Intelligent Early Warning Method and System for UAV Inspection for Construction in Subway Security Areas

Abstract

The present invention discloses an intelligent early warning method and system for UAV inspection for construction in the subway security area, which relates to the technical field of UAVs and includes the following steps: S10, planning the inspection flight path and flight parameters of the UAV; S20, using the UAV to collect inspection videos along the set flight path; S30, determining the location of the subway security area in the inspection video and drawing the boundary line of the subway security area; S40, using the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, and taking screenshots of the inspection video according to the detected risk construction equipment; S50, obtaining the GPS information of the screenshots, and generating early warning information by combining the screenshots, the GPS information of the screenshots and business data and sending it to the staff for processing. The beneficial effect of the present invention: It provides a solid and reliable guarantee for the effective supervision of the subway security area.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs). More specifically, it relates to an intelligent early warning method and system for UAV inspections for construction in subway security areas. Background Art

[0002] As an important livelihood project in big cities, the safe operation of the subway not only depends on engineering construction and operation guarantee, but is also affected by many external factors. To avoid the impact of external operations such as engineering construction and exploration around the track and to protect the safe use of the rail transit structure, the construction department will set up a subway safety protection area within a specific range around the rail transit line and related structures, and conduct inspections and investigations on high-risk illegal construction situations such as well drilling, drilling, pipe jacking, and earth excavation occurring in the security area. However, traditional inspections mainly rely on manual inspections, which are inefficient and labor-intensive. Moreover, for some complex sections, it is not easy for personnel and vehicles to reach, resulting in many difficulties and loopholes in the inspection work.

[0003] In the current era of the rapid development of the security industry and computer vision technology, to ensure the safety of the subway and prevent unauthorized construction activities from damaging the rail transit structure, using fixed cameras to monitor within the subway security area has become a mainstream method. However, this method has some obvious deficiencies in practical applications. First, in terms of the monitoring coverage area, the environment of the subway security area is complex, with not only undulating terrain but also various building facilities densely distributed. Due to the limitations of the installation location and viewing angle of fixed cameras, they cannot cover every corner in all directions. Especially, subway lines often extend for several kilometers, and their security areas also extend accordingly. If all areas are to be covered by cameras, not only is the cost extremely high, but many areas simply do not have the installation conditions. Second, from the perspective of target recognition, the occlusion and overlap phenomena between different targets are extremely common, which undoubtedly increases the difficulty of timely discovery of potential safety hazards.

[0004] Therefore, the present invention provides an intelligent early warning method and system for UAV inspections for construction in subway security areas, which can not only conduct comprehensive inspections of the subway security area, but also improve the probability of discovering risk construction equipment, and can quickly obtain the location information of risk construction equipment and send it to the staff in a timely manner, providing a solid and reliable guarantee for the effective supervision of the subway security area. Summary of the Invention

[0005] To overcome the deficiencies of the prior art, the present invention provides an intelligent early warning method and system for UAV inspections for construction in subway security areas, which can not only conduct comprehensive inspections of the subway security area, but also improve the probability of discovering risk construction equipment, and can quickly obtain the location information of risk construction equipment and send it to the staff in a timely manner, providing a solid and reliable guarantee for the effective supervision of the subway security area.

[0006] The technical solution adopted by the present invention to solve its technical problems is: an intelligent early warning method for drone inspection during subway security area construction. The improvement lies in that the drone inspection method for subway security area construction includes the following steps:

[0007] S10. Plan the inspection flight path and flight parameters of the drone;

[0008] S20. Use the drone to collect inspection videos along the set flight path;

[0009] S30. Determine the location of the subway security area in the inspection video and draw the boundary line of the subway security area;

[0010] S40. Use the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, and take screenshots of the inspection video according to the detected risk construction equipment;

[0011] S50. Obtain the GPS information of the screenshots, and generate early warning information by combining the screenshots, the GPS information of the screenshots and business data, and send it to the staff for processing.

[0012] Further, in step S10, the specific method for planning the inspection flight path of the drone is as follows:

[0013] Obtain the accurate map of the subway line to be inspected, covering the line direction, station locations and surrounding terrain;

[0014] Determine the scope of the subway security area at different intervals of the subway line: set the boundary line of the subway security area at L meters outside the center line of the subway track, where the value range of L is between 30 and 70;

[0015] Taking the subway line to be inspected as the reference, set the flight path of the drone directly above the center line of the subway track.

[0016] Further, in step 10, the flight parameters of the drone include flight altitude and flight speed. Among them, the flight altitude of the drone is set between 70m and 100m to ensure that the videos taken by the drone can completely cover the subway security areas on both sides of the subway line; the flight speed of the drone is set between 5m / s and 10m / s to meet the inspection accuracy.

[0017] Further, in step S20, during the process of the drone collecting inspection videos, it is also necessary to collect the GPS data of the drone in real time. The GPS data includes longitude, latitude, altitude and timestamp.

[0018] Further, in step S30, the specific method for determining the position of the subway security area in the inspection video and drawing the boundary line of the subway security area is as follows:

[0019] S301, draw a virtual straight line perpendicular to the horizontal direction and passing through the center point of the image in the inspection video as the center line, where the center point of the image is the projection of the position of the drone;

[0020] S302, determine the position of the boundary line of the subway security area in the inspection video according to the distance Pd between the boundary line of the subway security area and the image center line, and draw the boundary line of the subway security area in the inspection video; where the calculation expression of Pd is:

[0021] ; in the formula, represents the actual distance between the boundary line of the subway security area and the center line of the subway track; represents the width of the inspection video image; represents the height of the inspection video image; represents the actual distance of the diagonal line in the inspection video image, The calculation expression of is:

[0022] ; in the formula, represents the relative height of the drone from the ground; represents the field of view angle of the drone, that is, the maximum angle range that the drone can shoot, The calculation expression of is:

[0023] , in the formula, represents the size of the image sensor of the camera equipped on the drone; represents the lens focal length.

[0024] Further, in step S40, before using the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, it is necessary to first deploy the trained target recognition model to the drone, where the specific training method of the target recognition model is as follows:

[0025] S401, use the drone to collect construction equipment data in various construction scenarios at different time periods every day, including various aerial shooting heights and angles, and construct a dataset with an aerial view mainly composed of excavators, pile drivers, drilling machines, and cranes;

[0026] S402, design a target recognition model based on the construction equipment in the subway security area by improving the target detection model yolov11;

[0027] In S403, the designed target recognition model in S402 is used to load the dataset for multiple rounds of training, and the performance of the target recognition model is evaluated after each round of training. The optimal model is selected according to the evaluation results and deployed on the drone.

[0028] Furthermore, in step S402, the specific improvement method of the target detection model yolov11 is as follows: data is enhanced through geometric transformation and color space transformation at the input end; in the backbone network, the adaptive rotation convolution kernel ARconv2d is used to replace the conventional convolution kernel in the C3K2 module to extract target features with different directions in the image, so as to output feature maps with different directions; in the detection head part, a rotation angle path is added to each detection head to extract features related to the rotation angle.

[0029] Furthermore, in step S50, the process of obtaining the GPS information of the screenshot, and generating a warning message by combining the screenshot, the GPS information of the screenshot, and the business data and sending it to the staff for processing specifically includes the following steps:

[0030] Determine the longitude, latitude, and altitude of the drone in the screenshot from the GPS information data collected by the drone through the timestamp of the screenshot;

[0031] Design a data structure to store the screenshot, the GPS information of the screenshot, and the business data, and convert the data structure into JSON format to generate a warning message;

[0032] Report the warning message to the remote control platform of the drone through the network communication protocol, and the remote control platform of the drone will push the warning message to the terminal device of the staff for processing.

[0033] An intelligent warning system for drone patrol in the subway security area during construction, which is applied to an intelligent warning method for drone patrol in the subway security area during construction as described above. The improvement lies in that the intelligent warning system for drone patrol in the subway security area during construction includes:

[0034] A flight mission planning module, which is used to set the patrol route and flight parameters of the drone;

[0035] A video acquisition module, which is used to collect patrol videos in real time along the route;

[0036] A subway security area identification and marking module, which is used to determine the subway security area and draw the boundary line of the subway security area;

[0037] A risk construction equipment identification and detection module, which is used to detect risk construction equipment in the subway security area in the patrol video by using the target recognition model;

[0038] A risk construction equipment extraction module, which is used to capture screenshots of the location of risk construction equipment in the inspection video and extract the GPS information of the screenshots;

[0039] An early warning module, which is used to generate early warning information by combining the screenshots, the GPS information of the screenshots and business data and send it to the staff for processing.

[0040] The beneficial effects of the present invention are as follows: First, this solution uses a drone to collect inspection videos along the planned flight path according to the planned flight parameters, which can ensure that the perspective of the drone shooting the inspection video can completely cover the subway security areas on both sides of the subway track area and can meet the inspection accuracy; Then, by determining the location of the subway security area in the inspection video and drawing the boundary line of the subway security area, the subway security area can be accurately located in the inspection video; Next, the target recognition model is used to detect the risk construction equipment in the subway security area within the boundary line, and screenshots of the inspection video are taken according to the detected risk construction equipment to improve the detection efficiency of the risk construction equipment; Finally, by obtaining the GPS information of the screenshots and generating early warning information by combining the screenshots, the GPS information of the screenshots and business data and sending it to the staff for processing, it can help the staff quickly obtain the location information of the risk construction equipment for timely processing; Therefore, the present invention can not only conduct all-round inspections on the subway security area, but also improve the discovery probability of risk construction equipment, and can quickly obtain the location information of risk construction equipment and send it to the staff in time, providing a solid and reliable guarantee for the effective supervision of the subway security area. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flowchart of an intelligent early warning method for drone inspection for subway security area construction according to the present invention;

[0042] Figure 2 It is a block diagram of an intelligent early warning system for drone inspection for subway security area construction according to the present invention;

[0043] Figure 3 It is a hardware structure diagram of an electronic device shown in an exemplary embodiment;

[0044] Figure 4 It is a block diagram of an electronic device shown in an exemplary embodiment;

[0045] Figure 5 It is a path diagram of the increased rotation angle of the detection head;

[0046] Figure 6 It is a structural diagram of an improved asymmetric detection head. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The present invention will be further described below with reference to the drawings and embodiments.

[0048] The concept, specific structure and technical effects of the present invention will be clearly and completely described below in combination with the embodiments and the accompanying drawings, so as to fully understand the purpose, features and effects of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present invention. In addition, all connection / connection relationships involved in the patent do not simply refer to the direct connection of components, but refer to the formation of a more optimal connection structure by adding or reducing connection accessories according to specific implementation situations. The various technical features in the present invention can be combined interactively without conflicting with each other.

[0049] Refer to Figure 1 As shown, the present invention discloses an intelligent early warning method for drone patrol in the construction of subway security areas. The drone patrol method for the construction of subway security areas includes the following steps:

[0050] S10, plan the patrol flight path and flight parameters of the drone; in this embodiment, during the specific application of the drone, the remote control platform of the drone sets the patrol flight path and flight parameters of the drone. After the setting is completed, the remote control platform controls the drone to execute the patrol task; specifically, the specific method for planning the patrol flight path of the drone is as follows:

[0051] Obtain an accurate map of the subway line to be patrolled, covering the line direction, station locations and surrounding terrain;

[0052] Determine the scope of the subway security area at different interval positions of the subway line: set the boundary line of the subway security area at L meters outside the center line of the subway track, where the value range of L is between 30 and 70;

[0053] Taking the subway line to be patrolled as the reference, set the flight path of the drone directly above the center line of the subway track;

[0054] The flight parameters of the drone include flight altitude and flight speed. Among them, the flight altitude of the drone is set between 70m and 100m to ensure that the video captured by the drone can completely cover the subway security areas on both sides of the subway line; the flight speed of the drone is set between 5m / s and 10m / s to meet the patrol accuracy;

[0055] It should be noted that in the process of patrol route planning, the accurate map of the subway line to be patrolled is the basis for planning the drone patrol route. The existence of map data enables the remote control platform to accurately understand the actual layout of the subway line, help formulate the most appropriate patrol route, and provide necessary geographic information support for subsequent route planning, ensuring that the drone's patrol mission can cover all areas that need to be inspected and reduce omissions; the determination of the scope of the subway security area is to ensure that the patrol drone can cover the security areas on both sides of the subway line, which is helpful to discover potential safety hazards. In addition, due to the different geographical locations of different sections of the subway line, the actual distance L between the subway track center and the subway security area needs to be set after an on-site investigation of the geographical environment of the subway line. Adaptive adjustment is performed; the drone's flight path is set directly above the centerline of the subway track, which means that the drone's inspection path will fly vertically above the subway line. This setting ensures the symmetry of the drone's flight path, allowing it to fully cover the subway line and the subway security areas on both sides, avoiding missing any possible monitoring blind spots; in the planning process of the drone's inspection route, the drone's flight altitude is set between 70m and 100m, so that the drone can ensure a suitable observation range and clear field of view while effectively avoiding collisions with obstacles during the execution of the inspection mission; the drone's flight speed is generally maintained between 5m / s and 10m / s, and the flight speed can be appropriately reduced in areas that require key observation to facilitate more detailed inspections;

[0056] S20, using the drone to collect inspection videos along the set route; in addition, during the process of collecting the inspection videos, the drone also needs to collect GPS data of the drone in real time, the GPS data including longitude, latitude, altitude and timestamp;

[0057] It should be noted that, in this embodiment, the drone is equipped with a high-definition camera. During the execution of the patrol mission, the drone collects patrol videos in a vertical overhead shooting manner. Moreover, according to the set flight altitude, the images captured by the patrol video often exceed the actual subway security area, so as to completely cover the subway security area on both sides of the track without leaving any blind spots, thereby ensuring a full-scale patrol of the subway security area. In addition, since the drone itself is equipped with a GPS module, during the process of collecting the patrol video, the drone will collect its own GPS information in real time through the GPS module. The GPS information includes longitude, latitude, altitude and timestamp. The collection of GPS information can record the flight trajectory and position of the drone in real time. Combined with the collected video data, it can accurately provide geographic location information for the video, which can help with the subsequent analysis and positioning of the video data.

[0058] S30. Determine the location of the subway security area in the inspection video and draw the boundary line of the subway security area. In this embodiment, since the captured inspection video footage exceeds the actual subway security area, it is necessary to exclude the area outside the subway security area before detecting risk construction equipment to prevent false detection. Therefore, in this embodiment, the range of the subway security area in the inspection video is determined by drawing the boundary line of the security area in the inspection video image to ensure the accuracy of subsequent risk construction equipment detection. Specifically, the specific method for determining the location of the subway security area in the inspection video and drawing the boundary line of the subway security area is as follows:

[0059] S301. Draw a virtual straight line perpendicular to the horizontal direction and passing through the center point of the image in the inspection video as the center line, where the center point of the image is the projection of the location of the drone.

[0060] S302. Determine the location of the boundary line of the subway security area in the inspection video according to the distance Pd between the boundary line of the subway security area and the image center line, and draw the boundary line of the subway security area in the inspection video. The calculation expression of Pd is:

[0061] ; In the formula, represents the actual distance between the boundary line of the subway security area and the center line of the subway track; represents the width of the inspection video image; represents the height of the inspection video image; represents the actual distance of the diagonal line in the inspection video image, The calculation expression of is:

[0062] ; In the formula, represents the relative height of the drone from the ground; represents the field of view angle of the drone, that is, the maximum angle range that the drone can shoot, The calculation expression of is:

[0063] , in the formula, represents the size of the image sensor of the camera equipped with the drone; represents the focal length of the lens;

[0064] S40. Use the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, and take screenshots of the patrol video based on the detected risk construction equipment. In this embodiment, the subway security area in the patrol video is analyzed through the target recognition model to identify the construction equipment that may pose a threat to the safety of the subway structure. Therefore, before using the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, it is necessary to first deploy the trained target recognition model to the drone. The specific training method of the target recognition model is as follows:

[0065] S401. Use the drone to collect construction equipment data in various construction scenarios at different time periods every day, including various aerial shooting heights and angles, and construct a dataset with an aerial view mainly composed of excavators, pile drivers, drilling machines, and cranes. Among them, excavators, pile drivers, drilling machines, and cranes are the key construction equipment that the subway security area pays the most attention to.

[0066] S402. Design a target recognition model based on the construction equipment in the subway security area by improving the target detection model yolov11. The specific improvement method of the target detection model yolov11 is as follows:

[0067] Enhance the data through geometric transformation and color space transformation at the input end. Specifically, the improvement at the input end increases the probability of mosaic hybrid splicing from 50% to 70%, and increases the appearance frequency of small targets. The rotation enhancement increases from a probability of 15% to 50% to ensure that the model can adapt to objects at different angles.

[0068] In the backbone network, replace the conventional convolution kernel in the C3K2 module with an adaptive rotation convolution kernel ARconv2d to extract target features in different directions in the image, so as to output feature maps in different directions. Specifically, during training, for each conventional convolution , a rotation angle will be predicted and generated . First, according to the predicted rotation angle , construct a rotation matrix . Then, through the affine network generation function operate on to generate a sampled grid after affine transformation . Then, according to the rotated network coordinates, use the network sampling function to sample the original convolution to obtain the rotation convolution kernel . The calculation process is as follows:

[0069] , ; Finally, a rotated convolutional kernel is used to convolve with the input feature map to obtain an output feature map;

[0070] In the detection head part, a rotation angle path is added to each detection head to extract features related to the rotation angle. Specifically, for the feature maps of different scales output by the backbone network, yolov11 is respectively equipped with detection heads of corresponding scales. Each detection head has a classification path and a regression path. On this basis, a rotation angle path is added to each detection head, specifically for extracting features related to the rotation angle;

[0071] It should be noted that in this embodiment, with reference to Figure 5 - Figure 6 shown, where Figure 5 is the path map of the rotation angle. In the figure, ARConv represents the combination of ARconv2d+bn+silu, where ARconv2d represents a convolutional layer with a rotated convolutional kernel, bn represents a batch normalization layer, and silu represents an activation function; conv2d is a conventional convolutional layer, and the number of channels is set to 1 here for outputting the predicted angle. These detection heads are usually composed of multiple convolutional layers and adopt a symmetric structure, that is, the detection head structures of each scale are the same. Since the drone is far from the shooting target, the detection head corresponding to the feature map P3 (sampled 8 times) for small targets needs to be focused on. To improve the feature extraction ability, this embodiment adopts Figure 6 the design of the asymmetric structure shown. For the detection head corresponding to the feature map P3, keep its classification path unchanged and make the following improvements to its regression path. Add a CA (Coordinate Attention) attention module between two Conv layers to enhance the small target detection ability. Specifically, the CA (Coordinate Attention) attention module first performs global pooling on the feature map p output by the first Conv in the regression path of the feature map P3 (the feature map p is the intermediate output result of the regression path of the feature map P3) in the horizontal and vertical directions to generate feature vectors and , and then splice the feature vectors in these two directions and perform a convolution operation on the spliced feature to generate an intermediate feature ; separate the intermediate feature into a horizontal direction feature and a vertical direction feature , then, for and , perform convolution operations respectively to generate attention weights and ; the generated attention weights , Multiply element-wise with the input feature map p to obtain the output feature map of the CA layer; in addition, to reduce the computational complexity introduced by the CA layer, a Ghost operation is used for lightweight improvement in the second Conv in the regression path of the feature map P3; first, a part of the feature map is generated through a conventional convolution, and then redundant feature maps are generated using inexpensive operations (linear transformations), and then these two types of feature maps are concatenated to keep the same number of output channels as the original and input to the subsequent network; for the detection heads of the feature maps P4 and P5, except for adding a rotation angle path, the other paths remain unchanged in their original structures;

[0072] S403, load the dataset through the target recognition model designed in S402 for multiple rounds of training, and perform performance evaluation on the target recognition model after each round of training is completed. Select the optimal model according to the evaluation results and deploy it on the drone; it should be noted that during the training process, to accelerate the convergence speed, a pre-trained model based on the public data of the drone can be used for initialization, and at the same time, hyperparameters such as the learning rate, batch size, and number of training rounds are set; in addition, the target recognition model reads data from the training set, calculates the loss value through forward propagation, and updates the model parameters using backpropagation, and after each training round, the performance of the model is evaluated using the validation set;

[0073] S50, obtain the GPS information of the screenshot, and generate a warning message by combining the screenshot, the GPS information of the screenshot, and the service data and send it to the staff for processing; specifically, the steps of obtaining the GPS information of the screenshot, and generating a warning message by combining the screenshot, the GPS information of the screenshot, and the service data and send it to the staff for processing specifically include the following steps:

[0074] Determine the longitude, latitude, and altitude of the drone in the screenshot from the GPS information data collected by the drone through the timestamp of the screenshot;

[0075] Design a data structure to store the screenshot, the GPS information of the screenshot, and the service data, and convert the data structure into JSON format to generate a warning message; where the service data includes data related to the identified risk construction equipment such as the device id and event description;

[0076] Report the warning message to the remote control platform of the drone through the network communication protocol, and the remote control platform of the drone pushes the warning message to the terminal device of the staff for processing.

[0077] It should be noted that in this embodiment, by packing the screenshot, the GPS information of the screenshot, and the service data together to generate a warning message and sending it to the staff, it can help the staff quickly confirm the device information and location information of the risk construction equipment and process it in a timely manner, providing a solid and reliable guarantee for the effective supervision of the subway security area.

[0078] Refer toFigure 2 As shown in the figure, the present invention also discloses a drone inspection intelligent early warning system 600 for subway security area construction, which is applied to a drone inspection intelligent early warning method for subway security area construction as described in the above embodiments. The drone inspection intelligent early warning system 600 for subway security area construction includes:

[0079] A flight mission planning module 601, configured to set the inspection flight path and flight parameters of the drone;

[0080] A video acquisition module 602, configured to collect inspection videos in real time along the flight path;

[0081] A subway security area identification and marking module 603, configured to determine the subway security area and draw the boundary line of the subway security area;

[0082] A risk construction equipment identification and detection module 604, configured to detect risk construction equipment in the subway security area in the inspection video by using a target recognition model;

[0083] A risk construction equipment extraction module 605, configured to take a screenshot of the location of the risk construction equipment in the inspection video and extract the GPS information of the screenshot;

[0084] An early warning module 606, configured to generate an early warning message by combining the screenshot, the GPS information of the screenshot, and service data and send it to the staff for processing.

[0085] It should be noted that the drone inspection intelligent early warning strategy for subway security area construction provided in the above embodiments is only illustrated by the division of the above functional modules. In actual applications, the above functions can be allocated to different functional modules as needed, that is, the internal structure of the drone inspection intelligent early warning system 600 for subway security area construction will be divided into different functional modules to complete all or part of the functions described above;

[0086] In addition, the drone inspection intelligent early warning system 600 for subway security area construction provided in the above embodiments and the embodiments of the drone inspection intelligent early warning method for subway security area construction belong to the same concept. The specific ways in which each module performs operations have been described in detail in the method embodiments and will not be repeated here.

[0087] Figure 3 The structural schematic diagram of an electronic device shown according to an exemplary embodiment.

[0088] It should be noted that this electronic device is only an example adapted to the present invention and cannot be considered as providing any limitation to the scope of use of the present invention. This electronic device cannot also be interpreted as requiring dependence on or necessarily having Figure 3One or more components in the illustrated exemplary electronic device 2000.

[0089] The hardware structure of the electronic device 2000 may vary significantly due to different configurations or performances. For example, Figure 3 As shown, the electronic device 2000 includes: a power supply 210, an interface 230, at least one memory 250, and at least one central processing unit (CPU) 270.

[0090] Specifically, the power supply 210 is used to provide operating voltage for each hardware device on the electronic device 2000.

[0091] The interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted to the present invention, the interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input / output interface 235, and at least one USB interface 237, etc. As Figure 3 shown, this is not a specific limitation here.

[0092] The memory 250, as a carrier for resource storage, can be a read-only memory, a random access memory, a magnetic disk, or an optical disc, etc. The resources stored thereon include an operating system 251, application programs 253, and data 255, etc. The storage method can be temporary storage or permanent storage.

[0093] Among them, the operating system 251 is used to manage and control each hardware device and application program 253 on the electronic device 2000 to enable the central processing unit 270 to perform operations and processing on the massive data 255 in the memory 250. It can be Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSD TM, etc.

[0094] The application program 253 is a computer-readable instruction that completes at least one specific task based on the operating system 251. It may include at least one module ( Figure 3 not shown), and each module can separately contain computer-readable instructions for the electronic device 2000. For example, the device for evaluating task priorities can be regarded as an application program 253 deployed on the electronic device 2000.

[0095] The data 255 can be signal information, etc., and is stored in the memory 250.

[0096] The central processing unit 270 may include one or more processors, and is configured to communicate with the memory 250 via at least one communication bus to read computer-readable instructions stored in the memory 250, so as to perform operations and processing on the massive data 255 in the memory 250. For example, the intelligent early warning method for drone inspection for subway security area construction is completed by reading a series of computer-readable instructions stored in the memory 250 through the central processing unit 270.

[0097] In addition, the present invention can also be implemented by hardware circuits or a combination of hardware circuits and software. Therefore, the implementation of the present invention is not limited to any specific hardware circuit, software, or the combination of both.

[0098] Please refer to Figure 4 , in an embodiment of the present invention, an electronic device 4000 is provided. The electronic device 4000 may include: a desktop computer, a laptop computer, a server, etc. with sensor recognition capabilities.

[0099] In Figure 4 , the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0100] Among them, the data interaction between the processor 4001 and the memory 4003 can be realized through at least one communication bus 4002. The communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 4 in

[0101] it is only represented by a thick line, but it does not mean that there is only one bus or one type of bus.

[0102] The processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present invention. The processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0103] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory), or other types of dynamic storage devices that can store information and instructions. It can also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store desired program instructions or code in the form of instruction or data structures and can be accessed by the electronic device 4000, but is not limited thereto.

[0104] Computer-readable instructions are stored on the memory 4003, and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002.

[0105] The computer-readable instructions are executed by one or more processors 4001 to implement the intelligent early warning method for drone patrol in the subway security area construction in the above embodiments.

[0106] In addition, an embodiment of the present invention provides a storage medium on which computer-readable instructions are stored, and the computer-readable instructions are executed by one or more processors to implement the intelligent early warning method for drone patrol in the subway security area construction as described above.

[0107] Compared with the prior art, the beneficial effects of the present invention are as follows: First, using a drone to collect inspection videos along the planned flight path according to the planned flight parameters can ensure that the perspective of the drone shooting the inspection video can completely cover the subway security areas on both sides of the subway track area and can meet the inspection accuracy. Then, by determining the position of the subway security area in the inspection video and drawing the boundary line of the subway security area, the subway security area can be accurately located in the inspection video. Next, using the target recognition model to detect the risk construction equipment in the subway security area within the boundary line and taking screenshots of the inspection video according to the detected risk construction equipment can improve the detection efficiency of the risk construction equipment. Finally, by obtaining the GPS information of the screenshots and generating early warning information by combining the screenshots, the GPS information of the screenshots and business data and sending it to the staff for processing, it can help the staff quickly obtain the position information of the risk construction equipment for timely processing. Therefore, the present invention can not only conduct a comprehensive inspection of the subway security area, but also improve the discovery probability of risk construction equipment, and can quickly obtain the position information of the risk construction equipment and send it to the staff in time, providing a solid and reliable guarantee for the effective supervision of the subway security area.

[0108] The above is a specific description of the preferred embodiment of the present invention. However, the present invention is not limited to the described embodiment. Those skilled in the art can make various equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations or substitutions are all included in the scope defined by the claims of this application.

Claims

1. An intelligent early warning method for drone inspection during subway security area construction, characterized in that, The drone inspection method for construction in the subway security area includes the following steps: S10. Plan the inspection flight path and flight parameters of the drone. Specifically, the specific method for planning the inspection flight path of the drone is as follows: Obtain an accurate map of the subway line to be inspected, covering the line's direction, station locations, and surrounding terrain; Determine the scope of the subway security area at different intervals of the subway line: Set the boundary line of the subway security area at L meters outside the center line of the subway track, where the value range of L is between 30 and 70; Taking the subway line to be inspected as a reference, set the flight path of the drone directly above the center line of the subway track; The flight parameters of the drone include flight altitude and flight speed. Among them, the flight altitude of the drone is set between 70m and 100m to ensure that the video captured by the drone can completely cover the subway security areas on both sides of the subway line; the flight speed of the drone is set between 5m / s and 10m / s to meet the inspection accuracy; S20. Use the drone to collect inspection videos along the set flight path. During the process of collecting inspection videos by the drone, the GPS data of the drone also needs to be collected in real time. The GPS data includes longitude, latitude, altitude, and timestamp; S30. Determine the location of the subway security area in the inspection video and draw the boundary line of the subway security area. Specifically, the specific method for determining the location of the subway security area in the inspection video and drawing the boundary line of the subway security area is as follows: S301. Draw a virtual straight line perpendicular to the horizontal direction and passing through the center point of the image as the center line in the inspection video, where the center point of the image is the projection of the location of the drone; S302. Determine the position of the boundary line of the subway security area in the inspection video according to the distance Pd between the boundary line of the subway security area and the image center line, and draw the boundary line of the subway security area in the inspection video; among them, the calculation expression of Pd is: ; wherein, represents the actual distance from the boundary line of the subway security area to the center line of the subway track; represents the width of the inspection video image; represents the height of the inspection video image; represents the actual distance of the diagonal line in the inspection video image, The calculation expression of ; where, represents the relative height of the UAV with respect to the ground; represents the field of view angle of the UAV, that is, the maximum angular range that the UAV can capture, The calculation expression of , where represents the image sensor size of the camera equipped on the UAV; represents the lens focal length; S40. Use the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, and take screenshots of the inspection video according to the detected risk construction equipment; S50. Obtain the GPS information of the screenshot, and generate a warning message by combining the screenshot, the GPS information of the screenshot, and business data and send it to the staff for processing.

2. The intelligent early warning method for drone inspection in the subway security area construction according to claim 1, characterized in that, In step S40, before using the target recognition model to detect the risk construction equipment in the subway security area within the boundary line, the trained target recognition model needs to be deployed on the drone first. Among them, the specific training method of the target recognition model is: S401. Use the drone to collect construction equipment data in various construction scenarios at different time periods every day, including various aerial shooting heights and angles, and construct a dataset with an aerial view mainly composed of excavators, pile drivers, drilling machines, and cranes; S402. Design a target recognition model based on the construction equipment in the subway security area by improving the target detection model yolov11; In S403, the designed target recognition model in S402 is used to load the data set for multiple rounds of training. After each round of training is completed, the performance of the target recognition model is evaluated, and the optimal model is selected according to the evaluation results and deployed on the drone.

3. The intelligent early warning method for UAV inspection in the construction of subway security areas according to claim 2, characterized in that, In step S402, the specific improvement method of the target detection model yolov11 is as follows: data is enhanced through geometric transformation and color space transformation at the input end; in the backbone network, the conventional convolution kernel in the C3K2 module is replaced by the adaptive rotation convolution kernel ARconv2d to extract target features in different directions in the image, so as to output feature maps in different directions; in the detection head part, a rotation angle path is added to each detection head to extract features related to the rotation angle.

4. The intelligent early warning method for UAV inspection in the construction of subway security areas according to claim 3, characterized in that, In step S50, the specific steps of obtaining the GPS information of the screenshot, and generating a warning message by combining the screenshot, the GPS information of the screenshot, and the service data and sending it to the staff for processing are as follows: Determine the longitude, latitude, and altitude of the drone in the screenshot from the GPS information data collected by the drone through the timestamp of the screenshot; Design a data structure to store the screenshot, the GPS information of the screenshot, and the service data, and convert the data structure into JSON format to generate a warning message; Report the warning message to the remote control platform of the drone through the network communication protocol, and the remote control platform of the drone pushes the warning message to the terminal device of the staff for processing.

5. An intelligent early warning system for UAV inspection during subway security area construction, which applies an intelligent early warning method for UAV inspection during subway security area construction according to any one of claims 1-4, characterized in that, The intelligent warning system for drone patrol of subway security construction areas includes: A flight mission planning module for setting the patrol route and flight parameters of the drone; A video acquisition module for collecting patrol videos in real time along the route; A subway security area identification and marking module for determining the subway security area and drawing the boundary line of the subway security area; A risk construction equipment identification and detection module for detecting risk construction equipment in the subway security area in the patrol video by using the target recognition model; A risk construction equipment extraction module for taking screenshots of the positions of the risk construction equipment in the patrol video and extracting the GPS information of the screenshots; A warning module for generating a warning message by combining the screenshot, the GPS information of the screenshot, and the service data and sending it to the staff for processing.

Images