal point coordinate of the camera optical axis imaged on the picture. According to Zhang Zhengyou's calibration method, the intrinsic parameter matrix of the monocular camera is calibrated, and the height limit warning line is calculated according to the intrinsic parameter matrix, which specifically includes: Determine the camera's intrinsic parameter matrix using Zhang Zhengyou's calibration method and the preset four-parameter model; The height limit warning line is calculated based on the internal parameter matrix and the preset height limit warning surface standard.

2. The method for detecting and managing hidden dangers in power transmission line construction according to claim 1, It is characterized in that Before calibrating the intrinsic parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method and calculating the height limit warning line according to the intrinsic parameter matrix, the detection management method further includes: Acquire multiple historical construction hazard images, and respectively mark the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, and correspondingly obtain a first historical construction hazard image; Respectively scaling each first historical construction hazard image to a preset size to obtain a corresponding second historical construction hazard image; The second historical construction hazard image is input into the SSD-Mobile Net network for training to obtain a construction machinery detection model.

3. The method for detecting and managing hidden dangers in power transmission line construction according to any one of claims 1 to 2, It is characterized in that The detection management method also includes: According to the hidden danger detection result, determine whether remote early warning is needed; If necessary, the warning data is sent and stored in the historical warning data group.

4. The method for detecting and managing hidden dangers in power transmission line construction according to claim 3, It is characterized in that According to the hidden danger detection results, determine whether remote warning is needed, including: Determine whether there is an extremely high hidden danger in the hidden danger detection result; If there is an extremely high hidden danger, determine whether there is warning data in the historical warning data group; If there is no warning data in the historical warning data group, a remote warning is required; If there is warning data in the historical warning data group, then calculating the time interval between the first detection time corresponding to the hidden danger detection result and the second detection time corresponding to the warning data closest to the first detection time, and judging whether the time interval is greater than a preset interval threshold; If the time interval is greater than the preset interval threshold, a remote warning is required; If the time interval is not greater than the preset interval threshold, determining whether the first number of extremely high hidden dangers in the hidden danger detection result is greater than the second number of extremely high hidden dangers in the warning data corresponding to the majority of second detection times; If the first number is greater than the second number, a remote warning is required.

5. A detection and management device for hidden dangers in power transmission line construction, It is characterized in that The detection management device includes a calibration calculation unit, a mechanical identification unit and a detection judgment unit, wherein: The calibration calculation unit is used to calibrate the internal parameter matrix of the monocular camera according to the Zhang Zhengyou calibration method, and calculate the height limit warning line according to the internal parameter matrix; The machine identification unit is used to obtain the line construction image to be detected and managed, and input the line construction image into a preset construction machine detection model to obtain a detection result; the construction machine detection model has an SSD-Mobile Net neural network structure; the detection result includes the type of construction machine and the position of the construction machine; the construction machine position includes the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machine in the image; The detection and judgment unit is used to obtain a hidden danger detection result according to the detection result and the height limit warning line, including: Obtaining the ordinate of the highest point of the construction machinery in the image according to the image coordinates of the upper left point and the lower right point of the circumscribed rectangle of the construction machinery in the image in the detection result; Calculate the distance between the ordinate value of the highest point and the ordinate value of the image principal point coordinate of the height limit warning line. If the distance is greater than or equal to 0, the corresponding construction machinery is determined to be an overheight hidden danger. Otherwise, the corresponding construction machinery is determined to be a non-overheight hidden danger. The ordinate value of the image principal point coordinate is the principal point coordinate of the camera optical axis imaged on the picture. According to Zhang Zhengyou's calibration method, the intrinsic parameter matrix of the monocular camera is calibrated, and the height limit warning line is calculated according to the intrinsic parameter matrix, which specifically includes: Determine the camera's intrinsic parameter matrix using Zhang Zhengyou's calibration method and the preset four-parameter model; The height limit warning line is calculated based on the internal parameter matrix and the preset height limit warning surface standard.

6. The detection and management device for hidden dangers in power transmission line construction according to claim 5, It is characterized in that The detection management device further includes a model training unit, which is used to: Acquire multiple historical construction hazard images, and respectively mark the circumscribed rectangle and category of the construction machinery in each historical construction hazard image, and correspondingly obtain a first historical construction hazard image; Respectively scaling each first historical construction hazard image to a preset size to obtain a corresponding second historical construction hazard image; The second historical construction hazard image is input into the SSD-Mobile Net network for training to obtain a construction machinery detection model.

7. The detection and management device for hidden dangers in power transmission line construction according to claim 6, It is characterized in that The detection management device further includes a remote early warning unit, which is used to: According to the hidden danger detection result, determine whether remote early warning is needed; If necessary, the warning data is sent and stored in the historical warning data group.

8. A detection and management system for hidden dangers in power transmission line construction. It is characterized in that The detection management system includes a detection management module, a data storage module and a monocular camera. The detection management module, the data storage module and the monocular camera are communicatively connected. The data storage module is used to store all data. The detection management module is used to execute the detection management method for hidden dangers in construction of power transmission lines as described in any one of claims 1 to 4. The monocular camera is used to collect and send line construction images to be detected and managed to the detection management module.

9. The detection and management system for power transmission line construction hidden dangers according to claim 8, It is characterized in that The detection management system further comprises a user interaction module, and the user interaction module is used to receive the early warning data sent by the detection management module and send the early warning data to the user.

Images