Inspection system and inspection platform for overhead conductor

Abstract

The invention provides an inspection system and an inspection platform for an overhead conductor. The inspection system for the overhead conductor comprises an unmanned aerial vehicle, an image acquisition device, an edge processor and edge business processing equipment. Firstly, a panoramic image of an inspection area and image information of each inspection point location are collected through an image collection device, secondly, an edge processor determines an inspection path and first target image information according to the panoramic image and the inspection image information, and finally, an edge service processing device determines a shooting track strategy according to the first target image information. And analyzing defects and reasons in the second target image information, and sending the defects and reasons to the inspection terminal. According to the inspection system, the inspection image information is processed through the edge processor and the edge service processing equipment, and only the image defect information is sent to the inspection terminal, so that the data volume transmitted through a network is greatly reduced, the network transmission speed and efficiency are further improved, and meanwhile, the inspection reliability and efficiency are also improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to an inspection system and an inspection platform for overhead wires. Background Art

[0002] The use of drone inspections and AI-based image intelligent recognition technology to identify target defects has gradually replaced manual inspections and human judgment, becoming the main method of defect detection. However, this implementation method is generally that the enterprise-side intelligent recognition platform is remotely connected to the drone through the network. This implementation method is too dependent on network transmission, and the image data is generally large.

[0003] For example, in actual applications, drones need to inspect at least 40 towers simultaneously, and each tower needs to collect about 50 pictures, each picture is about 1 to 2M, which will generate dozens of GB of data. For each tower, if you want to perform real-time inspections and real-time processing, you need to transmit the collected dozens of GB of data through 4G or 5G network cards to the enterprise-side intelligent recognition platform for processing, which will inevitably lead to slow network transmission speeds. If you want to transmit video, the transmission data will be larger, which will lead to slower transmission speeds, and even the situation where the transmission cannot be transmitted. Especially in remote areas where the signal construction is not yet perfect, the network signal will often be interrupted at any time, which will lead to the interruption of image data transmission, causing the enterprise-side intelligent recognition platform to receive incomplete image data or even fail to receive image data.

[0004] With respect to the technical problems of low transmission efficiency and poor reliability of the above inspection method, no effective solution has been proposed so far. Summary of the invention

[0005] The main purpose of the present invention is to provide an inspection system and an inspection platform for overhead wires to solve the technical problems of low transmission efficiency and poor reliability of the inspection method in the prior art.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present invention, there is provided an inspection system for overhead wires, comprising: an unmanned aerial vehicle; an image acquisition device, the image acquisition device is mounted on the unmanned aerial vehicle, and the image acquisition device is used to acquire panoramic image information of a target inspection area and inspection image information of each inspection point in the target inspection area; an edge processor, the edge processor is mounted on the bottom of the unmanned aerial vehicle, and is electrically connected to the image acquisition device, determines an inspection path of the unmanned aerial vehicle according to the received panoramic image information, and sends the inspection path to the unmanned aerial vehicle, determines first target image information according to the received inspection image information, and sends the first target image information to an edge business processing device, and the first target image information is used to characterize the inspection image information. abnormal image information; an edge business processing device, the edge business processing device is electrically connected to the edge processor and the image acquisition device respectively, the edge business processing device is used to receive the first target image information, and determine the shooting trajectory strategy of the drone according to the first target image information; wherein the edge processor is also used to receive the second target image information acquired by the image acquisition device according to the shooting trajectory strategy, and send the second target image information to the edge business processing device, so that the edge business processing device determines the image defect information according to the second target image information, and sends the image defect information to the external inspection terminal, and the image defect information at least includes the defective image information existing in the second target image information and the reason for the existence of the defective image information.

[0007] Furthermore, the edge processor is provided with a lightweight processed and trained image recognition model, which is associated with the object to be inspected. The image recognition model is used to identify the object to be inspected that has been input and corresponds to the image recognition model to obtain the first target image information.

[0008] Furthermore, the edge business processing device is provided with a defect recognition model, each of the defect recognition models is mapped to each of the preset components, and the defect recognition model is used to perform defect recognition on the second target image information associated with the defect recognition model, and obtain image defect information after defect recognition, wherein the second target image information includes the preset components associated with the defect recognition model.

[0009] Furthermore, the image acquisition device is electrically connected to the edge processor, and the edge processor determines the shooting posture information of the image acquisition device according to the second target image information, and controls the image acquisition device to adjust its posture relative to the drone.

[0010] Furthermore, the edge processor determines the shooting posture information of the image acquisition device according to the second target image information, including: the edge processor determines whether there is image information of a preset component in the second target image information according to the second target image information; if so, obtaining the shooting posture information of the image acquisition device, and adjusting the posture of the image acquisition device according to the shooting posture information to obtain clear second target image information.

[0011] Furthermore, the edge processor determines the shooting posture information of the image acquisition device based on the second target image information, and also includes: the edge processor determines whether there is image information of a preset component in the second target image information based on the second target image information; if not, the current second target image information is sent to the edge business processing device.

[0012] Furthermore, the edge processor includes: a shell; a main edge image processing component, which is arranged in the shell, and the main edge image processing component interacts with the edge business processing device and the drone respectively; a backup edge image processing component, which is arranged in the shell, and the backup edge image processing component interacts with the edge business processing device and the drone respectively; a fault detection circuit, which is arranged in the shell, and the fault detection circuit interacts with the main edge image processing component and the edge business processing device respectively, and the fault detection circuit is used to output a fault signal; wherein the edge business processing device is also used to receive a fault signal and enable the backup edge image processing component based on the fault signal.

[0013] Furthermore, the shell is provided with a humidity sensor for sensing humidity, and the shell is also provided with a heat dissipation window and a shielding member. The humidity sensor and the shielding member are both electrically connected to the edge processor. The edge processor is used to control the shielding member to move to a shielding position for shielding the heat dissipation window when the received humidity value reaches a humidity threshold, and to control the shielding member to move to an avoidance position for exposing the heat dissipation window when the received humidity value is lower than or equal to the humidity threshold.

[0014] Furthermore, the inspection system has a humidity sensor, and the humidity sensor and the shielding component are respectively connected to the edge processor signal. The edge processor is used to receive the humidity signal sensed by the humidity sensor, and control the shielding component to switch between the shielding position and the avoidance position according to the humidity signal.

[0015] Furthermore, the inspection system also includes: an edge inspection mobile terminal, which interacts with the edge business processing device and the inspection terminal respectively, and the edge inspection mobile terminal is used to filter image defect information and send the filtered image defect information to the inspection terminal.

[0016] According to another aspect of the present invention, there is provided an inspection platform, the inspection platform comprising an inspection system and an inspection terminal, and the inspection system is the above-mentioned inspection system.

[0017] Applying the technical solution of the present invention, the inspection system for overhead wires includes a drone, an edge processor, and an edge business processing device. First, the panoramic image information of the target inspection area and the inspection image information of each inspection point are collected by an image acquisition device. Secondly, the edge processor determines the inspection path according to the panoramic image information, and determines the first target image information according to the inspection image information. Finally, the edge business processing device determines the shooting trajectory strategy according to the first target image information, analyzes the defects and causes in the second target image information, and sends it to the inspection terminal. The above-mentioned inspection system processes the inspection image information through the edge processor and the edge business processing device, and only sends the image defect information of the overhead wire to the inspection terminal, which greatly reduces the amount of data transmitted through the network, thereby improving the network transmission speed and efficiency, and also improves the inspection reliability and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. 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:

[0019] Figure 1 A schematic diagram showing a first embodiment of a patrol inspection system according to the present invention is shown;

[0020] Figure 2 A schematic diagram of the structure of an image acquisition device according to the present invention is shown;

[0021] Figure 3 A schematic diagram showing the structure of a first embodiment of an edge processor according to the present invention is shown;

[0022] Figure 4 A schematic diagram showing the structure of a second embodiment of an edge processor according to the present invention is shown;

[0023] Figure 5 A schematic structural diagram of a third embodiment of an edge processor according to the present invention is shown;

[0024] Figure 6 A schematic structural diagram of a second embodiment of the inspection system according to the present invention is shown.

[0025] The above drawings include the following reference numerals:

[0026] 1. Image acquisition device;

[0027] 2. Install the pallet rack;

[0028] 3. Telescopic rod freedom platform;

[0029] 4. Stabilize the pallet rack;

[0030] 5. Collector;

[0031] 11. Shielding parts;

[0032] 12. Motor;

[0033] 13. Driving rod;

[0034] 14. Heat dissipation window;

[0035] 15. Traction belt;

[0036] 16. Shell. DETAILED DESCRIPTION

[0037] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0038] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0039] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 terms used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein, for example. In addition, the terms "including" and "having" and any of their variations 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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0040] Now, exemplary embodiments according to the present application will be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments may be implemented in a variety of different forms and should not be construed as being limited to the embodiments described herein. It should be understood that these embodiments are provided to make the disclosure of the present application thorough and complete, and to fully convey the concepts of these exemplary embodiments to those of ordinary skill in the art. In the accompanying drawings, for the sake of clarity, the thickness of the layers and regions may be enlarged, and the same reference numerals are used to represent the same devices, and thus their descriptions will be omitted.

[0041] Combination Figures 1 to 6 As shown, according to a specific embodiment of the present application, a patrol inspection system for overhead wires is provided.

[0042] Specifically, the inspection system includes: a drone, an image acquisition device, an edge processor and an edge business processing device. The image acquisition device is used to collect panoramic image information of the target inspection area and inspection image information of each inspection point in the target inspection area; the edge processor is mounted on the bottom of the drone and is electrically connected to the image acquisition device, determines the inspection path of the drone according to the received panoramic image information, and sends the inspection path to the drone, so that the drone performs inspection according to the inspection path, and determines the first target image information according to the received inspection image information, and sends the first target image information to the edge business processing device, the first target image information is used to characterize abnormal image information in the inspection image information; the edge business processing device is electrically connected to the edge processor and the image acquisition device respectively, the edge business processing device is used to receive the first target image information, and determine the shooting trajectory strategy of the drone according to the first target image information; wherein the edge processor is also used to receive the second target image information collected by the image acquisition device according to the shooting trajectory strategy, and send the second target image information to the edge business processing device, so that the edge business processing device determines the image defect information according to the second target image information, and sends the image defect information to the external inspection terminal, the image defect information at least includes the defect image information existing in the second target image information and the reason for the existence of the defect image information.

[0043] Applying the technical solution of the present invention, the inspection system for overhead wires includes a drone, an edge processor, and an edge business processing device. First, the panoramic image information of the target inspection area and the inspection image information of each inspection point are collected by an image acquisition device. Secondly, the edge processor determines the inspection path according to the panoramic image information, and determines the first target image information according to the inspection image information. Finally, the edge business processing device determines the shooting trajectory strategy according to the first target image information, analyzes the defects and causes in the second target image information, and sends it to the inspection terminal. The above-mentioned inspection system processes the inspection image information through the edge processor and the edge business processing device, and only sends the image defect information of the overhead wire to the inspection terminal, which greatly reduces the amount of data transmitted through the network, thereby improving the network transmission speed and efficiency, and also improves the inspection reliability and efficiency.

[0044] Specifically, Figure 1 As shown, the inspection system includes: a drone, an image acquisition device, an edge processor, an edge business processing device and a router, and the edge processor is mounted on the drone.

[0045] like Figure 1 As shown, in this embodiment, there are multiple drones, and the number of drones can be determined according to the actual area and time to be inspected. Each drone is equipped with an image acquisition device and an edge processor. The panoramic image information collected by the drone is sent to the edge processor. After receiving the panoramic image information, the edge processor sends the inspection path to the drone, and the drone inspects according to the inspection path. At the same time, the image acquisition device shoots the overhead wires in the corresponding inspection path area and sends the inspection image to the edge processor. The edge processor identifies defects in the inspection image to obtain first target image information and sends the first target image information to the edge business processing device. The edge business processing device determines the shooting trajectory strategy of the drone according to the first target image information, and sends the shooting trajectory strategy to the drone and the image acquisition device, so that the drone assists the image acquisition device in performing inspection and shooting according to the shooting trajectory strategy to obtain second target image information. The image acquisition device sends the second target image information to the edge business processing device. The edge business processing device analyzes the cause of defects in the second target image information to obtain image defect information, and then sends the image defect information to the external inspection terminal.

[0046] It should be noted that the edge processor and the edge service processing device can interact with each other through wireless communication, and the edge service processing device interacts with the inspection terminal through a switching device such as a router.

[0047] As an embodiment, the edge processor is provided with an image recognition model that has been light-weighted and trained. The image recognition model is associated with the object to be inspected. The image recognition model is used to identify the object to be inspected that has been input and corresponds to the image recognition model, and obtain the first target image information. In view of the limited processing power of the edge processor, the image recognition model of this embodiment is a training model suitable for the edge processor obtained by training the set initial training model and light-weighting it, so as to be used for real-time recognition, analysis and processing of image data obtained during the inspection process, improve the inspection efficiency, accuracy and visualization, and help the inspection personnel to better complete the inspection task.

[0048] The working principle of the image recognition model: First, the input image data is preprocessed, including image enhancement, noise filtering, and threshold segmentation, so that it can be input into the model for processing. Next, a deep learning model such as a convolutional neural network (CNN) is used to extract features from the preprocessed image data to extract key features in the image, such as the shape, color, and position of the wire. After feature extraction, the model will input the extracted features into the fully connected layer and use the classifier to classify and identify the image. The classifier usually uses a softmax function to map the output feature vector to the probability distribution of each category. Finally, the model will output the image information with abnormalities, which is the first target image information.

[0049] It should be noted that the image recognition model in this embodiment is a business processing model suitable for edge processors and inspection areas to which the inspection paths belong after lightweight processing. In order to improve the recognition effect, different inspection areas have different corresponding image recognition models. For example, some inspection areas have small connectors, and small connectors are better recognized by the model. Based on this, an image recognition model suitable for different inspection areas is adopted for different inspection areas.

[0050] As an alternative implementation, the image recognition model analyzes the collected data information to obtain defect feature information used to represent the characteristics of defect hazards, and matches the obtained defect feature information with the set defect image features. If the match is successful, it indicates that the collected data information is defect image information, that is, the first target image information. Otherwise, it indicates that the collected data information is not defect image information.

[0051] As an embodiment, the edge business processing device is provided with a plurality of defect recognition models, each defect recognition model is in a mapping relationship with each preset component, and the defect recognition model is used to perform defect recognition on the second target image information associated with the defect recognition model, and obtain image defect information after defect recognition, wherein the second target image information includes the preset component associated with the defect recognition model. In view of the edge business processing device, since it does not need to be mounted on a drone like an edge processor, it means that it has a powerful computing and processing capability. Based on this, the defect recognition model does not need to be lightweight compared to the image recognition model in the edge processor. The defect recognition model is a model with high image recognition accuracy and is associated with the preset components. In other words, each defect recognition model can accurately identify the image containing the preset components associated with the defect recognition model, thereby automatically detecting problems and defects in the image information and improving the efficiency and accuracy of inspections.

[0052] In this embodiment, the preset components are generally components that frequently fail.

[0053] It should be noted that, in this embodiment, the edge service processing device is provided with a plurality of defect recognition models that have been trained and have high accuracy for identifying different types of components. The recognition accuracy of a defect recognition model for defect images of different specified types is limited. In order to improve the recognition accuracy, multiple defect recognition models each correspond to the defect images of the specified component types. The defect recognition model further processes the target graphic information, extracts features from the defect image, identifies possible defects of the wire, such as breakage, corrosion, looseness, etc., and determines the cause of the defect according to different defects, so as to achieve accurate identification and timely processing of wire defects and improve the safety and reliability of the inspection system.

[0054] As an embodiment, the image acquisition device is electrically connected to the edge processor, and the edge processor determines the shooting posture information of the image acquisition device according to the second target image information, and controls the image acquisition device to adjust the posture relative to the drone. The image acquisition device can transmit the captured image to the edge processor as soon as possible, and the shooting posture information of the image acquisition device can be adjusted in time after the image is processed, which greatly improves the transmission efficiency.

[0055] like Figure 2As shown, the image acquisition device in this embodiment includes: a collector 5, a mounting tray frame 2, a telescopic rod type freedom platform 3 and a stable tray frame 4. The collector 5 is a camera with an auto-focus lens, and the data information collected by the collector 5 can be image information or video information. The mounting tray frame 2 is installed at the bottom of the drone, and the mounting tray frame 2 can rotate relative to the drone. One end of the telescopic rod type freedom platform 3 is installed on the mounting tray frame 2, and the other end is installed on the stable tray frame 4. The collector 5 is installed on the stable tray frame 4.

[0056] As an embodiment, the telescopic rod type freedom platform 3 has six telescopic rods, and each telescopic rod is installed with a driving mechanism. In this embodiment, the driving mechanism is a linear cylinder, which is electrically connected to the edge processor. The telescopic rod is driven by the linear cylinder to move the telescopic rod type freedom platform 3. Each telescopic rod can work independently, and can control the moving range of the telescopic rod type freedom platform 3 in turn, thereby adjusting the shooting position of the collector 5.

[0057] As an embodiment, the mounting tray rack 2 is mounted on the bottom of the drone via a rotating cylinder, the rotating cylinder is electrically connected to the edge processor, and the mounting tray rack 2 is driven by the rotating cylinder to rotate relative to the drone.

[0058] As an embodiment, the edge processor determines the shooting posture information of the image acquisition device based on the second target image information, including: the edge processor determines whether there is image information of a preset component in the second target image information based on the second target image information; if so, the shooting posture information of the image acquisition device is obtained, and the posture of the image acquisition device is adjusted according to the shooting posture information to obtain clear second target image information.

[0059] As an embodiment, the edge processor determines the shooting posture information of the image acquisition device based on the second target image information, and also includes: the edge processor determines whether there is image information of a preset component in the second target image information based on the second target image information; if not, the current second target image information is sent to the edge business processing device.

[0060] It should be noted that the preset components are generally components that frequently fail. If there are components that frequently fail in the second target image information, the shooting posture information of the collector is obtained, and the image acquisition device 1 is controlled to adjust the shooting posture to capture clearer image information. If there are no components that frequently fail in the second target image information, the second target image information is sent to the edge business processing device in real time for further processing.

[0061] Specifically, in this embodiment, the shooting posture information of the image acquisition device includes rotation angle information and position information. The obtained shooting posture information enables the mounting tray rack 2 to rotate according to the rotation angle information, and the telescopic rod type freedom platform 3 is driven by the linear cylinder to move the telescopic rod type freedom platform 3 according to the position information, thereby driving the collector 5 to move. The collector 5 shoots the area to be photographed from different angles and orientations according to the shooting posture information to obtain a set number of high-definition and all-round image data, and sends it to the image recognition model to identify the image information. The edge processor performs a preliminary screening of the photographed pictures, and the pictures that meet the requirements are transmitted to the edge business processing equipment for the next step of processing. This method can greatly reduce the number of transmitted pictures and improve the transmission efficiency.

[0062] In a specific embodiment, one implementation method of the edge processor controlling the position and posture of the collector 5 can be: inputting the collected data information into the trained neural network model for controlling the collector 5, obtaining the shooting posture of the collector 5, and sending the obtained shooting posture to the drone, and the drone controls the collector 5 according to the current posture and control posture of the collector 5, so that the camera accurately captures the image information to be captured. In this embodiment, the image information captured in a specified time period is trained by using an existing deep learning target detection algorithm, and the above target detection algorithm includes FasterR-CNN, SSD, yolo and other algorithms, and the training goal is to determine the position and posture of the collector 5.

[0063] Furthermore, the edge processor includes: a shell, a main edge image processing component, a standby edge image processing component and a fault detection circuit, wherein the main edge image processing component, the standby edge image processing component and the fault detection circuit are all arranged in the shell, the main edge image processing component exchanges information with the edge business processing device and the drone respectively; the standby edge image processing component exchanges information with the edge business processing device and the drone respectively; the fault detection circuit exchanges information with the main edge image processing component and the edge business processing device respectively, and the fault detection circuit is used to output a fault signal; wherein the edge business processing device is also used to receive a fault signal and enable the standby edge image processing component based on the fault signal. The main edge image processing component and the standby edge image processing component are provided in the edge processor to ensure that the normal operation of the edge processor is not affected in the event of a fault in the main edge image processing component.

[0064] like Figure 3As shown, the edge processor also includes a one-way valve switch circuit and a start circuit, which are also arranged in the housing, and the start circuit interacts with the main edge image processing component and the start circuit is electrically connected to the one-way valve switch circuit, and the one-way valve switch circuit interacts with the standby edge image processing component and the fault detection circuit respectively. The two processing components, the main edge image processing component and the standby edge image processing component, are set in the edge processor to ensure that the normal operation of the edge processor is not affected when the main edge image processing component fails. When the fault detection circuit detects that the main edge image processing component fails, the fault detection circuit outputs a fault signal indicating the failure to the edge business processing device, and the edge business processing device sends a start signal to start the standby edge image processing component to the start circuit. The start circuit triggers the one-way valve switch circuit to disconnect the ground control center from the main edge image processing component, and connects the edge business processing device to the standby edge image processing component, so as to use the standby edge processing component to perform image recognition on the data information collected by the camera, and then sends the identified defect hidden danger image information to the ground control center through the edge business processing device and distributes it to the inspection terminal. Among them, the ground control center is a remote control.

[0065] As an embodiment, the shell is provided with a humidity sensor for sensing humidity, and the shell is provided with a heat dissipation window 14 and a shielding member 11. The humidity sensor and the shielding member are electrically connected to the edge processor. The edge processor is used to control the shielding member to move to a shielding position for shielding the heat dissipation window when the received humidity value reaches the humidity threshold, and to control the shielding member to move to a avoidance position for exposing the heat dissipation window when the received humidity value is lower than or equal to the humidity threshold. The design of the shielding member 11 in this embodiment can prevent the components in the edge processor from being damaged due to harsh environments, such as the invasion of rain or snow water when a drone performs a flight mission on rainy or snowy days, and the design of the shielding member 11 can prevent rain or snow water from entering the edge processor shell along the heat dissipation window 14 and damaging the edge processor components. The design of the heat dissipation window 14 can prevent the edge processor from processing a large amount of image data and causing heat, thereby increasing the heat dissipation effect of the edge processor.

[0066] Among them, a data port connected to the collector 5, the drone and the power supply is reserved on one side of the shell. In some embodiments, the inspection system also includes a data adapter, which is provided with at least a power supply connector for connecting to the drone power output port, a communication A connector for connecting to the communication port A, and a communication B connector for connecting to the communication port B. In this embodiment, the data adapter can expand the port on the drone into an accessory of a USB interface, and can be directly connected through a standard connection line on the market by using the data adapter.

[0067] like Figure 4, Figure 5 As shown, the edge processor also includes a driving rod 13, a motor 12 and a traction belt 15, wherein the shielding member 11 is a foldable waterproof cloth. The driving rod 13 is arranged along the length direction of the shell, and the driving rod 13 is rotatably connected to the shell 16, and both ends of the driving rod 13 extend outward from the shell 16; one end of the shielding member 11 is connected to the shell 16, and the other end of the shielding member 11 is connected to the driving rod 13 through the traction belt 15; the motor 12 is arranged in the shell 16, and the driving end of the motor 12 is connected to the driving rod 13 to drive the driving rod 13 to rotate relative to the shell 16. The motor 12 realizes the winding of the traction belt 15 on the driving rod 13 by forward and reverse rotation, thereby realizing the shielding member 11 to block and avoid the heat dissipation window 14.

[0068] As an embodiment, the inspection system is further provided with a humidity sensor, and the humidity sensor and the shielding member 11 are respectively connected to the edge processor signal, and the edge processor is used to receive the humidity signal sensed by the humidity sensor, and control the shielding member 11 to switch between the shielding position and the avoidance position according to the humidity signal. The shielding member 11 controls the shielding member 11 to switch between the shielding position and the avoidance position according to the signal of the humidity sensor, which can prevent rain or snow water from entering the edge processor housing and damaging the edge processor.

[0069] Specifically, when the humidity value received from the humidity sensor is greater than or equal to the set humidity threshold, the motor 12 is controlled to rotate forward, the traction belt 15 is completely wound around the driving rod 13, and the foldable waterproof cloth is unfolded to cover the heat dissipation window 14 to prevent rain or dust from falling into the heat dissipation window 14; when the humidity value received from the humidity sensor is less than the set humidity threshold, the motor 12 is controlled to reverse, so that part of the traction belt 15 is wound away from the driving rod to recycle the foldable waterproof cloth to avoid the heat dissipation window 14.

[0070] As an embodiment, the inspection system also includes: an edge inspection mobile terminal, which exchanges information with the edge service processing device and the inspection terminal respectively, and the edge inspection mobile terminal is used to filter image defect information and send the filtered image defect information to the inspection terminal. The edge service processing device sends the identified defect hidden danger image information to the edge inspection mobile terminal, so that the inspection personnel can select the target defect hidden danger image to be determined from the defect hidden danger image displayed by the edge inspection mobile terminal, further sort the defect hidden danger image information, further reduce the size of the defect hidden danger image information processed by the inspection terminal, and improve the efficiency of the inspection terminal in processing the defect hidden danger image information.

[0071] According to another specific embodiment of the present application, a patrol inspection platform is provided. The patrol inspection platform includes a patrol inspection system and a patrol inspection terminal. The patrol inspection system is the patrol inspection system in the above embodiment.

[0072] As an alternative implementation, Figure 6 As shown, the inspection platform includes an inspection system, an inspection terminal and an edge inspection mobile terminal. The edge business processing device exchanges information with the edge inspection mobile terminal through a router, and the edge inspection mobile terminal exchanges information with the inspection terminal. The edge inspection mobile terminal is set up so that the inspection personnel can select the target defect hidden danger image to be determined from the defect hidden danger image displayed by the edge inspection mobile terminal, and the edge inspection mobile terminal sends the selected target defect hidden danger image to the inspection terminal. That is, the edge inspection mobile terminal can further sort the defect hidden danger image information sent by the edge processor, further reduce the amount of information of the defect hidden danger image information processed by the inspection terminal, and improve the efficiency of the inspection terminal in processing the defect hidden danger image information.

[0073] From the above description, it can be seen that the above embodiments of the present invention achieve the following technical effects:

[0074] 1. The inspection system is specially equipped with an edge processor, which can directly and in real time detect and identify a large amount of data collected by the image acquisition device, and only send defective images that are detected and identified as having hidden defects to the edge business processing equipment.

[0075] 2. The inspection trajectory strategy is determined based on the defect image through the edge business processing equipment, and then the image defects and the causes of the image defects are determined based on the target image taken by the drone according to the inspection trajectory strategy. The defect hidden danger image information is further sorted by the edge inspection terminal and finally sent to the edge inspection terminal, which greatly reduces the data transmitted through the network, thereby improving the network transmission speed and efficiency, and also improving the inspection reliability and efficiency, bringing a good experience to users.

[0076] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0077] In addition to the above, it should be noted that "one embodiment", "another embodiment", "embodiment", etc. mentioned in this specification refer to the specific features, structures or characteristics described in conjunction with the embodiment included in at least one embodiment generally described in this application. The same expression appearing in multiple places in the specification does not necessarily refer to the same embodiment. Further, when describing a specific feature, structure or characteristic in conjunction with any embodiment, it is claimed that the realization of such feature, structure or characteristic in conjunction with other embodiments also falls within the scope of the present invention.

[0078] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A patrol inspection system for overhead wires, characterized in that: include: Drones; An image acquisition device, which is mounted on the drone and is used to acquire panoramic image information of a target inspection area and inspection image information of each inspection point in the target inspection area; an edge processor, the edge processor being mounted on the bottom of the drone and being electrically connected to the image acquisition device, and being used to determine an inspection path of the drone according to the received panoramic image information, and send the inspection path to the drone, determine first target image information according to the received inspection image information, and send the first target image information to an edge service processing device, wherein the first target image information is used to characterize abnormal image information in the inspection image information; An edge service processing device, the edge service processing device is electrically connected to the edge processor and the image acquisition device respectively, and the edge service processing device is used to receive the first target image information and determine the shooting trajectory strategy of the drone according to the first target image information; Among them, the edge processor is also used to receive the second target image information collected by the image acquisition device according to the shooting trajectory strategy, and send the second target image information to the edge business processing device, so that the edge business processing device determines the image defect information according to the second target image information, and sends the image defect information to the external inspection terminal, and the image defect information at least includes the defective image information existing in the second target image information and the reason for the existence of the defective image information.

2. The inspection system according to claim 1, characterized in that: The edge processor is provided with a lightweight and trained image recognition model, the image recognition model is associated with the object to be inspected, and the image recognition model is used to identify the input object to be inspected corresponding to the image recognition model to obtain the first target image information.

3. The inspection system according to claim 1, characterized in that: The edge business processing device is provided with a plurality of defect recognition models, each of which is mapped to each preset component. The defect recognition model is used to perform defect recognition on the second target image information associated with the defect recognition model, and obtain image defect information after defect recognition, wherein the second target image information includes the preset component associated with the defect recognition model.

4. The inspection system according to claim 1, characterized in that: The image acquisition device is electrically connected to the edge processor. The edge processor determines the shooting posture information of the image acquisition device according to the second target image information, and controls the image acquisition device to adjust its posture relative to the drone.

5. The inspection system according to claim 4, characterized in that: The edge processor determines the shooting posture information of the image acquisition device according to the second target image information, including: The edge processor determines whether there is image information of a preset component in the second target image information according to the second target image information; If it exists, the shooting posture information of the image acquisition device is acquired, and the posture of the image acquisition device is adjusted according to the shooting posture information to obtain the best second target image information.

6. The inspection system according to claim 4, characterized in that: The edge processor determines the shooting posture information of the image acquisition device according to the second target image information, and further includes: The edge processor determines whether there is image information of a preset component in the second target image information according to the second target image information; If not, the current second target image information is sent to the edge service processing device.

7. The inspection system according to claim 1, characterized in that: The edge processor comprises: case; A main edge image processing component, wherein the main edge image processing component is disposed in the housing and the main edge image processing component respectively exchanges information with the edge service processing device and the drone; A standby edge image processing component, the standby edge image processing component is arranged in the housing, and the standby edge image processing component respectively exchanges information with the edge service processing device and the drone; A fault detection circuit, wherein the fault detection circuit is disposed in the housing, the fault detection circuit respectively interacts with the primary edge image processing component and the edge service processing device, and the fault detection circuit is used to output a fault signal; The edge service processing device is further configured to receive the fault signal and enable the standby edge image processing component based on the fault signal.

8. The inspection system according to claim 7, characterized in that: The shell is provided with a humidity sensor for sensing humidity, and the shell is also provided with a heat dissipation window and a shielding member. The humidity sensor and the shielding member are both electrically connected to the edge processor. The edge processor is used to control the shielding member to move to a shielding position for shielding the heat dissipation window when the received humidity value reaches a humidity threshold, and to control the shielding member to move to an avoidance position for exposing the heat dissipation window when the received humidity value is lower than or equal to the humidity threshold.

9. The inspection system according to claim 1, characterized in that: The inspection system also includes: An edge inspection mobile terminal, wherein the edge inspection mobile terminal exchanges information with the edge service processing device and the inspection terminal respectively, and the edge inspection mobile terminal is used to filter the image defect information and send the filtered image defect information to the inspection terminal.

10. An inspection platform, characterized in that: The inspection platform includes an inspection system and an inspection terminal, and the inspection system is the inspection system according to any one of claims 1-9.

Images