Method and system for detecting multi-target invasion of power transmission line tower

Abstract

The invention discloses a power transmission line tower multi-target intrusion detection method and system, and relates to the technical field of intrusion detection, and the method comprises the steps: preprocessing millimeter wave radar data, screening effective targets, employing a DBSCAN algorithm to cluster output results, and extracting millimeter wave radar targets; an extended Kalman filtering algorithm is used to improve a measurement error, and a YOLOv7 network is used to carry out target detection on an image acquired by the infrared camera; and optimizing the data association effect of the millimeter-wave radar and the infrared camera by using a target-level fusion mode and a fusion strategy. The method improves the data processing efficiency, reduces the interference of invalid data, reduces the false alarm rate of the system, effectively improves the adaptability of the system to different environmental conditions, and maintains the stable target monitoring performance through fusing the information of different sensors. Therefore, reliable guarantee is provided for safe operation of the power transmission line tower.

Description

Technical Field

[0001] The present invention relates to the technical field of intrusion detection, and in particular to a multi-target intrusion detection method and system for a transmission line tower. Background Art

[0002] As a key node in the power transmission network, transmission line towers are directly related to the operating efficiency and safety of the entire power grid. With the continuous expansion of the scale and increase in complexity of the power grid, the monitoring and maintenance of transmission line towers has become an important issue in the power industry. In the past few decades, the monitoring technology of transmission line towers has undergone a transformation from traditional manual inspections to semi-automatic and automated monitoring. The development of these technologies has greatly improved the operation and maintenance efficiency of transmission line towers, but also exposed some technical bottlenecks. Early transmission line tower monitoring mainly relied on manual inspections. Although this method is simple and direct, it is inefficient and difficult to achieve real-time monitoring of large-area line towers. Subsequently, camera-based video surveillance systems were introduced. Although the real-time monitoring has been improved to a certain extent, it is greatly affected by external factors such as light and weather, especially at night or in severe weather conditions, the monitoring effect is greatly reduced; in addition, the camera system has limited detection capabilities for small targets at a long distance, and it is difficult to meet the needs of intrusion detection of transmission line towers.

[0003] There are many shortcomings in the existing transmission line tower intrusion detection technology. The millimeter wave radar lacks an effective target screening mechanism in data processing, resulting in a large number of invalid signals being mistakenly identified as targets in practical applications, increasing the false alarm rate of the system, which not only wastes valuable operation and maintenance resources, but also may reduce the reliability of the monitoring system due to frequent false alarms. The performance of infrared imaging technology is unstable under complex meteorological conditions, especially at night or in low visibility environments. The detection capability of infrared cameras is limited and cannot effectively identify and track targets, which limits its application in all-weather monitoring. The existing monitoring system has obvious deficiencies in data fusion. The respective advantages of millimeter wave radar and infrared cameras have not been effectively utilized. There is a lack of an efficient data fusion strategy to optimize the detection results of the two, resulting in the monitoring system being difficult to accurately identify and warn of multi-target intrusion events in practical applications. In short, the existing transmission line tower intrusion detection technology has many limitations in practical applications and cannot meet the power system's needs for high-precision and high-reliability monitoring. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: the existing intrusion detection technology has the problems of low millimeter-wave radar data processing efficiency, poor infrared camera target detection stability, poor millimeter-wave radar and infrared camera data fusion effect, and how to achieve all-weather, high-precision, and high-reliability multi-target intrusion detection.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for detecting multi-target intrusion of transmission line towers, comprising preprocessing millimeter-wave radar data, screening effective targets, clustering the output results using the DBSCAN algorithm, and extracting millimeter-wave radar targets; using an extended Kalman filter algorithm to improve measurement errors, and using a YOLOv7 network to perform target detection on images acquired by an infrared camera; using a target-level fusion method and fusion strategy to optimize the data association effect between the millimeter-wave radar and the infrared camera.

[0007] As a preferred solution of the multi-target intrusion detection method for transmission line towers of the present invention, the screening of effective targets includes judging the effectiveness of the targets according to the target history information output by the millimeter wave radar, and the conditions for judging the effective targets are expressed as follows:

[0008] {d i (n+1)-d i (n)>d t

[0009] {θ i (n+1)-θ i (n)}>θ t

[0010] {v i (n+1)v i (n)>v t

[0011] Where n is the sampling period number, n∈(1,2,3,…), d i is the relative distance of the i-th target point, θ i is the relative angle of the i-th target point, v i are the relative speed of the i-th target point, d t is the distance threshold within the sampling period interval, θ t is the relative angle threshold within the sampling period, v t is the threshold of relative speed change within the sampling period.

[0012] When the conditions for distinguishing a valid target are met, it is regarded as an invalid target and is removed from the detection targets of the millimeter-wave radar.

[0013] As a preferred solution of the multi-target intrusion detection method for transmission line towers described in the present invention, the extraction of millimeter-wave radar targets includes setting a radius threshold R and a minimum number of points minPts in the neighborhood, using a millimeter-wave radar to obtain radar scattering cross-sectional area information of the target as an additional quantity to supplement the data for clustering calculation, clustering the millimeter-wave radar targets within a distance less than or equal to a set range based on clustering accuracy, clustering the target point sets within a selected distance range, and improving the clustering accuracy by adaptively adjusting the radius threshold R.

[0014] As a preferred solution of the multi-target intrusion detection method for transmission line towers described in the present invention, the use of an extended Kalman filter algorithm to improve measurement errors includes modeling the target using a constant acceleration model based on the actual movement of the millimeter-wave radar target, the millimeter-wave radar outputs the relative distance, relative azimuth and relative speed of the measurement target, and uses an extended Kalman gain matrix to improve the error.

[0015] As a preferred solution of the multi-target intrusion detection method for transmission line towers described in the present invention, the method of performing target detection on images acquired by an infrared camera using a YOLOv7 network includes introducing an improved YOLOv7 network for training and performing target detection on images acquired by the infrared camera.

[0016] As a preferred solution of the multi-target intrusion detection method for transmission line towers described in the present invention, the optimization of the data association effect of the millimeter-wave radar and the infrared camera includes locating the center point of the rectangular tracking frame of the infrared image target as the starting point, starting from the starting point, finding the millimeter-wave radar target projection point matching the starting point, matching the infrared image detection frame to multiple millimeter-wave radar target projection points, and selecting the target point with the closest Euclidean distance to the sensor for matching.

[0017] If the sensor's target matching is successful, the position and speed information of the millimeter-wave radar and the category information of the infrared camera are extracted and combined and output as fused target information.

[0018] As a preferred solution of the multi-target intrusion detection method for transmission line towers described in the present invention, the data association effect of optimizing the millimeter-wave radar and the infrared camera also includes that when the infrared image algorithm successfully tracks the target and outputs the target information, and the millimeter-wave radar algorithm fails to track the target, the center point of the infrared image target frame is used as the representation target, and the representation target is converted to the millimeter-wave radar coordinate system to obtain the target's position information, and the target's speed information is calculated with the help of continuous infrared video frame detection results.

[0019] If the infrared image cannot detect the target, but the millimeter-wave radar outputs the target and there is a matching history with the infrared image target, the target features of the millimeter-wave radar are output as the fusion result. The target features are fused with the infrared image category features of the historical frame, which is called the radar tracking target.

[0020] If there is no matching history with the infrared image target, it means that it cannot be matched to the currently known category, which is called an unknown target. The position and speed information are provided by the tracking results of the millimeter wave radar, and the category of the target is unknown.

[0021] Another object of the present invention is to provide a multi-target intrusion detection system for transmission line towers, which can improve measurement errors by using an extended Kalman filter algorithm and perform target detection on images acquired by an infrared camera using a YOLOv7 network, thereby solving the problems of measurement error accumulation and insufficient detection stability in current infrared image target detection technology.

[0022] As a preferred solution of the transmission line tower multi-target intrusion detection system described in the present invention, it includes: a data screening and processing module, a target detection module, and a fusion strategy optimization module.

[0023] The data screening and processing module is used to preprocess the millimeter-wave radar data, screen effective targets, cluster the output results using the DBSCAN algorithm, and extract millimeter-wave radar targets; the target detection module is used to use the extended Kalman filter algorithm to improve measurement errors, and use the YOLOv7 network to perform target detection on images acquired by the infrared camera; the fusion strategy optimization module is used to use the target-level fusion method and fusion strategy to optimize the data association effect between the millimeter-wave radar and the infrared camera.

[0024] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a method for detecting multi-target intrusion of a transmission line tower.

[0025] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for detecting multi-target intrusion of a transmission line tower.

[0026] Beneficial effects of the present invention: The multi-target intrusion detection method for transmission line towers provided by the present invention preprocesses millimeter-wave radar data, screens effective targets, clusters the output results using the DBSCAN algorithm, extracts millimeter-wave radar targets, improves the accuracy and reliability of the detection system, reduces the false alarm rate, improves the target detection efficiency, and provides a data basis for subsequent target tracking and fusion. The extended Kalman filter algorithm is used to improve measurement errors, and the YOLOv7 network is used to perform target detection on images acquired by the infrared camera, which improves the stability and anti-interference ability of target detection, improves the accuracy and robustness of target detection, and reduces the influence of environmental factors on the detection results. The target-level fusion method and fusion strategy are used to optimize the data association effect of the millimeter-wave radar and the infrared camera, and the advantages of the millimeter-wave radar and the infrared camera are integrated to improve the comprehensiveness and accuracy of target recognition, effectively improve the system's detection capability of multi-target intrusion, reduce missed detection and false detection, and optimize the overall system performance. The present invention achieves better results in terms of target detection accuracy, system real-time and environmental adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0028] Figure 1 An overall flow chart of a multi-target intrusion detection method for a transmission line tower provided in the first embodiment of the present invention.

[0029] Figure 2 An overall flow chart of a multi-target intrusion detection system for a transmission line tower provided in a third embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0031] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a multi-target intrusion detection method for a transmission line tower, comprising:

[0032] S1: Preprocess the millimeter-wave radar data, screen valid targets, cluster the output results using the DBSCAN algorithm, and extract millimeter-wave radar targets.

[0033] Furthermore, screening effective targets includes judging the effectiveness of the targets based on the target history information output by the millimeter-wave radar. The conditions for judging effective targets are expressed as follows:

[0034] {d i (n+1)-d i (n)>d t

[0035] {θ i (n+1)-θ i (n)}>θ t

[0036] {v i (n+1)v i (n)>v t

[0037] Where n is the sampling period number, n∈(1,2,3,…), d i is the relative distance of the i-th target point, θ i is the relative angle of the i-th target point, v i are the relative speed of the i-th target point, d t is the distance threshold within the sampling period interval, θ t is the relative angle threshold within the sampling period, v t is the threshold of relative speed change within the sampling period.

[0038] When the conditions for distinguishing a valid target are met, it is regarded as an invalid target and is removed from the detection targets of the millimeter-wave radar.

[0039] It should be noted that the extraction of millimeter-wave radar targets includes setting a radius threshold R and the minimum number of points minPts in the neighborhood, using millimeter-wave radar to obtain the radar scattering cross-sectional area information of the target as an additional quantity to supplement the data for clustering calculation, and clustering the millimeter-wave radar targets with a distance less than or equal to the set range based on clustering accuracy, clustering the target point sets within the selected distance range, and improving the clustering accuracy by adaptively adjusting the radius threshold R.

[0040] It should also be noted that the signal-to-noise ratio of the data is improved through preprocessing steps such as denoising and signal enhancement, thereby providing a cleaner and more reliable data source for subsequent target identification. The application of the DBSCAN algorithm does not require the pre-setting of the number of clusters, and can adaptively identify valid targets in radar signals and effectively distinguish targets from background noise, greatly reducing the false detection rate. Through cluster analysis, the present invention can identify neighboring targets and avoid the problem of target merging or missed detection, which is particularly important for accurate target identification in transmission line tower monitoring, and lays a solid foundation for subsequent target tracking and fusion analysis, ensuring the accuracy and efficiency of the entire monitoring system in the initial data processing stage.

[0041] S2: Use the extended Kalman filter algorithm to improve the measurement error and use the YOLOv7 network to detect targets in images acquired by the infrared camera.

[0042] Furthermore, the use of the extended Kalman filter algorithm to improve the measurement error includes modeling the target using a constant acceleration model according to the actual movement of the millimeter-wave radar target, the millimeter-wave radar outputs the relative distance, relative azimuth and relative speed of the measured target, and the use of the extended Kalman gain matrix to improve the error.

[0043] It should be noted that using the YOLOv7 network to perform target detection on images acquired by the infrared camera includes introducing an improved YOLOv7 network for training and performing target detection on images acquired by the infrared camera.

[0044] It should also be noted that by extending the Kalman filter algorithm, it is possible to continuously estimate the state of targets in infrared images, effectively overcoming the measurement errors caused by sensor noise and environmental interference, and improving the accuracy and stability of target tracking. Through the recursive nature of the algorithm, the target state can be updated in real time in a dynamically changing environment. The use of the YOLOv7 network, using an efficient end-to-end design, achieves fast target detection while maintaining a high detection accuracy. Especially in low light or bad weather conditions, the YOLOv7 network exhibits excellent robustness, which not only improves the speed of infrared image target detection, but also enhances the system's adaptability in complex environments, providing strong technical support for the safe monitoring of transmission line towers.

[0045] S3: Use target-level fusion methods and fusion strategies to optimize the data association effect between millimeter-wave radar and infrared camera.

[0046] Furthermore, optimizing the data association effect of the millimeter-wave radar and the infrared camera includes locating the center point of the rectangular tracking frame of the infrared image target as the starting point, starting from the starting point, finding the millimeter-wave radar target projection point that matches the starting point, matching the infrared image detection frame to multiple millimeter-wave radar target projection points, and selecting the target point with the closest Euclidean distance to the sensor for matching.

[0047] If the sensor's target matching is successful, the position and speed information of the millimeter-wave radar and the category information of the infrared camera are extracted and combined and output as fused target information.

[0048] It should be noted that the effect of optimizing the data association between the millimeter-wave radar and the infrared camera also includes that when the infrared image algorithm successfully tracks the target and outputs the target information, and the millimeter-wave radar algorithm fails to track the target, the center point of the infrared image target frame is used as the representation target, and the representation target is converted to the millimeter-wave radar coordinate system to obtain the target's position information, and the target's speed information is calculated with the help of continuous infrared video frame detection results.

[0049] If the infrared image cannot detect the target, but the millimeter-wave radar outputs the target and there is a matching history with the infrared image target, the target features of the millimeter-wave radar are output as the fusion result. The target features are fused with the infrared image category features of the historical frame, which is called the radar tracking target.

[0050] If there is no matching history with the infrared image target, it means that it cannot be matched to the currently known category, which is called an unknown target. The position and speed information are provided by the tracking results of the millimeter wave radar, and the category of the target is unknown.

[0051] It should also be noted that the target-level fusion strategy effectively combines the respective advantages of the two sensors. The accuracy of millimeter-wave radar in distance and speed measurement and the clarity of infrared camera in target imaging complement each other, thereby achieving a more comprehensive and accurate description of target information. By optimizing the data association strategy, it can more effectively match and fuse target data from different sensors, reduce ambiguity and uncertainty in target tracking, and improve the continuity and accuracy of target tracking. The optimization of the fusion strategy improves the system's processing capability in multi-target scenarios, and can maintain good tracking performance even in the case of target intersection or occlusion, effectively improving the system's adaptability to different environmental conditions. Whether it is at night or in bad weather, it can maintain stable target monitoring performance by fusing information from different sensors, thereby providing reliable protection for the safe operation of transmission line towers.

[0052] Example 2 is an embodiment of the present invention, which provides a multi-target intrusion detection method for transmission line towers. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0053] First, the data collected by the millimeter-wave radar is preprocessed, which includes signal denoising, filtering and enhancement. Then, the DBSCAN algorithm is used to perform cluster analysis on the preprocessed data. The radius threshold R is set to 50 meters and the minimum number of points in the neighborhood minPts is set to 3. Through the algorithm, a total of 20 valid targets are screened out and 5 invalid targets are removed. The extended Kalman filter algorithm is used to estimate the state of the millimeter-wave radar target to improve the measurement error. At the same time, the improved YOLOv7 network is used to detect targets in images captured by the infrared camera. The network has been specially trained to adapt to the characteristics of infrared images. At this stage, a total of 18 targets are detected, and the matching rate with the millimeter-wave radar data reaches 90%. The target-level fusion method is adopted to optimize the millimeter-wave radar and infrared camera. The data association effect is improved. Through the matching algorithm, 17 targets are successfully associated. Among them, 2 targets cannot be associated due to the difference in sensor data. Through the fusion strategy, the position and speed information of the targets are successfully determined. A target-level fusion strategy is implemented to optimize the data association effect of the millimeter-wave radar and the infrared camera. Taking the center point of the rectangular tracking frame of the infrared image target as the starting point, the millimeter-wave radar target projection point matching it is found. By comparing the Euclidean distance of the sensor, the nearest target point is selected for matching. 17 targets are successfully associated. Among them, 2 targets cannot be associated due to the difference in sensor data. Through the fusion strategy, the position and speed information of these targets are determined. It can be obtained from the experimental results that the present invention has advantages and creativity in improving detection accuracy, reducing false alarm rate and enhancing system environmental adaptability.

[0054] Example 3, reference Figure 2 , as an embodiment of the present invention, provides a transmission line tower multi-target intrusion detection system, including a data screening and processing module, a target detection module, and a fusion strategy optimization module.

[0055] The data screening and processing module is used to preprocess the millimeter-wave radar data, screen effective targets, cluster the output results using the DBSCAN algorithm, and extract millimeter-wave radar targets; the target detection module is used to use the extended Kalman filter algorithm to improve the measurement error, and use the YOLOv7 network to perform target detection on the images obtained by the infrared camera; the fusion strategy optimization module is used to use the target-level fusion method and fusion strategy to optimize the data association effect between the millimeter-wave radar and the infrared camera.

[0056] If the function 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 the 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, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0057] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0058] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0059] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0060] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for detecting multi-target intrusion of transmission line towers, characterized in that: include: Preprocess the millimeter-wave radar data, screen effective targets, cluster the output results using the DBSCAN algorithm, and extract millimeter-wave radar targets; The extended Kalman filter algorithm is used to improve the measurement error, and the YOLOv7 network is used to detect targets in images acquired by the infrared camera; Use target-level fusion methods and fusion strategies to optimize the data association effect between millimeter-wave radar and infrared camera.

2. The method for detecting multi-target intrusion of a transmission line tower according to claim 1, characterized in that: The screening of effective targets includes judging the effectiveness of the targets according to the target history information output by the millimeter wave radar. The conditions for judging the effective targets are expressed as follows: {d i (n+1)-d i (n)}>d t {i} i (n+1)-θ i (n)}>θ t {v i (n+1)v i (n)>v t Where n is the sampling period number, n∈(1, 2, 3, ...), d i is the relative distance of the i-th target point, θ i is the relative angle of the i-th target point, v i are the relative speed of the i-th target point, d t is the distance threshold within the sampling period interval, θ t is the relative angle threshold within the sampling period, v t is the threshold value of relative speed change within the sampling period interval; When the conditions for distinguishing a valid target are met, it is regarded as an invalid target and is removed from the detection targets of the millimeter-wave radar.

3. The method for detecting multi-target intrusion of a transmission line tower according to claim 2, characterized in that: The method for extracting millimeter-wave radar targets includes setting a radius threshold R and a minimum number of points minPts in a neighborhood, using a millimeter-wave radar to obtain radar scattering cross-sectional area information of the target as an additional amount for supplementing data for clustering calculation, clustering millimeter-wave radar targets within a set range with a distance less than or equal to the set range based on clustering accuracy, clustering target point sets within a selected distance range, and improving clustering accuracy by adaptively adjusting the radius threshold R.

4. The method for detecting multi-target intrusion of a transmission line tower according to claim 3, characterized in that: The use of the extended Kalman filter algorithm to improve the measurement error includes modeling the target using a constant acceleration model according to the actual movement of the millimeter wave radar target, the millimeter wave radar outputs the relative distance, relative azimuth and relative speed of the measured target, and the extended Kalman gain matrix is ​​used to improve the error.

5. The method for detecting multi-target intrusion of a transmission line tower according to claim 4, characterized in that: The method of performing target detection on images acquired by an infrared camera using a YOLOv7 network includes introducing an improved YOLOv7 network for training and performing target detection on images acquired by the infrared camera.

6. The method for detecting multi-target intrusion of a transmission line tower according to claim 5, characterized in that: The optimization of the data association effect of the millimeter-wave radar and the infrared camera includes locating the center point of the rectangular tracking frame of the infrared image target as the starting point, starting from the starting point, finding the millimeter-wave radar target projection point matching the starting point, matching the infrared image detection frame to multiple millimeter-wave radar target projection points, and selecting the target point with the closest Euclidean distance to the sensor for matching; If the sensor's target matching is successful, the position and speed information of the millimeter-wave radar and the category information of the infrared camera are extracted and combined and output as fused target information.

7. The method for detecting multi-target intrusion of a transmission line tower according to claim 6, characterized in that: The optimization of the data association effect of the millimeter-wave radar and the infrared camera also includes when the infrared image algorithm successfully tracks the target and outputs the target information, and when the millimeter-wave radar algorithm fails to track the target, the center point of the infrared image target frame is used as the representation target, and the representation target is converted into the millimeter-wave radar coordinate system to obtain the position information of the target, and the speed information of the target is calculated with the help of continuous infrared video frame detection results; If the infrared image cannot detect the target, but the millimeter-wave radar outputs the target and there is a matching history with the infrared image target, the target features of the millimeter-wave radar are output as the fusion result. The target features are fused with the infrared image category features of the historical frame, which is called the radar tracking target. If there is no matching history with the infrared image target, it means that it cannot be matched to the currently known category, which is called an unknown target. The position and speed information are provided by the tracking results of the millimeter-wave radar, and the category of the target is unknown.

8. A system using the multi-target intrusion detection method for transmission line towers as claimed in any one of claims 1 to 7, characterized in that: Including data screening and processing module, target detection module, and fusion strategy optimization module; The data screening and processing module is used to pre-process the millimeter-wave radar data, screen effective targets, cluster the output results using the DBSCAN algorithm, and extract millimeter-wave radar targets; The target detection module is used to improve the measurement error by using the extended Kalman filter algorithm and perform target detection on the image acquired by the infrared camera by using the YOLOv7 network; The fusion strategy optimization module is used to optimize the data association effect of the millimeter wave radar and the infrared camera by using the target level fusion method and fusion strategy.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the multi-target intrusion detection method for transmission line towers described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-target intrusion detection method for transmission line towers according to any one of claims 1 to 7 are implemented.

Images