A transmission line key area visual monitoring system

Abstract

The application provides a power transmission line key area visual monitoring system, comprising a UAV acquisition module and a management and control module; the UAV acquisition module is used for matching the regional type information according to the current positioning information, and obtaining the power transmission line monitoring data corresponding to the regional type information from the current inspection position; the management and control module is used for extracting the corresponding positioning information according to the obtained power transmission line monitoring data, matching the corresponding regional type information according to the positioning information, calling the monitoring standard corresponding to the regional type, performing abnormal analysis on the power transmission line monitoring data according to the obtained monitoring standard, obtaining the abnormal analysis result, and integrating the corresponding power transmission line monitoring data and the abnormal analysis result into a map model for visual display. The application helps to improve the pertinence and adaptability of the power transmission line abnormal analysis.

Description

Technical Field

[0001] This invention relates to the field of power transmission line monitoring technology, and in particular to a visual monitoring system for key areas of power transmission lines. Background Technology

[0002] Currently, overhead transmission lines are an important component of power grid construction. The construction of transmission line networks is characterized by strong coverage areas. Therefore, the construction of overhead transmission lines inevitably needs to cross large areas, which increases the difficulty of monitoring overhead transmission lines.

[0003] In the existing technology, some methods have emerged that use drones to conduct aerial inspections along the transmission line's route. For example, patent CN115046532A discloses a method for detecting transmission line faults based on drones. This method uses drones to acquire video images and then uses these images to monitor for faults or anomalies in the transmission line. However, traditional video-based transmission line monitoring methods typically use fixed methods (such as for monitoring icing conditions on the line), which result in limited functionality and an inability to perform targeted analysis based on the specific characteristics of the area where the transmission line is located. This can easily lead to wasted resources or insufficient monitoring results. Summary of the Invention

[0004] To address the aforementioned problems, this invention aims to provide a visual monitoring system for key areas of power transmission lines.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] This invention proposes a visual monitoring system for key areas of power transmission lines, characterized by comprising a drone data acquisition module and a control module; wherein,

[0007] The drone data acquisition module is used to monitor power transmission lines along a preset inspection route. During the inspection, it matches the area type information with the current location information, obtains the power transmission line monitoring data corresponding to the area type information from the current inspection location, and transmits the obtained power transmission line monitoring data back to the control module. The power transmission line monitoring data carries the corresponding location information.

[0008] The control module is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding area type information based on the location information and retrieve the monitoring standards corresponding to the area type, perform anomaly analysis on the transmission line monitoring data based on the acquired monitoring standards, and obtain the anomaly analysis results; and to integrate the corresponding transmission line monitoring data and anomaly analysis results into the map model for visualization based on the location information of the transmission line monitoring data.

[0009] Preferably, the human-machine data acquisition module includes a positioning unit, a control unit, a monitoring unit, and a communication unit;

[0010] The positioning unit is used to acquire real-time positioning information from the UAV acquisition module;

[0011] The control unit is used to control the UAV data acquisition module to complete the inspection along the preset inspection route. During the inspection, it matches the area type information according to the current positioning information and collects the corresponding power transmission line monitoring data from the control test unit according to the area type information. The area types include forest areas, residential areas, heavily polluted areas, lightning areas, bird activity areas, areas damaged by external forces, flood areas, etc.

[0012] The monitoring unit is used to collect transmission line monitoring data according to the instructions of the control unit, including video monitoring data pointed at different devices or angles;

[0013] The communication unit is used to transmit the acquired transmission line monitoring data to the control module.

[0014] Preferably, the monitoring unit includes an image acquisition unit, a sensor unit, and a communication unit;

[0015] The image acquisition unit includes a camera unit with a pan-tilt head, wherein the camera's shooting angle and the pan-tilt head's angle are controlled by a control unit; the camera unit is used to capture video monitoring data at a specified angle.

[0016] The sensor unit is equipped with different types of sensors, and corresponding sensor monitoring data is acquired through each type of sensor. The sensor monitoring data includes temperature data, humidity data, etc.

[0017] Preferably, the control module includes an extraction unit, an anomaly analysis unit, and a visualization unit;

[0018] The extraction unit is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding area type information based on the location information, record the corresponding area type identifier, and match the corresponding monitoring standard based on the area type identifier.

[0019] The anomaly analysis unit is used to perform anomaly analysis based on the acquired monitoring data, analyze whether the monitoring data meets the corresponding monitoring standards, and obtain the anomaly analysis results.

[0020] The visualization unit is used to integrate transmission line monitoring data and corresponding anomaly analysis results into the visualization map model, and display them at the corresponding locations in the visualization map model based on the location information of the monitoring data.

[0021] Preferably, the anomaly analysis unit includes a video analysis unit and an environmental analysis unit;

[0022] The video analysis unit is used to retrieve the corresponding video analysis model to perform anomaly analysis on the video monitoring data based on the area type to which the acquired transmission line video monitoring data belongs, and obtain the video anomaly analysis results. The video analysis models include fire identification models, floating object identification models, bird damage identification models, corrosion identification models, and covering identification models based on video data.

[0023] The environmental analysis unit is used to compare and analyze the acquired sensor monitoring data with the environmental standards of the corresponding area type to obtain environmental anomaly analysis results.

[0024] Preferably, the video analysis unit includes a preprocessing unit and an anomaly analysis unit;

[0025] The preprocessing unit is used to preprocess the acquired video monitoring data to improve the clarity of the video monitoring data;

[0026] The anomaly analysis unit is used to retrieve the corresponding video analysis model to perform anomaly analysis on the video monitoring data based on the area type to which the acquired transmission line video monitoring data belongs, and obtain the video anomaly analysis results.

[0027] Preferably, it also includes a backend management module;

[0028] The backend management module is used to manage the area type classification information based on location information, so that different location information corresponds to one or more area types;

[0029] It also manages the corresponding monitoring standards for each type of region, including setting corresponding monitoring standards for different types of regions and specific judgment indicators for each monitoring standard; and associates the pre-stored data processing model with the corresponding monitoring standard so that it can be called when the acquired transmission line monitoring data is used for anomaly analysis using the monitoring standard.

[0030] The beneficial effects of this invention are as follows: When inspecting power transmission lines using a drone data acquisition module, this invention can collect relevant monitoring data based on the drone's positioning, achieving differentiated monitoring of different key areas of the power transmission line. This allows monitoring of different key areas to adapt to the characteristics of those areas, improving the adaptability and targeting of power transmission line inspection data collection. Furthermore, the monitoring data collected from different key areas can be analyzed using corresponding analysis methods and standards to obtain targeted anomaly analysis results for those key areas, enhancing the targeting and adaptability of power transmission line anomaly analysis. Moreover, the anomaly analysis results are visualized based on a map model, improving the intuitive display of overall and local monitoring results for the power transmission line, and assisting managers in further operational and maintenance scheduling management of anomalies. Attached Figure Description

[0031] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0032] Figure 1 This is a framework diagram of a visualization monitoring system for key areas of a power transmission line, as shown in an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram of the module setup of a visualization monitoring system for key areas of a power transmission line, as shown in an embodiment of the present invention. Detailed Implementation

[0034] The present invention will be further described in conjunction with the following application scenarios.

[0035] See Figure 1 The embodiment illustrates a visualization monitoring system for key areas of power transmission lines, comprising a drone data acquisition module and a control module;

[0036] The drone data acquisition module is used to monitor power transmission lines along a preset inspection route. During the inspection, it matches the area type information with the current location information, obtains the power transmission line monitoring data corresponding to the area type information from the current inspection location, and transmits the obtained power transmission line monitoring data back to the control module. The power transmission line monitoring data carries the corresponding location information.

[0037] The control module is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding area type information based on the location information and retrieve the monitoring standards corresponding to the area type, perform anomaly analysis on the transmission line monitoring data based on the acquired monitoring standards, and obtain the anomaly analysis results; and to integrate the corresponding transmission line monitoring data and anomaly analysis results into the map model for visualization based on the location information of the transmission line monitoring data.

[0038] In the above embodiments, when inspecting transmission lines using a drone data acquisition module, the corresponding monitoring data can be collected based on the drone's positioning. This allows for differentiated monitoring of different key areas of the transmission line, ensuring that monitoring of different key areas is adapted to the characteristics of those areas and improving the adaptability and relevance of transmission line inspection data collection. Furthermore, the monitoring data collected from different key areas can be analyzed using appropriate analytical methods and standards to obtain targeted anomaly analysis results for those key areas, enhancing the relevance and adaptability of transmission line anomaly analysis. Moreover, the anomaly analysis results are further visualized using a map model, improving the intuitive presentation of overall and local monitoring results for the transmission line and assisting managers in further operational and maintenance scheduling management of anomalies.

[0039] The regional types include forest areas, residential areas, heavily polluted areas, minefields, bird activity areas, areas damaged by external forces, and flood-prone areas.

[0040] Different types of regions experience different forms of abnormal interference to transmission lines. Therefore, when monitoring key areas of different types, it is necessary to collect different monitoring data or obtain monitoring data through different methods so that the acquired monitoring data can be adapted to the characteristics of the region, which helps to improve the accuracy and relevance of monitoring key areas of transmission lines.

[0041] In one scenario, specifically in forested areas, the presence of numerous trees beneath power transmission lines makes them highly susceptible to damage from forest fires due to their rapid and widespread impact. Therefore, when drones patrol forested areas, the video surveillance range should be expanded to analyze the entire forest area for potential fires, enabling early detection of any abnormal fire conditions. In contrast, in open, flooded areas, fires have a smaller impact on power transmission lines. Instead, issues like riverbank collapses and land subsidence significantly affect the stability of the transmission lines. Therefore, when drones patrol flooded areas, video surveillance data should be used to monitor the surrounding land conditions around power transmission towers, analyzing for collapses or corrosion to prevent damage to the transmission lines due to unstable ground. Meanwhile, for highly polluted areas with concentrated industries, the strategy should be adjusted to monitor the appearance of the transmission lines, such as dirt and corrosion. In this case, the video monitoring data acquisition strategy should be adjusted to acquire close-up images of the transmission lines, so as to conduct anomaly analysis on the appearance of the transmission lines through video monitoring data and obtain targeted anomaly analysis results.

[0042] Preferred, see Figure 2 The UAV data acquisition module includes a positioning unit, a control unit, a monitoring unit, and a communication unit;

[0043] The positioning unit is used to acquire real-time positioning information from the UAV acquisition module;

[0044] The control unit is used to control the UAV data acquisition module to complete the inspection along the preset inspection route. During the inspection, it matches the area type information according to the current positioning information and collects the corresponding power transmission line monitoring data from the control test unit according to the area type information.

[0045] The monitoring unit is used to collect transmission line monitoring data according to the instructions of the control unit, including video monitoring data pointed at different devices or angles;

[0046] The communication unit is used to transmit the acquired transmission line monitoring data to the control module.

[0047] During flight inspections, the UAV data acquisition module can match the corresponding area type based on real-time location information and adopt appropriate strategies to complete the data collection. This enables targeted anomaly analysis of the acquired data in the corresponding area, improving the focus of information collection in key areas of power transmission lines. Furthermore, adjusting the monitoring strategy based on location information helps to enhance the intelligence level of data acquisition.

[0048] Preferably, the monitoring unit includes an image acquisition unit, a sensor unit, and a communication unit;

[0049] The image acquisition unit includes a camera unit with a pan-tilt head, wherein the shooting angle of the camera and the angle of the pan-tilt head are controlled by a control unit; the camera unit is used to capture video monitoring data at a specified angle;

[0050] The sensor unit is equipped with different types of sensors, and corresponding sensor monitoring data is acquired through each type of sensor. The sensor monitoring data includes temperature data, humidity data, etc.

[0051] The transmission line monitoring data also includes sensor monitoring data.

[0052] In addition to collecting video image data, it can also collect environmental monitoring data of the area through corresponding sensors, thus enriching the comprehensiveness of monitoring data collection.

[0053] Preferably, the control module includes an extraction unit, an anomaly analysis unit, and a visualization unit;

[0054] The extraction unit is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding area type information based on the location information, record the corresponding area type identifier, and match the corresponding monitoring standard based on the area type identifier.

[0055] The anomaly analysis unit is used to perform anomaly analysis based on the acquired monitoring data, analyze whether the monitoring data meets the corresponding monitoring standards, and obtain the anomaly analysis results.

[0056] The visualization unit is used to integrate transmission line monitoring data and corresponding anomaly analysis results into the visualization map model, and display them at the corresponding locations in the visualization map model based on the location information of the monitoring data.

[0057] Corresponding to the settings of the UAV data acquisition module, when the control module obtains the transmission line monitoring data transmitted back by the UAV data acquisition module, it obtains the area type of the area where the transmission line is located based on the location information of the transmission line monitoring data, and extracts the corresponding monitoring standards (including analysis indicators, analysis methods, and anomaly standards, etc.) based on the area type. The anomaly analysis model performs anomaly analysis on the specified monitoring indicators according to the monitoring standards and adopts the corresponding methods to obtain the corresponding anomaly analysis results. Based on the visualization map module, the acquired monitoring data and the corresponding anomaly analysis results are integrated into the map module for visualization display, thereby improving the level of intelligence in data processing.

[0058] The control module can be built on cloud servers, local servers, or local smart terminals to adapt to the needs of power transmission line monitoring data processing in different scenarios.

[0059] Preferably, the anomaly analysis unit includes a video analysis unit and an environmental analysis unit;

[0060] The video analysis unit is used to retrieve the corresponding video analysis model to perform anomaly analysis on the video monitoring data based on the area type to which the acquired transmission line video monitoring data belongs, and obtain the video anomaly analysis results. The video analysis models include fire identification models, floating object identification models, bird damage identification models, corrosion identification models, and covering identification models based on video data.

[0061] The environmental analysis unit is used to compare and analyze the acquired sensor monitoring data with the environmental standards of the corresponding area type to obtain environmental anomaly analysis results.

[0062] The video analysis unit uses image processing technology to analyze and process the acquired video monitoring data. Different analysis models are used for different area types to analyze the video monitoring data and obtain corresponding video anomaly analysis results. The environmental analysis unit further analyzes the acquired sensor monitoring data in conjunction with the indicators of the current area type to determine whether there are any abnormal environmental indicators in the area and obtains environmental anomaly analysis results.

[0063] In one scenario, for a specific area type, one or more video analysis models can be used to analyze the acquired video monitoring data and obtain corresponding video anomaly analysis results. For example, for all area types, abnormal coverings and bird damage can be identified based on the collected close-up monitoring images of power transmission lines, while for forest areas, fire identification can be further performed based on distant monitoring images of power transmission lines.

[0064] Preferably, the video analysis unit includes a preprocessing unit and an anomaly analysis unit;

[0065] The preprocessing unit is used to preprocess the acquired video monitoring data to improve the clarity of the video monitoring data;

[0066] The anomaly analysis unit is used to retrieve the corresponding video analysis model based on the region type to which the pre-transmission line video monitoring data belongs, and to perform anomaly analysis on the processed video monitoring data to obtain the video anomaly analysis results.

[0067] Considering that during drone patrols, fluctuations caused by the environment or the drone's own flight (such as turbulence and shaking) can easily lead to unclear transmission line monitoring data acquired by the drone's acquisition module, a dedicated preprocessing unit is included in the video analysis unit. This unit preprocesses the acquired video monitoring data to improve image clarity, which helps the subsequent anomaly analysis unit perform further anomaly analysis based on the preprocessed transmission line video monitoring data. This effectively avoids situations where image quality issues affect the anomaly analysis results.

[0068] Preferably, the preprocessing unit performs preprocessing based on the acquired video monitoring data, specifically including:

[0069] 1) Perform image frame processing on the acquired video monitoring data and extract the video monitoring image X from each frame;

[0070] 2) Detect foreground targets based on the acquired video monitoring frame X. When a foreground target is detected, mark the video monitoring frame X as a close-up image X∈X. N Otherwise, if no foreground target can be detected, the video monitoring frame is marked as a distant image X∈X. F ;

[0071] This includes foreground target detection based on the acquired video monitoring footage, specifically:

[0072] Edge detection is performed on the video monitoring image, edge pixels in the video monitoring image are extracted and marked, and the edge pixels in the image are counted into the edge pixel set K.

[0073] Based on the marked edge pixels, the edge density factor of each pixel in the image is calculated. The edge density factor calculation function used is as follows:

[0074]

[0075] Where, ρ K (x,y) represents the edge density factor of pixel (x,y), s n×n (x,y) represents the range of an n×n rectangular region centered at pixel (x,y), where (a,b)∈s. n×n (x,y) represents pixel point (a,b) in region s. n×n The pixel within (x,y) is used as a condition, indicating that the pixel (a,b) is marked as an edge pixel. This represents the number of pixels that meet the judgment condition (a,b)∈K, and n represents the side length of the preset rectangular range.

[0076] Further calculations are performed on the edge variation factors of each edge pixel, using the following edge variation phonetic calculation function:

[0077]

[0078] Where, ρ K (x,y) represents the edge variation factor of pixel (x,y), s m×m (x,y) represents the range of an m×m rectangular region centered at pixel (x,y), where (a,b)∈s. m×m (x,y) represents pixel point (a,b) in region s. m×m Pixels within (x,y) and Representing regions s respectively m×m The maximum and minimum values ​​of the edge density factor of each pixel within (x,y);

[0079] Calculate the grayscale variation factor for each edge pixel (x,y)∈K, where the grayscale variation factor calculation function is:

[0080]

[0081] Where hg(x,y) represents the grayscale variation factor of pixel (x,y), where (x,y)∈K, s m×m (x,y) represents the range of an m×m rectangular region centered at pixel (x,y), where (a,b)∈s. m×m (x,y) represents pixel point (a,b) in region s. m×m Pixels within (x, y), This indicates that pixel (a, b) is a non-edge pixel. and Representing regions s respectively m×m The maximum and minimum gray values ​​of non-edge pixels within (x,y);

[0082] Foreground features are determined for edge pixels based on the obtained edge density factor and grayscale variation factor. The foreground feature determination function used is as follows:

[0083]

[0084] Here, ture1(x,y) represents the foreground feature judgment function for edge pixel (x,y), where (x,y)∈K. When the corresponding conditions are met, pixel (x,y) is marked as an edge pixel that does not belong to the foreground feature. Or belong to the foreground feature edge pixel (x,y)∈acK, and represents AND / AND logical operation; pgT represents the preset edge change threshold, where pgT∈[0.1,0.4]; hgT represents the preset grayscale change threshold, where pgT∈[30,70]; pgT2 represents the preset edge feature threshold, where pgT2∈[0.15,0.3];

[0085] Based on the connectivity properties of edge pixels, edge pixels connected to foreground feature edge pixels are marked as foreground edge pixels, and the region enclosed by the foreground edge pixels is further marked as the preliminary foreground region A. - And further based on each preparatory foreground area A - The area S occupied - The area S occupied by the prepared foreground region - The foreground area that is larger than the preset area standard value ST is marked as foreground area A, and the remaining areas other than the foreground area are marked as background area B;

[0086] When a foreground target region A is detected in the image, the video monitoring frame X is marked as a close-up image X∈X. N Otherwise, if no foreground target can be detected, the video monitoring frame X is marked as a distant image X∈X. F ;

[0087] 3) For close-up images X∈X N Enhancement processing is applied to the marked foreground region, including:

[0088] Convert the video monitoring image to the Lab color space and extract the luminance channel sub-image L, color channel sub-image a, and color channel sub-image b from the video monitoring image;

[0089] The brightness of the foreground region A is enhanced based on the obtained brightness channel sub-image L, where the brightness enhancement function used is:

[0090]

[0091] Where L′(x,y) represents the luminance channel value of pixel (x,y) after luminance enhancement adjustment, where pixel (x,y)∈A, and L(x,y) represents the luminance channel value of pixel (x,y) before luminance enhancement adjustment, where (a,b)∈A. (x,y) This indicates that pixel (a, b) belongs to the foreground region A where pixel (x, y) is located. (x,y) In the image, L(a,b) represents the luminance channel value of pixel (a,b). and These represent the foreground regions A and B, respectively. (x,y)The maximum and minimum values ​​of the luminance channel values ​​of each pixel; LT1 represents the preset first luminance channel value, where LT1∈[30,35]; LT2 represents the preset second luminance channel value, where LT2∈[30,40], where LT1 and LT2 are set accordingly. When LT1 is larger, LT2 is smaller, and vice versa.

[0092] Further brightness adaptation adjustments are made to the background area, using the following brightness adaptation adjustment function:

[0093]

[0094] Among them, L * (x,y) represents the luminance channel value of pixel (x,y) after luminance adaptation adjustment, where pixel (x,y)∈B, d2A((x,y)) represents the pixel distance from pixel (x,y) to the nearest foreground region A, and dz represents the preset standard value of pixel distance; (a,b)∈A mind2A L'(a,b) represents the pixel point (a,b) belonging to the foreground region A, which is closest to the pixel point (x,y); L'(a,b) represents the brightness channel value of the pixel point (a,b) after brightness enhancement adjustment, L(x,y) represents the brightness channel value of the pixel point (x,y) before brightness adaptation adjustment, and LT3 represents the preset third brightness channel value, where LT3∈[75,85];

[0095] After performing brightness enhancement adjustment on foreground region A and brightness adaptation adjustment on background region B respectively, an enhanced brightness channel sub-image is obtained. Then, based on the enhanced brightness channel sub-image, color channel sub-image a and color channel sub-image b, an inverse color space transformation is performed to obtain the pre-processed video surveillance image. The pre-processed video surveillance image is then input into the anomaly analysis unit.

[0096] 4) For the distant image X∈X F Brightness enhancement adjustment is performed on the entire video monitoring screen, including:

[0097] Convert the video monitoring image to the Lab color space and extract the luminance channel sub-image L, color channel sub-image a, and color channel sub-image b from the video monitoring image;

[0098] The brightness enhancement adjustment is performed based on the obtained brightness channel sub-image L, where the brightness enhancement adjustment function used is:

[0099]

[0100] Among them, L ′′(x,y) represents the luminance channel value of pixel (x,y) after brightness enhancement adjustment, where pixel (x,y)∈X, L(x,y) represents the luminance channel value of pixel (x,y) before brightness enhancement adjustment, (a,b)∈X means that pixel (a,b) belongs to the pixel in the video monitoring frame X, and L(a,b) represents the luminance channel value of pixel (a,b). and LT4 and LT5 represent the maximum and minimum values ​​of the brightness channel for each pixel in the video monitoring image X, respectively; LT4 represents the preset fourth brightness channel value, where LT4∈[15,25]; LT5 represents the preset fifth brightness channel value, where LT5∈[50,70], where LT4 and LT5 are set accordingly. When LT4 is larger, LT5 is smaller, and vice versa.

[0101] After performing brightness enhancement adjustment on foreground region A and brightness adaptation adjustment on background region B respectively, an enhanced brightness channel sub-image is obtained. Then, based on the enhanced brightness channel sub-image, color channel sub-image a and color channel sub-image b, an inverse color space transformation is performed to obtain the pre-processed video surveillance image. The pre-processed video surveillance image is then input into the anomaly analysis unit.

[0102] Optionally, based on the connectivity characteristics of the edge pixels, edge pixels connected to the foreground feature edge pixels are marked as foreground edge pixels, and the area enclosed by the foreground edge pixels is further marked as foreground region A, and the remaining areas other than the foreground region are marked as background region B.

[0103] Optional, standard area value Where S′ represents the total (pixel) area of ​​the video surveillance image.

[0104] Optionally, the edge detection operators include the Canny operator, the Sobel operator, the Prewitt operator, etc.

[0105] Considering that in actual anomaly analysis, the video monitoring data acquired may vary in format due to adjustments in shooting angle and content based on the actual location, traditional image processing methods may be unable to perform targeted preprocessing of the video monitoring data, thus affecting its quality. Furthermore, in subsequent anomaly analysis based on the video monitoring data, the specific anomaly analysis model invoked varies depending on the type of video monitoring data. Therefore, inconsistent image monitoring data quality will also affect the accuracy and reliability of subsequent video analysis. The present invention proposes a technical solution for preprocessing acquired video monitoring data based on a preprocessing unit. This solution adaptively classifies distant and near-field images according to the video monitoring data, effectively improving the targeting of video monitoring data preprocessing. (When acquiring targeted video monitoring data (e.g., for floating object monitoring, covering object monitoring, etc.), it is necessary to acquire image monitoring data from a relatively wide shooting angle and use the video monitoring data to monitor the region of interest in the image; however, during video acquisition, it is impossible to pre-mark whether the video monitoring data belongs to the distant or near-field (for example, normally a distant image is acquired, but if floating objects are present, it becomes a near-field image)). Specifically, the proposed segmentation scheme is based on the edge change characteristics and grayscale change characteristics of image edge pixels, combined with a proposed foreground feature judgment function to accurately extract edge pixels belonging to the foreground region. This accurately extracts the foreground region in the video monitoring image to distinguish between distant and near-field types, helping to improve the adaptability and intelligence level of subsequent targeted enhancement processing.

[0106] For close-up images with foreground targets in video monitoring footage, a brightness enhancement scheme is proposed to improve the clarity and concentration of the foreground area. This scheme specifically enhances the clarity of the foreground region while simultaneously making adaptive adjustments to the background region. This improves the overall image clarity while highlighting the foreground target, effectively enhancing the accuracy of subsequent anomaly analysis based on video monitoring data, such as detecting floating objects, bird damage, and covering materials where key features exist in the foreground area. For distant images, adaptive brightness enhancement is applied to the entire image, effectively improving the clarity of each region within the image. This enhances the accuracy of subsequent anomaly analysis based on video monitoring data, such as fire identification where key features exist in a wide area. Overall, this improves the adaptability and reliability of anomaly analysis based on video monitoring data.

[0107] The preprocessed video monitoring data is then input into the corresponding video analysis model through the anomaly analysis unit for targeted analysis, which can effectively improve the reliability and accuracy of anomaly analysis.

[0108] The fire identification model is based on a convolutional neural network and consists of an input layer, three consecutive convolutional layers, a fully connected layer, and an output layer. The convolutional layers use a 3×3 kernel and 32 channels. Each convolutional layer is followed by a pooling layer with a 2×2 kernel and max pooling. The activation function is ReLU. The output layer uses softmax to output the result of whether a fire was detected or not.

[0109] The floating object recognition model is built using a template-based recognition model. It extracts floating objects from the background area of ​​video images and matches the extracted floating object images with a preset template to identify the type and specific location of the floating objects in the images. The model then compares and analyzes the floating object type and location with the preset safety range of power transmission equipment to obtain the floating object anomaly recognition results.

[0110] Based on a pre-trained video analysis model, it is possible to complete specified anomaly analysis tasks using acquired transmission line video monitoring data. Specifically, fire identification models, floating object identification models, bird damage identification models, corrosion identification models, and covering material identification models can also achieve their corresponding functions using pre-trained video analysis models in existing technologies. That is, based on the acquired transmission line video monitoring data, corresponding intelligent monitoring and anomaly analysis tasks can be completed; this invention does not impose specific limitations on these models.

[0111] Preferably, the system also includes a backend management module;

[0112] The backend management module is used to manage the area type classification information based on location information, so that different location information corresponds to one or more area types;

[0113] It also manages the corresponding monitoring standards for each type of region, including setting corresponding monitoring standards for different types of regions and specific judgment indicators for each monitoring standard; and associates the pre-stored data processing model with the corresponding monitoring standard so that it can be called when the acquired transmission line monitoring data is used for anomaly analysis using the monitoring standard.

[0114] Through the back-end management module, managers can pre-set the area type of the transmission line coverage area according to the actual situation, and also set the monitoring standards for different area types. The setting information is then transmitted to the drone data collection module and control module to realize the input of setting information.

[0115] The backend management module can also input the data analysis model used for anomaly analysis into the control module, so that the control module can call it when performing anomaly analysis on specific monitoring data for specific regions.

[0116] It should be noted that the functional units / modules in the various embodiments of the present invention can be integrated into one processing unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated into one unit / module. The integrated unit / module described above can be implemented in hardware or in the form of software functional units / modules.

[0117] From the above description of the embodiments, those skilled in the art will clearly understand that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field-Programmable Gate Array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein, or combinations thereof. For software implementation, some or all of the processes of the embodiments can be implemented by a computer program instructing the associated hardware. During implementation, the program can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. Computer-readable media can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should be able to analyze that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A visual monitoring system for key areas of power transmission lines, characterized in that, Includes a drone data acquisition module and a control module; The drone data acquisition module is used to acquire power transmission line monitoring data corresponding to the area type information from the current inspection location, and then transmit the acquired power transmission line monitoring data back to the control module. The control module is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding monitoring standards, perform anomaly analysis on the transmission line monitoring data based on the acquired monitoring standards, and obtain the anomaly analysis results. And for integrating the corresponding transmission line monitoring data and anomaly analysis results into the map model for visualization; The control module includes: preprocessing the acquired video monitoring data; The corresponding video analysis model is invoked to perform anomaly analysis on the video monitoring data, and the video anomaly analysis results are obtained. The preprocessing of the acquired video monitoring data includes: 1) Perform image frame processing on the acquired video monitoring data and extract the video monitoring image X from each frame; 2) Extract edge pixels from the acquired video monitoring image X, and determine foreground features based on the edge change features and grayscale change features of the edge pixels; The foreground feature judgment function used is: in, Represents edge pixels Foreground feature judgment function, where When the corresponding conditions are met, the pixel will be... Pixels marked as not belonging to the foreground feature edge Or belong to the edge pixels of the foreground features , and indicates the AND / AND logical operation; This represents the preset edge change threshold, where ; This represents the preset grayscale change threshold, where ; The preset edge feature threshold is shown, where ; Represents pixels Edge change factor, Represents pixels Gray-scale variation factor; Represents the set of edge pixels; Shown by pixels Centered The rectangular area range, Represents pixels For the region Pixels within; Represents pixels Edge density factor; Based on the connectivity characteristics of edge pixels, edge pixels connected to foreground feature edge pixels are marked as foreground edge pixels. Foreground region A is obtained based on the area enclosed by the foreground edge pixels, and the remaining areas other than the foreground region are marked as background region B. When a foreground area is detected in the image, the video monitoring frame is marked as a close-up image. Otherwise, mark it as a distant view image. ; 3) For close-up images, enhance the foreground area and adjust the brightness of the background area to obtain the pre-processed video surveillance image; 4) For distant images, the brightness of the entire video monitoring screen is enhanced and adjusted to obtain the pre-processed video monitoring screen.

2. The visualization monitoring system for key areas of transmission lines according to claim 1, characterized in that, The human-machine data acquisition module includes a positioning unit, a control unit, a monitoring unit, and a communication unit; The positioning unit is used to acquire real-time positioning information from the UAV acquisition module; The control unit is used to control the UAV data acquisition module to complete the inspection along the preset inspection route. During the inspection, it matches the area type information according to the current positioning information and collects the corresponding power transmission line monitoring data from the control test unit according to the area type information. The area types include forest areas, residential areas, heavily polluted areas, lightning areas, bird activity areas, areas damaged by external forces, and flood areas. The monitoring unit is used to collect transmission line monitoring data according to the instructions of the control unit, including video monitoring data pointed at different devices or angles; The communication unit is used to transmit the acquired transmission line monitoring data to the control module.

3. The visualization monitoring system for key areas of transmission lines according to claim 2, characterized in that, The monitoring unit includes an image acquisition unit, a sensor unit, and a communication unit; The image acquisition unit includes a camera unit with a pan-tilt head, wherein the shooting angle of the camera and the angle of the pan-tilt head are controlled by a control unit; the camera unit is used to capture video monitoring data at a specified angle; The sensor unit is equipped with different types of sensors, and the corresponding sensor monitoring data is acquired through each type of sensor. The sensor monitoring data includes temperature data and humidity data.

4. The visualization monitoring system for key areas of transmission lines according to claim 2, characterized in that, The control module includes an extraction unit, an anomaly analysis unit, and a visualization unit; The extraction unit is used to extract the corresponding location information based on the acquired transmission line monitoring data, match the corresponding area type information based on the location information, record the corresponding area type identifier, and match the corresponding monitoring standard based on the area type identifier. The anomaly analysis unit is used to perform anomaly analysis based on the acquired monitoring data, analyze whether the monitoring data meets the corresponding monitoring standards, and obtain the anomaly analysis results. The visualization unit is used to integrate transmission line monitoring data and corresponding anomaly analysis results into the visualization map model, and display them at the corresponding locations in the visualization map model based on the location information of the monitoring data.

5. A visualization monitoring system for key areas of transmission lines according to claim 4, characterized in that, The anomaly analysis unit includes a video analysis unit and an environmental analysis unit; The video analysis unit is used to retrieve the corresponding video analysis model to perform anomaly analysis on the video monitoring data based on the area type to which the acquired transmission line video monitoring data belongs, and obtain the video anomaly analysis results. The video analysis models include fire identification model, floating object identification model, bird damage identification model, corrosion identification model, and covering identification model based on video data. The environmental analysis unit is used to compare and analyze the acquired sensor monitoring data with the environmental standards of the corresponding area type to obtain environmental anomaly analysis results.

6. The visualization monitoring system for key areas of transmission lines according to claim 5, characterized in that, The video analysis unit includes a preprocessing unit and an anomaly analysis unit; The preprocessing unit is used to preprocess the acquired video monitoring data to improve the clarity of the video monitoring data; The anomaly analysis unit is used to retrieve the corresponding video analysis model to perform anomaly analysis on the video monitoring data based on the region type to which the acquired transmission line video monitoring data belongs, and obtain the video anomaly analysis results.

7. The visualization monitoring system for key areas of transmission lines according to claim 1, characterized in that, It also includes a backend management module; The backend management module is used to manage the area type classification information based on location information, so that different location information corresponds to one or more area types; It also manages the corresponding monitoring standards for each type of region, including setting corresponding monitoring standards for different types of regions and specific judgment indicators for each monitoring standard; and associates the pre-stored data processing model with the corresponding monitoring standard so that it can be called when the acquired transmission line monitoring data is used for anomaly analysis using the monitoring standard.

8. The visualization monitoring system for key areas of transmission lines according to claim 1, characterized in that, In the preprocessing unit, edge pixels are extracted from the acquired video monitoring image X, and foreground target detection is performed based on the edge change features and grayscale change features of the edge pixels. Specifically, this includes: Edge detection is performed on the video monitoring image, edge pixels in the video monitoring image are extracted and marked, and the edge pixels in the image are counted into the edge pixel set K. Based on the marked edge pixels, the edge density factor of each pixel in the image is calculated. The edge density factor calculation function used is as follows: in, Represents pixels edge density factor, Represented by pixels Centered The rectangular area range, Represents pixels For the region Pixels within; As a condition for judgment, it represents a pixel. Marked as edge pixels This indicates that the judgment condition is met. The number of pixels, This indicates the side length of the preset rectangular area; The edge variation factor of each edge pixel is further calculated, and the edge variation factor calculation function used is: in, Represents pixels Edge change factor, Represented by pixels Centered The rectangular area range, Represents pixels For the region Pixels within, and Representing regions The maximum and minimum values ​​of the edge density factor for each pixel within the range; Calculate each edge pixel separately The grayscale variation factor, wherein the grayscale variation factor calculation function is: in, Represents pixels The grayscale variation factor, where , Represented by pixels Centered The rectangular area range, Represents pixels For the region Pixels within, Represents pixels These are non-edge pixels; and Representing regions The maximum and minimum grayscale values ​​of non-edge pixels within the range; Foreground features are determined for edge pixels based on the obtained edge density factor and grayscale variation factor. The foreground feature determination function used is as follows: in, Represents edge pixels Foreground feature judgment function, where When the corresponding conditions are met, the pixel will be... Pixels marked as not belonging to the foreground feature edge Or belong to the edge pixels of the foreground features , This represents the AND / AND logical operation; This represents the preset edge change threshold, where ; This represents the preset grayscale change threshold, where ; This represents the preset edge feature threshold, where ; Based on the connectivity properties of edge pixels, edge pixels connected to foreground feature edge pixels are marked as foreground edge pixels, and the region enclosed by the foreground edge pixels is further marked as the preliminary foreground region. And further based on each preparatory foreground area Area size The size of the area occupied by the prepared foreground region Larger than the preset area standard value The prepared foreground area is marked as foreground area A, and the remaining areas other than the foreground area are marked as background area B; When a foreground region A is detected in the image, the video monitoring frame X is marked as a close-up image. Otherwise, if no foreground area can be detected, the video monitoring frame X is marked as a distant image. .

9. A visualization monitoring system for key areas of transmission lines according to claim 1, characterized in that, In the preprocessing unit, for close-up images Enhancement processing is applied to the marked foreground region, specifically including: Convert the video monitoring image to the Lab color space and extract the luminance channel sub-image L, color channel sub-image a, and color channel sub-image b from the video monitoring image; The brightness of the foreground region A is enhanced based on the obtained brightness channel sub-image L, where the brightness enhancement function used is: in, Indicates the pixel after brightness enhancement adjustment The brightness channel value, where the pixel is , Indicates the pixel value before brightness enhancement adjustment. The brightness channel value, Represents pixels Belongs to pixels Foreground area The pixels in Represents pixels The brightness channel value, and Representing the foreground area The maximum and minimum values ​​of the brightness channel values ​​for each pixel; This represents the preset first brightness channel value, where ; This represents the preset value of the second brightness channel, where ,in and Corresponding settings, when The larger, the better The smaller, the better. The smaller, the better The larger; Further brightness adaptation adjustments are made to the background area, using the following brightness adaptation adjustment function: in, Indicates the number of pixels after brightness adjustment The brightness channel value, where the pixel is , Represents pixels The pixel distance to the nearest foreground region A. This represents the preset standard value for pixel distance; Represents pixels Belongs to pixels The nearest pixel in foreground region A; Indicates the pixel after brightness enhancement adjustment The brightness channel value, Indicates the pixel before brightness adaptation adjustment The brightness channel value, This represents the preset value of the third brightness channel, where ; After performing brightness enhancement adjustment on foreground region A and brightness adaptation adjustment on background region B respectively, an enhanced brightness channel sub-image is obtained. Then, based on the enhanced brightness channel sub-image, color channel sub-image a and color channel sub-image b, an inverse color space transformation is performed to obtain the pre-processed video surveillance image. The pre-processed video surveillance image is then input into the anomaly analysis unit.

10. A visualization monitoring system for key areas of transmission lines according to claim 1, characterized in that, In the preprocessing unit, for distant images Brightness enhancement adjustment is performed on the entire video monitoring screen, including: Convert the video monitoring image to the Lab color space and extract the luminance channel sub-image L, color channel sub-image a, and color channel sub-image b from the video monitoring image; The brightness enhancement adjustment is performed based on the obtained brightness channel sub-image L, where the brightness enhancement adjustment function used is: in, Indicates the pixel after brightness enhancement adjustment The brightness channel value, where the pixel is , Indicates the pixel value before brightness enhancement adjustment. The brightness channel value, Represents pixels The pixel that belongs to X in the video monitoring image. Represents pixels The brightness channel value, and These represent the maximum and minimum values ​​of the brightness channel for each pixel in the video monitoring image X, respectively. This represents the preset value of the fourth brightness channel, where ; This represents the preset value of the fifth brightness channel, where ,in and Corresponding settings, when The larger, the better The smaller, the better. The smaller, the better The larger; After completing the brightness enhancement adjustment, an enhanced brightness channel sub-image is obtained. Then, based on the enhanced brightness channel sub-image, color channel sub-image a and color channel sub-image b, an inverse color space transformation is performed to obtain the pre-processed video monitoring image. The pre-processed video monitoring image is then input into the anomaly analysis unit.

Images