Inspection method and device for power construction and electronic equipment

Abstract

The invention discloses an inspection method and device for power construction and electronic equipment. Relates to the field of electric power, and the method comprises the steps: obtaining a wide-angle image of an electric power construction site, the wide-angle image comprising an image obtained by carrying out the image collection of the electric power construction site through a wide-angle camera; based on the wide-angle image, a first operation area is determined, the first operation area comprises an area including an operation main body, and the operation main body comprises facilities which are operated by the operated object in the electric power construction site; and based on a pan-tilt inverse mapping strategy, performing inspection on the first operation area and the second operation area through a pan-tilt camera to determine whether a safety risk exists in the power construction site or not. The technical problem that the whole construction area is difficult to cover when a safety officer inspects a power construction site in the related technology is solved.

Description

Technical Field

[0001] The present invention relates to the field of electric power, and in particular to an inspection method, device and electronic equipment for electric power construction. Background Art

[0002] At power construction sites, the operational landscape exhibits a striking pattern of large scenes and small targets. Large-scale machinery such as tower cranes and large cranes is frequently used. These machines are not only bulky but also require complex and tedious operating procedures, involving numerous specialized techniques and safety regulations. In contrast, the workers themselves appear extremely small within the vast construction space, often occupying a tiny pixel area on surveillance footage, making them difficult to accurately capture and identify. However, the personal safety of workers remains paramount during power construction, especially in a construction environment fraught with uncertainty and complexity. The safety of workers directly impacts the smooth progress of the entire construction process. Once a safety incident occurs, it not only poses a serious threat to the lives and health of workers but can also trigger a chain reaction of delays, equipment damage, and other consequences, resulting in significant losses and adverse effects.

[0003] Currently, safety monitoring at construction sites is usually done manually by safety officers. However, since power construction sites are generally large in scale and have dispersed workers, and there is usually a considerable safety distance between them and various types of machinery and equipment, daily monitoring of construction sites faces severe challenges. In other words, with limited time and energy, safety officers find it difficult to comprehensively and meticulously cover the entire construction area, and there are major omissions in controlling potential risks.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] The embodiments of the present invention provide a patrol inspection method, device and electronic equipment for electric power construction, so as to at least solve the technical problem in the related art that it is difficult for safety officers to patrol the electric power construction site and cover the entire construction area.

[0006] According to one aspect of an embodiment of the present invention, there is provided a patrol method for electric power construction, comprising: acquiring a wide-angle image of an electric power construction site, wherein the wide-angle image comprises: an image obtained by acquiring an image of the electric power construction site by a wide-angle camera; determining a first working area based on the wide-angle image, wherein the first working area comprises: an area containing a working subject, wherein the working subject comprises: facilities in the electric power construction site where the working subject performs work; based on a pan-tilt inverse mapping strategy, patrolling the first working area and the second working area respectively by a pan-tilt camera to determine whether there is a safety risk in the electric power construction site, wherein the pan-tilt inverse mapping strategy comprises: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, the priority of patrolling the first working area is higher than the priority of patrolling the second working area, and the second working area is the area in the electric power construction site that does not contain the working subject.

[0007] Furthermore, based on the pan-tilt inverse mapping strategy, the first working area and the second working area are inspected respectively by the pan-tilt camera, including: determining all working points in the first working area and all working points in the second working area by the pan-tilt camera; adopting the pan-tilt inverse mapping strategy, merging all working points in the first working area and all working points in the second working area to obtain a target working point set; based on preset inspection parameters, each working point in the target working point set is inspected by the pan-tilt camera, and during the inspection process, the working object in each working point is tracked, wherein the preset inspection parameters are used to control the frequency of the pan-tilt camera inspecting each working point.

[0008] Furthermore, during the inspection process, the work object in each of the work points is tracked, including: after the pan-tilt camera inspects the target work point, obtaining the target parameters of the pan-tilt camera, wherein the target parameters include: the magnification of the pan-tilt camera for the picture, the vertical altitude angle and the horizontal altitude angle, wherein the target work point is a pre-set work point in the target work point set; determining the detection frame of the target work object of the target work point in the pan-tilt camera picture to obtain the target detection frame; based on the target parameters, moving the center point of the pan-tilt camera picture to the center point of the target detection frame to track the work object in each of the work points.

[0009] Furthermore, the gimbal inverse mapping strategy is adopted to merge all the work points in the first work area and all the work points in the second work area to obtain a target work point set, including: adopting the gimbal inverse mapping strategy to map the coordinate positions of all the work points in the first work area to the wide-angle space to obtain a first work point set, wherein the first work point set includes: the coordinates of all the work points in the first work area in the wide-angle space, and the wide-angle space includes: the coordinate system of the image area covered by the wide-angle camera; determining the second work point set associated with the second work area, wherein the second work point set includes: the coordinates of all the work points in the second work area in the wide-angle space; merging the first work point set and the second work point set to obtain a target work point set.

[0010] Furthermore, all work points in the first work area are determined by the pan-tilt camera, including: dividing the first work area in the wide-angle image according to a preset dividing direction to obtain M first sub-areas, where M is a positive integer; for the power construction site, setting multiple pan-tilt camera positions in each first sub-area; based on the positions of multiple pan-tilt cameras associated with each first sub-area, determining the work points of each first sub-area through the pan-tilt camera to obtain all work points in the first work area.

[0011] Furthermore, all work points in the second work area are determined by the pan-tilt camera, including: dividing the first work area in the wide-angle image into grids to obtain N second sub-areas, where N is an integer greater than M; eliminating target sub-areas of the N second sub-areas to obtain S second sub-areas, where the target sub-areas include: a second sub-area whose degree of overlap with any second sub-area exceeds a preset threshold, and S is a positive integer less than N; determining the work point of each second sub-area to obtain all work points in the second work area.

[0012] Furthermore, after adopting the pan-tilt inverse mapping strategy to merge all the operating points in the first operating area and all the operating points in the second operating area to obtain a target operating point set, it also includes: updating the preset inspection parameters and the target operating point set.

[0013] Furthermore, in the process of inspecting the first working area and the second working area respectively through the pan-tilt camera based on the pan-tilt inverse mapping strategy, it also includes: identifying whether there is a safety risk at the power construction site, and when it is identified that there is a safety risk at the power construction site, determining the coordinates of the working point with the safety risk in the pan-tilt camera screen to obtain a first coordinate; mapping the first coordinate to the coordinates of the image area covered by the wide-angle camera to obtain a second coordinate; obtaining the alarm area information in the wide-angle image, wherein the alarm area information includes: information of the alarm area and the non-alarm area; based on the alarm area information, determining whether the second coordinate is in the alarm area, and issuing an alarm prompt when the second coordinate is in the alarm area.

[0014] According to another aspect of an embodiment of the present invention, a patrol device for power construction is also provided, including: an acquisition unit for acquiring a wide-angle image of the power construction site, wherein the wide-angle image includes: an image obtained by collecting images of the power construction site by a wide-angle camera; a determination unit for determining a first working area based on the wide-angle image, wherein the first working area includes: an area containing a working subject, and the working subject includes: facilities in the power construction site where the working object performs work; an inspection unit for respectively inspecting the first working area and the second working area through a pan-tilt camera based on a pan-tilt inverse mapping strategy to determine whether there is a safety risk in the power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, the priority of inspecting the first working area is higher than the priority of inspecting the second working area, and the second working area is the area in the power construction site that does not contain the working subject.

[0015] Furthermore, the inspection unit includes: a determination subunit, used to determine all work points in the first work area and all work points in the second work area through the pan-tilt camera; a merging subunit, used to adopt the pan-tilt inverse mapping strategy to merge all work points in the first work area and all work points in the second work area to obtain a target work point set; an inspection subunit, used to inspect each work point in the target work point set through the pan-tilt camera based on preset inspection parameters, and track the work objects in each of the work points during the inspection process, wherein the preset inspection parameters are used to control the frequency of the pan-tilt camera inspecting each of the work points.

[0016] Furthermore, the inspection sub-unit includes: an acquisition module, used to obtain the target parameters of the pan-tilt camera after the pan-tilt camera inspects the target work point, wherein the target parameters include: the magnification of the pan-tilt camera to the picture, the vertical altitude angle and the horizontal altitude angle, wherein the target work point is a pre-set work point in the target work point set; a first determination module, used to determine the detection frame of the target work object of the target work point in the pan-tilt camera picture to obtain a target detection frame; a tracking module, used to move the center point of the pan-tilt camera picture to the center point of the target detection frame based on the target parameters, so as to track the work object in each of the work points.

[0017] Furthermore, the merging subunit includes: a first mapping module, used to adopt the gimbal inverse mapping strategy to map the coordinate positions of all work points in the first work area to the wide-angle space to obtain a first work point set, wherein the first work point set includes: the coordinates of all work points in the first work area in the wide-angle space, and the wide-angle space includes: the coordinate system of the image area covered by the wide-angle camera; a second mapping module, used to determine the second work point set associated with the second work area, wherein the second work point set includes: the coordinates of all work points in the second work area in the wide-angle space; a merging module, used to merge the first work point set and the second work point set to obtain a target work point set.

[0018] Furthermore, the determination subunit includes: a first segmentation module, used to segment the first working area in the wide-angle image according to a preset segmentation direction to obtain M first sub-areas, where M is a positive integer; a setting module, used to set multiple pan-tilt camera positions in each first sub-area for the power construction site; a second determination module, used to determine the working points of each first sub-area through the pan-tilt camera based on the positions of multiple pan-tilt cameras associated with each first sub-area, and obtain all working points in the first working area.

[0019] Furthermore, the determination subunit includes: a second segmentation module, used to divide the first working area in the wide-angle image into grids to obtain N second sub-areas, wherein N is an integer greater than M; a elimination module, used to eliminate target sub-areas of the N second sub-areas to obtain S second sub-areas, wherein the target sub-areas include: a second sub-area whose overlap with any second sub-area exceeds a preset threshold, and S is a positive integer less than N; a third determination module, used to determine the working point of each second sub-area to obtain all working points in the second working area.

[0020] Furthermore, the inspection unit also includes: an updating subunit, which is used to update the preset inspection parameters and the target operation point set after merging all the operation points in the first operation area and all the operation points in the second operation area by adopting the pan-tilt inverse mapping strategy to obtain the target operation point set.

[0021] Furthermore, the inspection unit also includes: a first processing subunit, used to identify whether there is a safety risk at the power construction site during the inspection of the first working area and the second working area respectively through the pan-tilt camera based on the pan-tilt inverse mapping strategy, and when it is identified that there is a safety risk at the power construction site, determine the coordinates of the working point with the safety risk in the pan-tilt camera screen to obtain a first coordinate; a mapping subunit, used to map the first coordinate to the coordinate of the image area covered by the wide-angle camera to obtain a second coordinate; an acquisition subunit, used to acquire alarm area information in the wide-angle image, wherein the alarm area information includes: information of the alarm area and the non-alarm area; a second processing subunit, used to determine whether the second coordinate is in the alarm area based on the alarm area information, and issue an alarm prompt when the second coordinate is in the alarm area.

[0022] According to another aspect of an embodiment of the present invention, an electronic device is also provided, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned inspection methods for power construction by executing the executable instructions.

[0023] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, which stores a computer program, wherein when the computer program runs, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned inspection methods for power construction.

[0024] In the present invention, a wide-angle image of the power construction site is obtained, wherein the wide-angle image includes: an image obtained by collecting images of the power construction site through a wide-angle camera; based on the wide-angle image, a first working area is determined, wherein the first working area includes: an area containing a working subject, and the working subject includes: facilities in the power construction site where the working subject performs work; based on a pan-tilt inverse mapping strategy, the first working area and the second working area are inspected separately by a pan-tilt camera to determine whether there are safety risks at the power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, and the priority of inspecting the first working area is higher than the priority of inspecting the second working area, and the second working area is an area in the power construction site that does not contain a working subject. This solves the technical problem in the related art that it is difficult to cover the entire construction area when inspecting the power construction site by a safety officer.

[0025] In the present invention, based on the wide-angle image of the power construction site and the coordinate conversion strategy between the pan-tilt camera and the wide-angle camera, the power construction site is inspected for safety, thereby achieving the purpose of automatically and accurately inspecting the power construction site, thereby realizing the technical effect of improving the efficiency and comprehensiveness of safety monitoring of the power construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1 is a flow chart of an optional inspection method for power construction according to an embodiment of the present invention;

[0028] Figure 2 is a schematic diagram of an optional power construction scenario according to an embodiment of the present invention;

[0029] Figure 3 is a schematic diagram of an optional inspection device according to an embodiment of the present invention;

[0030] Figure 4 This is an optional inspection flow chart of a power construction site according to an embodiment of the present invention;

[0031] Figure 5 is a schematic diagram of an optional tower polling according to an embodiment of the present invention;

[0032] Figure 6 is a schematic diagram of an optional wide-angle polling according to an embodiment of the present invention;

[0033] Figure 7is a schematic diagram of an optional inspection device for power construction according to an embodiment of the present invention;

[0034] Figure 8 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described 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 making creative efforts should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0037] It should be noted that the user information (including but not limited to user device information, user personal information, etc.), collected information and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0038] Example 1

[0039] According to an embodiment of the present invention, an optional method embodiment of an inspection method for power construction is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0040] Figure 1 FIG. 1 is a flow chart of an optional inspection method for power construction according to an embodiment of the present invention, such as Figure 1 As shown, the method includes the following steps:

[0041] Step S101 : Acquire a wide-angle image of the electric power construction site, wherein the wide-angle image includes: an image acquired by capturing the electric power construction site through a wide-angle camera.

[0042] The wide-angle image mentioned above may include: an image obtained by capturing images of the power construction site using a wide-angle camera.

[0043] Step S102 : determining a first operation area based on the wide-angle image, wherein the first operation area includes an area containing an operation subject, and the operation subject includes facilities where the operation subject performs operation in the power construction site.

[0044] The aforementioned working subject may include: a facility at a power construction site where work is being performed by a work object, such as a power tower. The aforementioned working object may be an object performing a work action, such as a worker or a robot performing the work. In this embodiment, an area containing the working subject may be detected in a wide-angle image to obtain a first working area. Figure 2 is a schematic diagram of an optional power construction scenario according to an embodiment of the present invention, such as Figure 2 As shown, it is a wide-angle image of a power construction site with an electric tower as the main operating body. The first operating area can be determined by identifying the wide-angle image.

[0045] Step S103: Based on the pan-tilt inverse mapping strategy, the first operating area and the second operating area are inspected separately by the pan-tilt camera to determine whether there is a safety risk at the power construction site. The pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera. The priority of inspecting the first operating area is higher than the priority of inspecting the second operating area. The second operating area is an area in the power construction site that does not contain an operating entity.

[0046] In this embodiment, based on the pan-tilt inverse mapping strategy, the first working area and the second working area can be inspected separately by the pan-tilt camera to determine whether there are safety risks at the power construction site. The frequency of inspections of the working points in the first working area can be higher than the inspection frequency of the working points in the second working area.

[0047] In the power construction scenario, the first step of autonomous monitoring is to automatically locate the work point. Common types of operations are generally divided into special operations and ordinary operations. Special operations include high-altitude operations, hoisting operations, and hot work operations. These operations have a higher risk factor. Therefore, the operation area containing special operations (ie, the first operation area) needs to be monitored intensively. In contrast, general operations on the ground are considered ordinary operations and are less dangerous. Usually, the operation area of ordinary operations (ie, the second operation area) does not need to be monitored intensively. During the construction process, the focus of monitoring can be determined based on the risk factor of the operation type to ensure that high-risk operations are strictly monitored, while ordinary operations are subject to routine monitoring.

[0048] Because high-altitude tower assembly involves both high-altitude work and hoisting, it carries a high risk factor in power construction scenarios. Therefore, every worker on the tower needs to be closely monitored. This requirement can be met by individually polling the towers to identify every worker on the towers.

[0049] Figure 3 is a schematic diagram of an optional inspection device according to an embodiment of the present invention, such as Figure 3 As shown, it includes: zoom visible light camera, speaker, laser radar, high-pixel visible light wide-angle camera, pan / tilt, switch, display, network port and power cord, which can be connected through Figure 3 The inspection equipment shown is used to inspect the power construction site.

[0050] In this embodiment, through the above steps, a safety inspection of the power construction site is conducted based on wide-angle images of the site and the coordinate conversion strategy between the pan-tilt camera and the wide-angle camera. This achieves the goal of automatically and accurately conducting safety inspections of the power construction site, thereby achieving the technical effect of improving the efficiency and comprehensiveness of safety monitoring at the power construction site. This solves the technical problem in related technologies whereby safety officers cannot cover the entire construction area during inspections of the power construction site.

[0051] Figure 4 This is an optional inspection flow chart of a power construction site according to an embodiment of the present invention, such as Figure 4 As shown, a wide-angle camera image (corresponding to a wide-angle image) can be obtained to locate the working pole tower (corresponding to the working subject), and then a pan-tilt camera can be used to patrol the tower to determine the location of the working point. After that, the pan-tilt camera can be used to patrol other areas (for example, other areas outside the area where the working pole tower is located) to determine the location of the working point (the location of the working point in other areas). Then, the working point can be observed according to the camera scheduling algorithm, and staff members can be selected within the working point for tracking and observation (i.e., inspection). According to the type of personnel operation, the corresponding algorithm is called for safety monitoring to re-establish the working point.

[0052] For example, in this embodiment, the pan-tilt camera performs full scene polling to identify and mark the work points in the picture. Then, through intelligent algorithms, the device can automatically identify key work points in the construction area and accurately record the information in the picture space of the wide-angle camera, providing work site positioning information for subsequent monitoring and analysis.

[0053] After identifying the work point, the direction of the pan-tilt camera can be automatically adjusted for observation based on the information of the point to be observed (work point). At the same time, in order to improve the accuracy of the alarm, a pan-tilt self-linkage strategy can be adopted. In this strategy, the pan-tilt can automatically adjust the rotation direction according to the movement information of the operator, thereby realizing dynamic observation of the operator. It can not only improve the accuracy of observation, but also ensure that the activities of the operator can be captured in a vast construction site, reducing the omission of effective information. However, during the automatic movement of the pan-tilt, some special areas may be scanned, such as rest areas, non-working areas, etc. Even if these areas appear in the monitoring screen, they are not key areas that require security monitoring, so the alarm should not be triggered. In order to solve this problem, a pan-tilt inverse mapping strategy can be adopted. According to the preset area information, the personnel therein are not algorithmically identified to ensure that the system's alarm module is more intelligent and accurate;

[0054] It's important to note that after locating construction workers (i.e., the objects being worked on), the pan-tilt camera can be used to capture image information, and intelligent recognition technology can be used to determine the type of work being performed. Based on different work scenarios, the corresponding violation recognition algorithm is automatically invoked to analyze violations in real time during the construction process, such as failure to wear safety equipment or illegal operations. This process then issues alerts, records relevant information, and provides instant feedback to safety officers.

[0055] Optionally, based on the pan-tilt inverse mapping strategy, the first work area and the second work area are inspected separately by the pan-tilt camera, including: determining all work points in the first work area and all work points in the second work area by the pan-tilt camera; adopting the pan-tilt inverse mapping strategy, merging all work points in the first work area and all work points in the second work area to obtain a target work point set; based on preset inspection parameters, each work point in the target work point set is inspected by the pan-tilt camera, and during the inspection process, the work object in each work point is tracked, wherein the preset inspection parameters are used to control the frequency of the pan-tilt camera inspecting each work point.

[0056] Because the first work area contains the main operating entities and has a higher risk factor than the second work area, the work points in the first and second work areas are determined differently. Specifically, the first and second work areas can be polled separately to determine their work points. A PTZ inverse mapping strategy can then be used to merge all work points in the first and second work areas to obtain the target work point set.

[0057] The above-mentioned preset inspection parameters may include: the inspection frequency of each operating point, and the inspection frequency of the operating points in the first operating area may be higher than the inspection frequency of the operating points in the second operating area.

[0058] After determining the target work point set, each work point in the target work point set can be inspected through the pan-tilt camera based on the preset inspection parameters. During the inspection process, the work objects in each work point can be tracked, achieving the purpose of automatic monitoring of the power construction scene.

[0059] Optionally, during the inspection process, the work object in each work point is tracked, including: after the pan-tilt camera inspects the target work point, obtaining the target parameters of the pan-tilt camera, wherein the target parameters include: the magnification of the pan-tilt camera to the picture, the vertical altitude angle and the horizontal altitude angle, wherein the target work point is a pre-set work point in the target work point set; determining the detection frame of the target work object of the target work point in the pan-tilt camera picture to obtain the target detection frame; based on the target parameters, moving the center point of the pan-tilt camera picture to the center point of the target detection frame to track the work object in each work point.

[0060] In this embodiment, the mapping from the wide-angle camera to the PTZ camera can be used to locate the work point (corresponding to the target work point). By mapping the coordinates of any point in the wide-angle camera, the horizontal azimuth angle and vertical altitude angle in the PTZ camera coordinate system can be converted. The conversion method is shown in the following formula:

[0061]

[0062] Among them, f x ,f y ,c x ,c y The parameter values can be obtained through camera calibration methods, and the coordinate mapping matrix A can be obtained by matching the wide-angle image and the PTZ camera image using an image matching algorithm.

[0063] For example, if the work object is a worker, after the gimbal is positioned at the work point, the gimbal camera can randomly select a worker within the field of view to track them. This involves the gimbal's automatic tracking strategy, and the specific process is as follows:

[0064] Define x pp The horizontal rotation angle of the gimbal corresponding to the movement of one pixel in the x-axis direction of the center point of the image at 1x focal length. pp The vertical rotation angle of the gimbal corresponds to the movement of one pixel in the y-axis direction of the center point of the image at 1x focal length. Assume that there is a human frame in the gimbal image at (x c ,y c ,w,h), the current magnification of the gimbal camera is z, and the vertical altitude angle and horizontal azimuth angle of the gimbal in the current state are (α c ,β c ), where x c ,y c ,w,h represent the center point position of the human body detection frame and the width and height of the human body detection respectively. Then, the center point of the gimbal is focused from the current position to the human body detection frame through the coordinate mapping method:

[0065]

[0066] The formula for determining the focal length of the gimbal camera is:

[0067]

[0068] Among them, S wc is the diagonal length of the human detection box.

[0069] Through the automatic linkage of the pan-tilt system, the tracking of operators can be effectively realized. During the tracking process, attribute analysis can be performed to improve the analysis efficiency. By introducing the automatic tracking of the pan-tilt system and the inverse mapping of the pan-tilt system, the accuracy of tracking operators can be improved.

[0070] Optionally, a gimbal inverse mapping strategy is adopted to merge all the work points in the first work area and all the work points in the second work area to obtain a target work point set, including: adopting a gimbal inverse mapping strategy to map the coordinate positions of all the work points in the first work area to the wide-angle space to obtain a first work point set, wherein the first work point set includes: the coordinates of all the work points in the first work area in the wide-angle space, and the wide-angle space includes: the coordinate system of the image area covered by the wide-angle camera; determining a second work point set associated with the second work area, wherein the second work point set includes: the coordinates of all the work points in the second work area in the wide-angle space; merging the first work point set and the second work point set to obtain a target work point set.

[0071] In this embodiment, the PTZ camera can calculate the inverse mapping formula F according to the gun-ball linkage inv (Corresponding to the PTZ inverse mapping strategy) S in the ball camera space ptz The coordinate position of the human body on the tower (x ptz ,y ptz ,α,β,z) is mapped to the wide-angle space S w and record the corresponding position (x sw_ptz ,y sw_ptz ). For wide-angle space S w All recorded points are numbered. If there are n points in total, then we get the set P t = {1,...,n} (corresponding to the first set of working points), then the coordinates of the working points in the second working area in the wide-angle space can be determined. Assuming that there are m points in the second working area, then the set P g ={1,...,m} (corresponding to the second operation point set).

[0072] In this embodiment, the set P can also be t and set P g Merge, since the set P t and set P g All elements in the wide-angle space S w , so we can directly merge to get the set P:

[0073] P=P t ∪P g

[0074] Number the obtained P = {p1, p2, ..., p k}, where k = m + n;

[0075] Assume that the access weight of each seat is equal at the beginning of polling:

[0076] F base =1 / k

[0077] The subsequent access weight calculation formula is as follows:

[0078] F i =F base *(1+α*D i / T i )

[0079] Among them, F i is the access frequency of the ith PTZ position, D i is the number of times the i-th PTZ camera detects the operating object (e.g., operator), T i is the total number of visits to the system, W iis the importance weight of the i-th PTZ camera position. In this embodiment, the camera position at the tower head has a higher access weight. α controls the degree of influence of human body detection results on the frequency. If you want the human body detection results to have a greater impact on the frequency, increase the value of α. β controls the degree of influence of camera position importance on the frequency. If you want the camera position importance to have a greater impact on the frequency, increase the value of β.

[0080] Optionally, all work points in the first work area are determined by a pan-tilt camera, including: dividing the first work area in the wide-angle image according to a preset dividing direction to obtain M first sub-areas, where M is a positive integer; for the power construction site, setting multiple pan-tilt camera positions in each first sub-area; based on the positions of multiple pan-tilt cameras associated with each first sub-area, determining the work points of each first sub-area by the pan-tilt camera to obtain all work points in the first work area.

[0081] The above-mentioned preset dividing direction can be from top to bottom. In this embodiment, the first working area can be divided into M first sub-areas. The pan-tilt camera can poll the M first sub-areas in turn. The working point of each first sub-area can be determined through the pan-tilt camera to obtain all the working points of the first working area.

[0082] Taking the operation subject of the first operation area as an example, the tower can be divided into multiple sections from top to bottom. Finally, the tower is cut into sections and polled in turn (i.e., tower polling). During the polling process, one section of the tower is photographed each time.

[0083] Figure 5 This is an optional tower polling diagram according to an embodiment of the present invention, and the tower is cut into pieces as shown in FIG. Figure 5 As shown, the black rectangular frame is the location of the power tower detected by the wide-angle image. The tower is divided into three sections according to its height, marked with black Arabic numerals. Each section is divided into multiple PTZ camera positions, marked with white Arabic numerals. Focus on each camera position and perform human body detection.

[0084] If an operator is found during the polling process, the PTZ camera calculates the inverse mapping formula F according to the gun-ball linkage inv S in the ball machine space ptz The coordinate position of the human body on the tower (x ptz ,y ptz ,α,β,z) is mapped to the wide-angle space S w and record the corresponding position (x sw_ptz ,y sw_ptz ). After the tower polling is completed, the wide-angle space S w All recorded points are numbered. If there are n points in total, then we get the set P t={1,...,n}, achieving the technical effect of accurately determining the operators in the first operating area

[0085] Optionally, all work points in the second work area are determined by the pan-tilt camera, including: dividing the first work area in the wide-angle image into grids to obtain N second sub-areas, where N is an integer greater than M; eliminating target sub-areas of the N second sub-areas to obtain S second sub-areas, where the target sub-areas include: a second sub-area whose degree of overlap with any second sub-area exceeds a preset threshold, and S is a positive integer less than N; determining the work point of each second sub-area to obtain all work points in the second work area.

[0086] For example, to more accurately analyze wide-angle images and ensure effective monitoring, wide-angle images are sliced and polled after completing tower polling. Figure 6 is a schematic diagram of an optional wide-angle polling according to an embodiment of the present invention, such as Figure 6 As shown, considering the device parameters and the requirements of the actual application scenario, the wide-angle image can be divided into preset equal parts (for example, 5 equal parts) in the x-axis and y-axis directions to form a uniform grid. The segmentation method is as follows Figure 6 In this way, each monitoring screen is divided into multiple small areas (corresponding to N second sub-areas), which facilitates more accurate tracking and analysis.

[0087] Next, the wide-angle camera's monitoring points are numbered for subsequent data recording and analysis. During this processing, a key judgment condition can be set: if a point overlaps with 80% of the detection frame area on the upper part of the tower, the point is considered invalid and the monitoring task at that point is abandoned.

[0088] The locations of workers in the area where the tower is located (corresponding to the first operating area) are determined by tower polling, while the locations of workers who may appear near the ground of the tower base (the second operating area) are determined by wide-angle polling, which reduces repeated detection and improves monitoring efficiency.

[0089] like Figure 6 As shown in the figure, the black rectangle is the detection frame of the tower, the white rectangle inside the black rectangle is the 80% detection frame area of the upper part of the tower, where 80% of the data is converted from the black rectangle, and the gray grid rectangle is the polling point. All points (operating points) that meet the conditions are numbered. Assuming there are m points in total, then the set P will be obtained. g ={1,...,m}(within the second working area).

[0090] Optionally, after adopting the pan-tilt inverse mapping strategy to merge all the operating points in the first operating area and all the operating points in the second operating area to obtain the target operating point set, the method further includes: updating the preset inspection parameters and the target operating point set.

[0091] Due to the implementation of the operation point scheduling strategy, different weights can be assigned to different operation points according to their priorities, thereby affecting the access frequency of the operation points.

[0092] In practice, certain work points, requiring frequent inspection and monitoring, can be accessed more frequently. Meanwhile, access frequencies can be reduced for points without active work. This helps optimize resource allocation, reduces unnecessary monitoring, and improves overall inspection efficiency. Frequently reestablishing work points is impractical, especially at points without personnel. Therefore, a timer mechanism can be introduced to determine whether a work point needs to be reestablished. Typically, the reestablishment period for work points is set at 15 minutes. This interval is chosen based on the actual construction site conditions, particularly changes in facilities such as tower height. As tower construction progresses, tower height and the layout of the work area constantly change. Therefore, regularly reestablishing work points ensures that the monitoring system can maintain accurate and effective monitoring of the construction site despite these changes.

[0093] Optionally, in the process of inspecting the first working area and the second working area respectively through the pan-tilt camera based on the pan-tilt inverse mapping strategy, it also includes: identifying whether there is a safety risk at the power construction site, and when it is identified that there is a safety risk at the power construction site, determining the coordinates of the working point with the safety risk in the pan-tilt camera screen to obtain the first coordinate; mapping the first coordinate to the coordinates of the image area covered by the wide-angle camera to obtain the second coordinate; obtaining the alarm area information in the wide-angle image, wherein the alarm area information includes: information of the alarm area and the non-alarm area; based on the alarm area information, determining whether the second coordinate is in the alarm area, and issuing an alarm prompt when the second coordinate is in the alarm area.

[0094] In order to further increase the number of effective alarms and eliminate the influence of irrelevant areas, a non-alarm area S (corresponding to the non-alarm area) can be marked in the wide-angle image. When an alarm is generated, the current magnification of the gimbal camera is obtained as z, the vertical altitude angle and the horizontal azimuth angle are (α, β), and the center point of the gimbal camera at this time (cx ptz ,cy ptz ) corresponds to the coordinate point of the wide-angle camera (X, Y), let A -1 , a, b are brought into the coordinate system conversion formula of the wide-angle-pan-dome camera and simplified to obtain:

[0095]

[0096] Among them, cosα, sinα, sinβ, and A -1 is known.

[0097] The wide-angle corresponding coordinates are solved based on the least squares method, and the alarms in irrelevant areas (non-alarm areas) are filtered out through the coordinates. Alarms can be issued for the coordinates in the alarm area.

[0098] In this embodiment, automatic identification of the operation type can be achieved through multi-sensor fusion to ensure accurate detection of the operation scene and personnel: since the inspection equipment is equipped with a lidar, a wide-angle camera and a pan-tilt camera, specifically, the wide-angle camera provides a wide field of view to ensure that monitoring covers every corner of the construction site; the pan-tilt camera performs flexible viewing angle adjustment, tracks the dynamics of key areas or personnel in real time, and provides clear input data; the radar system provides precise spatial positioning capabilities; and the integrated intelligent recognition technology automatically identifies operators, equipment and potential safety hazards, and issues real-time alarms and marks dangerous areas.

[0099] Sensor calibration establishes conversion relationships between different coordinate systems, enabling highly integrated monitoring capabilities. Specifically, calibration establishes coordinate conversion relationships between the LiDAR and wide-angle camera, and between the PTZ camera and wide-angle camera. Furthermore, the PTZ camera can also accurately establish the corresponding relationship between the PTZ and wide-angle cameras through inverse mapping technology. Therefore, the effective combination of these three devices enables the system to leverage the visual perception capabilities of the PTZ and the spatial perception capabilities of the radar to collaboratively identify the type of operation and detect subsequent violations.

[0100] Through the technical means of multi-sensor fusion, the safety monitoring system can detect cranes in two-dimensional space, and combine radar information to accurately determine whether the crane is operating, thereby determining whether the current scene is a lifting operation scene. This recognition capability provides the basis for subsequent analysis of illegal behaviors.

[0101] For example, during a lifting operation, the system combines 2D images with 3D spatial information to identify violations such as whether someone is standing beneath the object or working beneath the boom. Furthermore, by combining the position of the person, the crane, and the boom, the system can monitor the safety of the lifting operation in real time.

[0102] It is also possible to combine human body detection in two-dimensional images with radar information to determine whether the operator is working at a height, and further identify whether the operator is wearing a safety rope and whether he complies with the "one step, one hang" operating safety regulations, thereby reducing the safety risks of working at height.

[0103] For hot work scenarios, the system detects people, gas cylinders, and flames in a 2D image to determine whether the work is currently hot work. It also analyzes whether the operator is wearing protective gear such as goggles and face shields. The system also uses radar information to detect the safe distance between oxygen and acetylene cylinders to ensure compliance with safety regulations.

[0104] For ordinary work scenarios, the system determines in real time whether the workers are wearing necessary safety equipment such as helmets and work clothes while the gimbal is tracking them.

[0105] Since the specific violation behavior identification algorithm is not the focus of this solution, it will not be elaborated in detail. However, through the combination of the above methods, the safety supervision system can comprehensively identify safety risks in various operating scenarios.

[0106] In an optional example, autonomous inspection of a power construction scene may include the following steps:

[0107] Step 1: Autonomous spatial positioning of operators, tower polling strategy, and formulation and allocation of operation point weights;

[0108] Step 2: Use the dome camera's automatic linkage strategy to locate the work point, automatically track the operator, and mark special areas in the wide-angle image for alarm processing;

[0109] Step 3: Automatically identify the type of work performed through multi-sensor fusion to ensure accurate detection of work scenes and personnel;

[0110] Step 4: Re-establish the operation point based on factors such as priority and access frequency.

[0111] In this embodiment, a laser radar can also be used to scan the scene, acquiring a 3D point cloud and providing two-dimensional position information for each element in the construction scene. Combined with multi-sensor information, this enhances the ability to analyze and identify personnel behavior, enriching the means of construction safety monitoring. This can better address the challenges of personnel safety monitoring in complex power construction scenarios, such as accurately identifying the behaviors of personnel engaged in special operations such as high-altitude work, hoisting, and hot work, thereby reducing safety risks. This system also enables the fusion of wide-angle and dome camera images of targets with different fields of view within the same scene. This not only expands scene coverage, ensuring no blind spots in the construction site, but also effectively tracks and senses dynamic changes in detected targets, promptly identifying safety hazards. For example, in areas where large machinery is operating, it can quickly capture changes in the relative position of personnel and equipment, improving overall monitoring effectiveness and timeliness. The dome camera's automatic linkage strategy not only enables automatic tracking of personnel, but also significantly reduces the camera's computing resource consumption. This not only enhances the intelligence of the safety monitoring system, enabling it to quickly respond to personnel activities, but also improves real-time performance, ensuring timely processing and feedback of monitoring information, ensuring efficient and stable operation of construction site safety monitoring.

[0112] In this embodiment, through the fusion and intelligent analysis of multi-sensor data, by combining two-dimensional image information and radar information, it is possible to monitor the power construction site in real time and accurately, realize automatic identification of the operation scene, and further identify various types of violations, timely discover potential risks and take corresponding preventive measures to ensure the safety of the operation process.

[0113] Example 2

[0114] The second embodiment of the present invention provides an optional inspection device for electric power construction, and each implementation unit in the inspection device corresponds to each implementation step in the first embodiment.

[0115] Figure 7 FIG. 1 is a schematic diagram of an optional inspection device for power construction according to an embodiment of the present invention, as shown in FIG. Figure 7 As shown, it includes: an acquisition unit 71, a determination unit 72 and an inspection unit 73.

[0116] The acquisition unit 71 is configured to acquire a wide-angle image of the electric power construction site, wherein the wide-angle image includes: an image acquired by capturing the electric power construction site through a wide-angle camera;

[0117] The determining unit 72 is configured to determine a first operation area based on the wide-angle image, wherein the first operation area includes an area containing an operation subject, and the operation subject includes facilities where the operation subject performs operation at the power construction site;

[0118] The inspection unit 73 is used to inspect the first working area and the second working area respectively through the pan-tilt camera based on the pan-tilt inverse mapping strategy to determine whether there is a safety risk at the power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, the priority of inspecting the first working area is higher than the priority of inspecting the second working area, and the second working area is an area in the power construction site that does not contain the working subject.

[0119] In the inspection device for electric power construction provided in the second embodiment of the present invention, a wide-angle image of the electric power construction site can be obtained by an acquisition unit 71, wherein the wide-angle image includes: an image obtained by capturing the electric power construction site through a wide-angle camera; a first operation area is determined based on the wide-angle image by a determination unit 72, wherein the first operation area includes: an area containing an operation subject, and the operation subject includes: facilities in the electric power construction site where the operation subject is working; and an inspection unit 73, based on a pan-tilt inverse mapping strategy, inspects the first operation area and the second operation area respectively using a pan-tilt camera to determine whether there is a safety risk at the electric power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera; the priority of inspecting the first operation area is higher than the priority of inspecting the second operation area, and the second operation area is an area in the electric power construction site that does not contain an operation subject. This solves the technical problem in the related art that it is difficult for safety officers to inspect the electric power construction site and cover the entire construction area. In this embodiment, based on the wide-angle image of the power construction site and the coordinate conversion strategy between the pan-tilt camera and the wide-angle camera, a safety inspection of the power construction site is carried out, thereby achieving the purpose of automatically conducting accurate safety inspections of the power construction site, thereby realizing the technical effect of improving the efficiency and comprehensiveness of safety monitoring of the power construction site.

[0120] Optionally, in the inspection device for power construction provided in the second embodiment of the present invention, the inspection unit includes: a determination subunit, used to determine all work points in the first work area and all work points in the second work area through a pan-tilt camera; a merging subunit, used to adopt a pan-tilt inverse mapping strategy to merge all work points in the first work area and all work points in the second work area to obtain a target work point set; an inspection subunit, used to inspect each work point in the target work point set through a pan-tilt camera based on preset inspection parameters, and track the work object in each work point during the inspection process, wherein the preset inspection parameters are used to control the frequency of the pan-tilt camera inspecting each work point.

[0121] Optionally, in the inspection device for power construction provided in the second embodiment of the present invention, the inspection sub-unit includes: an acquisition module, which is used to obtain the target parameters of the pan-tilt camera after the pan-tilt camera inspects the target work point, wherein the target parameters include: the magnification of the pan-tilt camera to the picture, the vertical altitude angle and the horizontal altitude angle, wherein the target work point is a pre-set work point in the target work point set; a first determination module, which is used to determine the detection frame of the target work object of the target work point in the pan-tilt camera picture to obtain the target detection frame; a tracking module, which is used to move the center point of the pan-tilt camera picture to the center point of the target detection frame based on the target parameters, so as to track the work object in each work point.

[0122] Optionally, in the inspection device for power construction provided in Example 2 of the present invention, the merging sub-unit includes: a first mapping module, used to adopt a pan-tilt inverse mapping strategy to map the coordinate positions of all work points in the first work area to the wide-angle space to obtain a first work point set, wherein the first work point set includes: the coordinates of all work points in the first work area in the wide-angle space, and the wide-angle space includes: the coordinate system of the image area covered by the wide-angle camera; a second mapping module, used to determine the second work point set associated with the second work area, wherein the second work point set includes: the coordinates of all work points in the second work area in the wide-angle space; a merging module, used to merge the first work point set and the second work point set to obtain a target work point set.

[0123] Optionally, in the inspection device for power construction provided in Example 2 of the present invention, the determination subunit includes: a first segmentation module, used to segment the first working area in the wide-angle image according to a preset segmentation direction to obtain M first sub-areas, where M is a positive integer; a setting module, used to set the positions of multiple pan-tilt cameras in each first sub-area for the power construction site; a second determination module, used to determine the working points of each first sub-area through the pan-tilt camera based on the positions of multiple pan-tilt cameras associated with each first sub-area, and obtain all working points in the first working area.

[0124] Optionally, in the inspection device for power construction provided in the second embodiment of the present invention, the determination subunit includes: a second segmentation module, used to segment the first working area in the wide-angle image into grids to obtain N second sub-areas, where N is an integer greater than M; a elimination module, used to eliminate target sub-areas of the N second sub-areas to obtain S second sub-areas, where the target sub-areas include: a second sub-area whose degree of overlap with any second sub-area exceeds a preset threshold, and S is a positive integer less than N; a third determination module, used to determine the working point of each second sub-area to obtain all working points in the second working area.

[0125] Optionally, in the inspection device for power construction provided in Example 2 of the present invention, the inspection unit also includes: an updating subunit, which is used to merge all the work points in the first work area and all the work points in the second work area by adopting a pan-tilt inverse mapping strategy to obtain a target work point set, and then update the preset inspection parameters and the target work point set.

[0126] Optionally, in the inspection device for power construction provided in the second embodiment of the present invention, the inspection unit also includes: a first processing sub-unit, used to identify whether there is a safety risk at the power construction site during the inspection of the first working area and the second working area respectively through the pan-tilt camera based on the pan-tilt inverse mapping strategy, and when it is identified that there is a safety risk at the power construction site, determine the coordinates of the working point with the safety risk in the pan-tilt camera screen to obtain the first coordinate; a mapping sub-unit, used to map the first coordinate to the coordinate of the image area covered by the wide-angle camera to obtain the second coordinate; an acquisition sub-unit, used to acquire the alarm area information in the wide-angle image, wherein the alarm area information includes: information of the alarm area and the non-alarm area; a second processing sub-unit, used to determine whether the second coordinate is in the alarm area based on the alarm area information, and issue an alarm prompt when the second coordinate is in the alarm area.

[0127] The above-mentioned inspection device for power construction may also include a processor and a memory. The above-mentioned acquisition unit 71, determination unit 72 and inspection unit 73 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0128] The processor includes a core that retrieves the corresponding program unit from memory. One or more cores can be configured. By adjusting core parameters, the system performs safety inspections of power construction sites based on wide-angle images of the site and the coordinate conversion strategy between the pan-tilt camera and the wide-angle camera. This achieves the goal of automatically and accurately conducting safety inspections of power construction sites, thereby improving the efficiency and comprehensiveness of safety monitoring at power construction sites.

[0129] The above-mentioned memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0130] According to another aspect of an embodiment of the present invention, an electronic device is also provided, including: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned inspection methods for power construction by executing the executable instructions.

[0131] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, which stores a computer program, wherein when the computer program runs, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned inspection methods for power construction.

[0132] Figure 8is a schematic diagram of an electronic device according to an embodiment of the present invention, such as Figure 8 As shown, an embodiment of the present invention provides an electronic device 80, which includes a processor, a memory, and a program stored in the memory and runnable on the processor. When the processor executes the program, any one of the above-mentioned inspection methods for power construction is implemented.

[0133] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0134] In the above embodiments of the present invention, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only exemplary. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0136] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0137] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0138] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc. Various media that can store program codes.

[0139] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A patrol inspection method for electric power construction, characterized in that: include: Acquire a wide-angle image of the electric power construction site, wherein the wide-angle image includes: an image acquired by capturing the electric power construction site through a wide-angle camera; Determining a first operation area based on the wide-angle image, wherein the first operation area includes: an area containing an operation subject, and the operation subject includes: facilities where the operation subject performs operation in the power construction site; Based on the pan-tilt inverse mapping strategy, the first working area and the second working area are inspected respectively by the pan-tilt camera to determine whether there is a safety risk in the power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, the priority of inspecting the first working area is higher than the priority of inspecting the second working area, and the second working area is the area in the power construction site that does not contain the working entity.

2. The inspection method according to claim 1, characterized in that: Based on the pan-tilt inverse mapping strategy, the first operation area and the second operation area are inspected respectively by the pan-tilt camera, including: Determine all the working points in the first working area and all the working points in the second working area by using the pan-tilt camera; Adopting the pan-tilt inverse mapping strategy, all the operating points in the first operating area and all the operating points in the second operating area are merged to obtain a target operating point set; Based on the preset inspection parameters, each work point in the target work point set is inspected by the pan-tilt camera, and during the inspection process, the work object in each work point is tracked, wherein the preset inspection parameters are used to control the frequency of the pan-tilt camera inspecting each work point.

3. The inspection method according to claim 2, characterized in that: During the inspection process, the operating objects in each of the operating points are tracked, including: After the PTZ camera inspects a target operation point, obtaining target parameters of the PTZ camera, wherein the target parameters include: a magnification of the PTZ camera for an image, a vertical altitude angle, and a horizontal altitude angle, wherein the target operation point is a pre-set operation point in the target operation point set; Determine a detection frame of a target operation object at the target operation point in the pan / tilt camera image to obtain a target detection frame; Based on the target parameters, the center point of the pan-tilt camera image is moved to the center point of the target detection frame to track the operating object in each of the operating points.

4. The inspection method according to claim 2, characterized in that: Adopting the pan-tilt inverse mapping strategy, all the operating points in the first operating area and all the operating points in the second operating area are merged to obtain a target operating point set, including: Using the gimbal inverse mapping strategy, the coordinate positions of all work points in the first work area are mapped to a wide-angle space to obtain a first work point set, wherein the first work point set includes: the coordinates of all work points in the first work area in the wide-angle space, and the wide-angle space includes: the coordinate system of the image area covered by the wide-angle camera; Determining a second operating point set associated with the second operating area, wherein the second operating point set includes: coordinates of all operating points in the second operating area in the wide-angle space; The first operation point set and the second operation point set are merged to obtain a target operation point set.

5. The inspection method according to claim 2, characterized in that: Determining all the operating points in the first operating area by using the pan-tilt camera includes: Segmenting the first operating area in the wide-angle image according to a preset segmentation direction to obtain M first sub-areas, where M is a positive integer; For the power construction site, multiple pan-tilt camera positions are set in each of the first sub-areas; Based on the positions of multiple pan-tilt cameras associated with each first sub-area, the operating points of each first sub-area are determined by the pan-tilt cameras to obtain all the operating points in the first operating area.

6. The inspection method according to claim 2, characterized in that: Determining all the operating points in the second operating area by using the pan-tilt camera includes: The first operating area in the wide-angle image is divided into grids to obtain N second sub-areas, where: N is an integer greater than M; Eliminate target subregions from the N second subregions to obtain S second subregions, where the target subregions include: second subregions whose overlap with any second subregion exceeds a preset threshold, and S is a positive integer less than N; An operating point in each second sub-area is determined to obtain all operating points in the second operating area.

7. The inspection method according to claim 2, characterized in that: After adopting the pan-tilt inverse mapping strategy to merge all the operating points in the first operating area and all the operating points in the second operating area to obtain a target operating point set, the method further includes: updating the preset inspection parameters and the target operating point set.

8. The inspection method according to claim 1, characterized in that: In the process of inspecting the first operation area and the second operation area respectively by using the pan-tilt camera based on the pan-tilt inverse mapping strategy, the process further includes: Identifying whether there is a safety risk at the power construction site, and if it is identified that there is a safety risk at the power construction site, determining the coordinates of the work point where the safety risk exists in the image of the pan-tilt camera to obtain a first coordinate; Mapping the first coordinates to coordinates of an image area covered by the wide-angle camera to obtain second coordinates; Acquiring warning area information in the wide-angle image, wherein the warning area information includes: information of a warning area and a non-warning area; Based on the warning area information, it is determined whether the second coordinate is in the warning area, and if the second coordinate is in the warning area, an alarm prompt is issued.

9. A patrol inspection device for electric power construction, characterized in that: include: An acquisition unit is configured to acquire a wide-angle image of the electric power construction site, wherein the wide-angle image comprises an image acquired by acquiring an image of the electric power construction site through a wide-angle camera; a determining unit configured to determine a first operation area based on the wide-angle image, wherein the first operation area includes an area containing an operation subject, and the operation subject includes a facility where an operation subject performs operation at the power construction site; An inspection unit is used to inspect the first working area and the second working area respectively through a pan-tilt camera based on a pan-tilt inverse mapping strategy to determine whether there is a safety risk in the power construction site, wherein the pan-tilt inverse mapping strategy includes: a strategy for coordinate conversion between the pan-tilt camera and the wide-angle camera, the priority of inspecting the first working area is higher than the priority of inspecting the second working area, and the second working area is an area in the power construction site that does not contain the working subject.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the inspection method for power construction as described in any one of claims 1 to 8.

Images