State detection method and device of power supply lightning protection device and electronic equipment
By collecting and analyzing the target images of the power supply lightning protector in real time and judging its status using pre-trained models, the problem of traditional manual detection is solved, and more efficient and accurate status monitoring is achieved.
Patent Information
- Application Number
- CN202410954579.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-19
AI Technical Summary
The traditional power supply lightning protection device status detection relies on manual inspection, which is time-consuming and labor-intensive and vulnerable to subjective factors, resulting in inaccurate detection results.
By collecting the target image of the power supply lightning protector in real time, input it to the pre-trained target detection model, marking the bounding box of the lightning protector area and the observation window area, cropping the target area of interest, determining the color channel mean, and according to the color channel mean relationship Determine the status of the lightning protector.
It improves the efficiency and accuracy of power supply lightning protector status monitoring, reduces manual intervention, and ensures the reliability of detection results.
Smart Images

Figure CN118982650A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of equipment monitoring, and in particular to a state detection method, device and electronic equipment of a power supply lightning arrester. Background Art
[0002] Power lightning arresters play a vital role in power systems. Their main function is to protect electrical equipment from damage caused by lightning strikes and power surges. During use, the status indicator window of the lightning arrester will display red or green to indicate the functional status of the lightning arrester: red indicates that the lightning arrester has failed, and green indicates that the lightning arrester functions normally. In order to ensure the safe operation of the power system, it is particularly important to detect the status of the lightning arrester in a timely and accurate manner.
[0003] Traditional lightning arrester status detection mainly relies on manual inspection, which is not only time-consuming and labor-intensive, but also easily affected by subjective factors, resulting in inaccurate detection results. Summary of the invention
[0004] The embodiments of the present disclosure at least provide a method, device and electronic device for detecting the state of a power supply lightning arrester, which can improve the efficiency and accuracy of monitoring the state of the power supply lightning arrester.
[0005] The present disclosure provides a method for detecting the state of a power supply lightning arrester, including:
[0006] Real-time acquisition of target images corresponding to power supply lightning arresters in the area to be detected;
[0007] Inputting the target image into a pre-trained target detection model, marking a first bounding box including a lightning arrester area and a second bounding box including an observation window area corresponding to the lightning arrester in the target image;
[0008] Cut out a target region of interest in each of the second bounding boxes, and determine a pixel color mean value of each color channel corresponding to the target region of interest;
[0009] According to the size relationship between the pixel color mean values of different color channels, the target color corresponding to the observation window is determined, and the operating state of the power supply lightning arrester is determined according to the target color.
[0010] In an optional implementation manner, the method further includes:
[0011] According to the attribution relationship between the observation window and the corresponding lightning arrester, matching the first bounding box and the second bounding box into a plurality of bounding box pairs;
[0012] When it is detected that the operating state of the power supply lightning arrester is that the lightning arrester function fails, determining a corresponding target second boundary box;
[0013] Determine, according to the bounding box pair, a target first bounding box that matches the target second bounding box, and determine a target lightning arrester corresponding to the target first bounding box;
[0014] A lightning arrester function failure alarm corresponding to the target lightning arrester is sent to the user.
[0015] In an optional implementation, according to the attribution relationship between the observation window and the corresponding lightning arrester, matching the first bounding box and the second bounding box into a plurality of bounding box pairs specifically includes:
[0016] For each of the first bounding boxes, determining an area of an intersection between the first bounding box and each of the second bounding boxes;
[0017] When the ratio between the area of the intersection region and the area of the corresponding second boundary box is 1, it is determined that the observation window corresponding to the second boundary box belongs to the lightning arrester corresponding to the first boundary box.
[0018] In an optional implementation, cutting out a target region of interest from each of the second bounding boxes and determining a pixel color mean value of each color channel corresponding to the target region of interest specifically includes:
[0019] For each pixel in the target region of interest, determine a pixel color value corresponding to the pixel in each color channel, wherein the types of the color channels include a red channel, a green channel, and a blue channel;
[0020] For each color channel, determine the average value of the pixel color values corresponding to the color channel of all pixels in the target region of interest.
[0021] In an optional implementation, determining the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels specifically includes:
[0022] When the pixel color mean value corresponding to the red channel is greater than the pixel color mean value corresponding to the green channel, determining that the observation window is red;
[0023] When the pixel color mean value corresponding to the red channel is less than the pixel color mean value corresponding to the green channel, it is determined that the observation window is green.
[0024] In an optional implementation, the target detection model is trained based on the following steps:
[0025] Collect sample lightning arrester images taken in a variety of different environments and at different angles, and mark the position of the lightning arrester contained in the sample lightning arrester image and the position of the observation window in the lightning arrester as data labels;
[0026] Performing data enhancement on the sample lightning arrester image, and constructing the sample lightning arrester image after data enhancement and the corresponding data label into a sample image data set;
[0027] The sample image data set is divided into a training set and a validation set, the training set is used to train the target detection model, and the validation set is used to evaluate the trained target detection model.
[0028] In an optional implementation, the target region of interest is a rectangular region centered at the center point of the second bounding box, with a length and a width both being half the length and the width of the second bounding box.
[0029] The present disclosure also provides a state detection device for a power supply lightning arrester, including:
[0030] An image acquisition module is used to acquire the target image corresponding to the power supply lightning arrester in the area to be detected in real time;
[0031] A region recognition module, used for inputting the target image into a pre-trained target detection model, marking a first bounding box including a lightning arrester region and a second bounding box including an observation window region corresponding to the lightning arrester in the target image;
[0032] An observation window color determination module, used to cut out a target region of interest in each second boundary frame, and determine a pixel color mean value corresponding to each color channel of the target region of interest;
[0033] The lightning arrester status determination module is used to determine the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determine the operating status of the power supply lightning arrester according to the target color.
[0034] An embodiment of the present disclosure also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the state detection method of the power supply lightning arrester or the steps in any possible implementation of the state detection method of the power supply lightning arrester are executed.
[0035] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the state detection method of the power supply lightning arrester or the steps in any possible implementation of the state detection method of the power supply lightning arrester is executed.
[0036] The embodiment of the present disclosure also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the above-mentioned power supply lightning arrester status detection method, or the steps in any possible implementation of the above-mentioned power supply lightning arrester status detection method.
[0037] The disclosed embodiment provides a state detection method, device and electronic device for a power lightning arrester, which collects the target image corresponding to the power lightning arrester in the detection area in real time; inputs the target image into a pre-trained target detection model, marks the first bounding box including the lightning arrester area and the second bounding box including the observation window area corresponding to the lightning arrester in the target image; cuts out the target region of interest in each second bounding box, determines the pixel color mean of each color channel corresponding to the target region of interest; determines the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determines the operating state of the power lightning arrester according to the target color. The efficiency and accuracy of the state monitoring of the power lightning arrester can be improved.
[0038] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.
[0040] Figure 1 A flow chart of a method for detecting the state of a power supply lightning arrester provided by an embodiment of the present disclosure is shown;
[0041] Figure 2 A flow chart of a method for training a target detection model provided by an embodiment of the present disclosure is shown;
[0042] Figure 3 A schematic diagram of a state detection device for a power supply lightning arrester provided in an embodiment of the present disclosure is shown;
[0043] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0044] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.
[0045] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0046] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.
[0047] Research has found that power lightning arresters play a vital role in power systems. Their main function is to protect electrical equipment from damage caused by lightning strikes and surges. During use, the status indicator window of the lightning arrester will display red or green to indicate the functional status of the lightning arrester: red indicates that the lightning arrester has failed, and green indicates that the lightning arrester functions normally. In order to ensure the safe operation of the power system, it is particularly important to detect the status of the lightning arrester in a timely and accurate manner. Traditional lightning arrester status detection mainly relies on manual inspections, which is not only time-consuming and labor-intensive, but also easily affected by subjective factors, resulting in inaccurate detection results.
[0048] Based on the above research, the present disclosure provides a state detection method, device and electronic device for a power lightning arrester, which collects the target image corresponding to the power lightning arrester in the area to be detected in real time; inputs the target image into a pre-trained target detection model, and marks the first bounding box including the lightning arrester area and the second bounding box including the observation window area corresponding to the lightning arrester in the target image; cuts out the target region of interest in each second bounding box, and determines the pixel color mean of each color channel corresponding to the target region of interest; determines the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determines the operating state of the power lightning arrester according to the target color. The efficiency and accuracy of the state monitoring of the power lightning arrester can be improved.
[0049] To facilitate understanding of this embodiment, a state detection method of a power lightning arrester disclosed in the embodiment of the present disclosure is first introduced in detail. The execution subject of the state detection method of the power lightning arrester provided in the embodiment of the present disclosure is generally a computer device with a certain computing capability, and the computer device includes, for example: a terminal device or a server or other processing device, and the terminal device can be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the state detection method of the power lightning arrester can be implemented by a processor calling a computer-readable instruction stored in a memory.
[0050] See also Figure 1 As shown, it is a flow chart of a state detection method of a power supply lightning arrester provided by an embodiment of the present disclosure, the method comprises steps S101 to S104, wherein:
[0051] S101, collecting target images corresponding to power supply lightning arresters in the area to be detected in real time.
[0052] In a specific implementation, the actual working area of the power supply lightning arrester is determined as the area to be detected, and the target image corresponding to the power supply lightning arrester in the area to be detected is collected in real time.
[0053] Here, each target image includes at least a power lightning arrester and a corresponding observation window.
[0054] S102: input the target image into a pre-trained target detection model, and mark in the target image a first bounding box including a lightning arrester area and a second bounding box including an observation window area corresponding to the lightning arrester.
[0055] In a specific implementation, a target image containing a power supply lightning arrester collected in real time in the area to be detected is input into a pre-trained target detection model. The target detection model output includes multiple bounding boxes, each of which represents a detected target, including a lightning arrester and an observation window.
[0056] Preferably, the target detection model can be a YOLOv5 model.
[0057] Here, the bounding boxes of all detected lightning arresters and observation windows are extracted from the target detection results. The bounding box information usually includes the position (upper left corner and lower right corner coordinates) and size (width and height).
[0058] The first boundary box is a boundary box including the area of the power supply lightning arrester, and the second boundary box is a boundary box including the observation window area corresponding to the power supply lightning arrester.
[0059] As a possible implementation, after determining a first bounding box including the lightning arrester area and a second bounding box including the observation window area corresponding to the lightning arrester, the first bounding box and the second bounding box can be matched into multiple bounding box pairs according to the relationship between the observation window and the corresponding lightning arrester.
[0060] In a specific implementation, for each first bounding box, the area of the intersection region between the first bounding box and each second bounding box is determined; when the ratio between the area of the intersection region and the corresponding second bounding box is 1, it is determined that the observation window corresponding to the second bounding box belongs to the lightning arrester corresponding to the first bounding box.
[0061] Here, the IOU calculation is performed on the second bounding box of each pair of observation windows and the first bounding box of the lightning arrester to find the area of their intersection area. The intersection area is the overlapping part of the two bounding boxes. If the width or height of the intersection area is less than or equal to zero, it means that the two bounding boxes have no overlapping part, and the area of the intersection area is 0.
[0062] The IOU is calculated using the ratio between the area of the intersection region and the area of the second bounding box, and the IOU value range is 0 to 1, where 1 indicates complete overlap, that is, the observation window is completely contained in the bounding box of the lightning arrester.
[0063] Furthermore, the pairing results of the power lightning arrester and the observation window are stored, and each lightning arrester boundary box is associated with one or more observation window boundary boxes. This information will be used in the subsequent state detection step.
[0064] As a possible implementation, see Figure 2 FIG. 2 is a flowchart of a method for training a target detection model provided by an embodiment of the present disclosure, wherein the method includes steps S201 to S204, wherein:
[0065] S201. Collect sample lightning arrester images taken in a variety of different environments and at different angles, and mark the position of the lightning arrester contained in the sample lightning arrester image and the position of the observation window in the lightning arrester as data labels.
[0066] S202: Perform data enhancement on the sample lightning arrester image, and construct the sample lightning arrester image after data enhancement and the corresponding data label into a sample image data set.
[0067] S203, dividing the sample image data set into a training set and a validation set, using the training set to train the target detection model, and using the validation set to evaluate the trained target detection model.
[0068] In the specific implementation, a large amount of sample lightning arrester image data containing lightning arresters is collected, and these sample lightning arrester images should cover a variety of different environments and angles to ensure the generalization ability of the model.
[0069] Here, the data sources of sample lightning arrester images can include field shooting, historical data, and public data sets. For field shooting, take pictures of lightning arresters in different scenes such as power facilities, substations, and distribution rooms. These scenes should include various environmental conditions, such as indoors and outdoors, day and night, sunny and rainy days, etc.; shoot lightning arresters from different angles and heights to ensure that the position and size of lightning arresters in the image are diverse. For example, shoot from multiple angles such as the front, side, top, and bottom. For historical data, collect lightning arrester pictures recorded by power companies during equipment maintenance. These pictures usually have rich scene and annotation information; capture pictures containing lightning arresters from surveillance videos of substations and distribution rooms to ensure that these pictures can reflect the real working environment and status. For public data sets, obtain public lightning arrester picture data sets from the Internet. These data sets are usually professionally annotated and can be used to enrich training data.
[0070] Among them, in order to improve the generalization ability of the model, the collected data needs to be diverse. Specific requirements include: environmental diversity, equipment diversity, and shooting diversity. For environmental diversity: Lighting conditions: including pictures under various lighting conditions such as strong light, weak light, shadow, backlight, etc.; Weather conditions: including pictures under various weather conditions such as sunny days, rainy days, foggy days, snowy days, etc.; Background complexity: including pictures of simple backgrounds and complex backgrounds to ensure that the model can identify lightning arresters under different backgrounds. For equipment diversity: collect pictures of lightning arresters of different models and manufacturers to ensure that the model can identify various types of lightning arresters; collect pictures of lightning arresters in various states such as normal, abnormal, and damaged to ensure that the model can identify the working status of lightning arresters; For shooting diversity: shoot lightning arresters from different angles, including the front, side, obliquely above, obliquely below, etc.; shoot lightning arresters from different distances, including close distance, medium distance, and long distance.
[0071] Furthermore, after collecting the sample lightning arrester images, the lightning arrester needs to be labeled. Professional labeling tools, such as LabelImg, RectLabel, etc., are used to label the lightning arrester in the image. The labeled content includes the position of the lightning arrester and the position of the corresponding observation window of the lightning arrester.
[0072] The annotation data is saved as an annotation file in YOLO format. Each image corresponds to a annotation file, and the file content includes the location of the lightning arrester (the coordinates of the bounding box) and the label of the location of the observation window corresponding to the lightning arrester.
[0073] Here, after data annotation, the IMGAUG data enhancement tool is used to expand the data to improve the recognition accuracy and robustness of the model. The data enhancement process includes: blurring, Gaussian blurring of the image; plane rotation, randomly rotating the image by a certain angle; mirror flipping, horizontally and vertically flipping the image; Gaussian noise, adding random noise to the image.
[0074] Specifically, the labeled data is converted into a format corresponding to the target detection model to construct a sample image dataset, such as the YOLOv5 format. Each sample lightning arrester image in the sample image dataset corresponds to a labeling file, which contains the category and bounding box coordinates of the target in the sample lightning arrester image.
[0075] Here, the sample image dataset is divided into a training set and a validation set. The training set is used for model training, and the validation set is used for model evaluation. Initialize the model parameters, and select an appropriate optimizer (such as SGD or Adam) and loss function (such as cross entropy loss). Use the training set to train the model, input image data batch by batch, calculate the loss, and adjust the model parameters through the back propagation algorithm. After each training round, use the validation set to evaluate the model performance and calculate indicators such as accuracy, recall, and F1 score. By observing the changes in loss values and evaluation indicators, determine whether the model has converged, and adjust hyperparameters as needed.
[0076] During the training process, the model weight file is saved regularly. The optimal model in the training process can be saved to avoid overfitting or underfitting of the model.
[0077] Furthermore, the trained model is evaluated using validation and test sets, and the accuracy, recall, F1 score, and mAP (mean Average Precision) of the model on different datasets are calculated. The model's error detection is analyzed to identify common false detections and missed detections, analyze their causes, and improve them through further data enhancement or adjustment of the model architecture.
[0078] Optionally, based on the evaluation results, adjust the model's hyperparameters, such as learning rate, batch size, etc., to further improve model performance. Prune and quantize the model to reduce the number of model parameters and computational complexity, and improve the model's running speed and efficiency, which is particularly suitable for embedded devices and real-time applications. If the training dataset is small, you can use a pre-trained model for transfer learning. Pre-train the model using a large public dataset, and then fine-tune it using the lightning arrester dataset.
[0079] S103: Cut out a target region of interest in each second boundary frame, and determine a pixel color mean value of each color channel corresponding to the target region of interest.
[0080] In a specific implementation, the target region of interest is cropped out in the second bounding box, and the target region of interest is a rectangular area centered on the center point of the second bounding box, and the length and width are both half of the length and width of the second bounding box. By calculating the mean of the RGB three-channel pixels of the target region of interest, we can effectively determine the color of the observation window, thereby inferring the working status of the lightning arrester.
[0081] Specifically, for each pixel in the target region of interest, the pixel color value corresponding to the pixel in each color channel is determined, wherein the types of color channels include red channel, green channel and blue channel; for each color channel, the average value of the pixel color values corresponding to the color channel of all pixels in the target region of interest is determined.
[0082] Here, for each pixel in the ROI area, the values of its RGB (red, green, blue) three channels are extracted, the RGB values of all pixels in the ROI area are summed up, and the mean is calculated.
[0083] S104, determining a target color corresponding to the observation window according to a size relationship between the pixel color means of different color channels, and determining an operating state of the power supply lightning arrester according to the target color.
[0084] In a specific implementation, the lightning arrester observation window has only two colors, red and green. In the embodiment of the present application, the color can be judged by comparing the mean of the red channel (R) and the green channel (G). If the pixel color mean corresponding to the red channel is greater than the pixel color mean corresponding to the green channel, it is inferred that the area is red. At this time, the lightning arrester observation window displays red, indicating that the lightning arrester function has failed. If the pixel color mean corresponding to the red channel is less than the pixel color mean corresponding to the green channel, it means that the area is green. At this time, the lightning arrester observation window displays green, indicating that the lightning arrester function is normal.
[0085] It should be noted that this method is not limited to distinguishing red from green. The three colors of red, green and blue can all be identified and distinguished using the above method based on the size relationship between the pixel color mean values.
[0086] The disclosed embodiment provides a state detection method for a power lightning arrester, which collects a target image corresponding to the power lightning arrester in the area to be detected in real time; inputs the target image into a pre-trained target detection model, and marks a first bounding box including the lightning arrester area and a second bounding box including the observation window area corresponding to the lightning arrester in the target image; cuts out a target region of interest in each of the second bounding boxes, and determines the pixel color mean of each color channel corresponding to the target region of interest; determines the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determines the operating state of the power lightning arrester according to the target color. The efficiency and accuracy of the state monitoring of the power lightning arrester can be improved.
[0087] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0088] Based on the same inventive concept, the embodiment of the present disclosure also provides a status detection device for a power supply lightning arrester corresponding to the status detection method for a power supply lightning arrester. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the status detection method for the power supply lightning arrester in the embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.
[0089] See also Figure 3 , Figure 3 A schematic diagram of a state detection device for a power supply lightning arrester provided in an embodiment of the present disclosure. Figure 3 As shown in the figure, the state detection device 300 of the power supply lightning arrester provided in the embodiment of the present disclosure includes:
[0090] The image acquisition module 310 is used to acquire the target image corresponding to the power supply lightning arrester in the area to be detected in real time.
[0091] The area recognition module 320 is used to input the target image into a pre-trained target detection model, and mark a first bounding box including the lightning arrester area and a second bounding box including the observation window area corresponding to the lightning arrester in the target image.
[0092] The observation window color determination module 330 is used to cut out a target region of interest in each of the second boundary frames, and determine the pixel color mean value of each color channel corresponding to the target region of interest.
[0093] The lightning arrester status determination module 340 is used to determine the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determine the operating status of the power supply lightning arrester according to the target color.
[0094] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.
[0095] The state detection device of a power lightning arrester provided by the embodiment of the present disclosure collects the target image corresponding to the power lightning arrester in the detection area in real time; inputs the target image into a pre-trained target detection model, marks the first bounding box including the lightning arrester area and the second bounding box including the observation window area corresponding to the lightning arrester in the target image; cuts out the target region of interest in each second bounding box, determines the pixel color mean of each color channel corresponding to the target region of interest; determines the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determines the operating state of the power lightning arrester according to the target color. The efficiency and accuracy of the state monitoring of the power lightning arrester can be improved.
[0096] Corresponds to Figure 1 The state detection method of the power supply lightning arrester in the present disclosure also provides an electronic device 400, such as Figure 4 FIG. 4 is a schematic diagram of the structure of an electronic device 400 provided in an embodiment of the present disclosure, including:
[0097] Processor 41, memory 42, and bus 43; memory 42 is used to store execution instructions, including memory 421 and external memory 422; memory 421 here is also called internal memory, which is used to temporarily store operation data in processor 41 and data exchanged with external memory 422 such as hard disk. Processor 41 exchanges data with external memory 422 through memory 421. When the electronic device 400 is running, the processor 41 communicates with the memory 42 through bus 43, so that the processor 41 executes Figure 1 Steps of the status detection method of the power supply lightning arrester.
[0098] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the state detection method of the power supply lightning arrester described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.
[0099] The present disclosure also provides a computer program product, which includes computer instructions. When the computer instructions are executed by a processor, the steps of the state detection method of the power supply lightning arrester described in the above method embodiment can be executed. For details, please refer to the above method embodiment, which will not be repeated here.
[0100] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).
[0101] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.
[0102] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0103] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0104] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0105] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A state detection method for a power supply lightning arrester, characterized in that: include: Real-time acquisition of target images corresponding to power supply lightning arresters in the area to be detected; Inputting the target image into a pre-trained target detection model, marking a first bounding box including a lightning arrester area and a second bounding box including an observation window area corresponding to the lightning arrester in the target image; Cut out a target region of interest in each of the second bounding boxes, and determine a pixel color mean value of each color channel corresponding to the target region of interest; According to the size relationship between the pixel color mean values of different color channels, the target color corresponding to the observation window is determined, and the operating state of the power supply lightning arrester is determined according to the target color.
2. The method according to claim 1, characterized in that The method further comprises: According to the attribution relationship between the observation window and the corresponding lightning arrester, matching the first bounding box and the second bounding box into a plurality of bounding box pairs; When it is detected that the operating state of the power supply lightning arrester is that the lightning arrester function fails, determining a corresponding target second boundary box; Determine, according to the bounding box pair, a target first bounding box that matches the target second bounding box, and determine a target lightning arrester corresponding to the target first bounding box; A lightning arrester function failure alarm corresponding to the target lightning arrester is sent to the user.
3. The method according to claim 2, characterized in that According to the attribution relationship between the observation window and the corresponding lightning arrester, matching the first bounding box and the second bounding box into a plurality of bounding box pairs specifically includes: For each of the first bounding boxes, determining an area of an intersection between the first bounding box and each of the second bounding boxes; When the ratio between the area of the intersection region and the area of the corresponding second boundary box is 1, it is determined that the observation window corresponding to the second boundary box belongs to the lightning arrester corresponding to the first boundary box.
4. The method according to claim 1, characterized in that: Clipping a target region of interest in each of the second bounding boxes and determining a pixel color mean value of each color channel corresponding to the target region of interest specifically includes: For each pixel in the target region of interest, determine a pixel color value corresponding to the pixel in each color channel, wherein the types of the color channels include a red channel, a green channel, and a blue channel; For each color channel, determine the average value of the pixel color values corresponding to the color channel of all pixels in the target region of interest.
5. The method according to claim 4, characterized in that Determining the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels specifically includes: When the pixel color mean value corresponding to the red channel is greater than the pixel color mean value corresponding to the green channel, determining that the observation window is red; When the pixel color mean value corresponding to the red channel is less than the pixel color mean value corresponding to the green channel, it is determined that the observation window is green.
6. The method according to claim 1, characterized in that The object detection model is trained based on the following steps: Collect sample lightning arrester images taken in a variety of different environments and at different angles, and mark the position of the lightning arrester contained in the sample lightning arrester image and the position of the observation window in the lightning arrester as data labels; Performing data enhancement on the sample lightning arrester image, and constructing the sample lightning arrester image after data enhancement and the corresponding data label into a sample image data set; The sample image data set is divided into a training set and a validation set, the training set is used to train the target detection model, and the validation set is used to evaluate the trained target detection model.
7. The method according to claim 1, characterized in that: The target region of interest is a rectangular region centered at the center point of the second bounding box, with a length and a width both being half of the length and the width of the second bounding box.
8. A state detection device for a power supply lightning arrester, characterized in that: include: An image acquisition module is used to acquire the target image corresponding to the power supply lightning arrester in the area to be detected in real time; A region recognition module, used for inputting the target image into a pre-trained target detection model, marking a first bounding box including a lightning arrester region and a second bounding box including an observation window region corresponding to the lightning arrester in the target image; An observation window color determination module, used to cut out a target region of interest in each second boundary frame, and determine a pixel color mean value corresponding to each color channel of the target region of interest; The lightning arrester status determination module is used to determine the target color corresponding to the observation window according to the size relationship between the pixel color means of different color channels, and determine the operating status of the power supply lightning arrester according to the target color.
9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the state detection method of the power supply lightning arrester as described in any one of claims 1 to 7 are performed.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the state detection method of a power supply lightning arrester as claimed in any one of claims 1 to 7 are executed.
Citation Information
Patent Citations
Novel lightning arrester monitoring device
CN106093620A
Indicator lamp identification method and system
CN107392116A
Round indicator light positioning and state identification method for power distribution cabinet
CN113409238A
Equipment operation state identification method and device
CN114820676A
Indicator lamp state identification method based on color channel and related equipment
CN116580178A
Cited By
Conduction detection system and method for lightning protection device of wind power generation blade
CN119861313A
