Power line foreign matter identification method and system based on image processing

Abstract

The invention relates to the technical field of power management, and discloses a power line foreign matter identification method and system based on image processing, and the method comprises the steps: shooting an image of a power line region through a camera disposed in the power line region; performing color space conversion on the image; performing image enhancement on the image; analyzing the image by using a pre-trained deep learning model, extracting features in the image, and identifying foreign matters from the image according to the features in the image; calculating a risk index of the foreign matter according to a preset strategy; and according to the risk index of the foreign matter, generating a treatment measure for the foreign matter. According to the technical scheme of the invention, through color space conversion and image enhancement, different environmental conditions can be adapted. And dynamically calculating the risk index of the foreign matter according to the type and size of the foreign matter, the risk type and the environmental parameters. A risk assessment strategy is dynamically adjusted through a machine learning model, so that the system can better cope with various complex conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of power management, and more specifically, to a method and system for identifying foreign objects in power lines based on image processing. Background Art

[0002] Power lines are an important part of modern power systems, responsible for transmitting the electricity generated by power stations to various users. However, during operation, power lines may be affected by various foreign objects, such as plastic bags, branches, birds, etc. These foreign objects may cause safety accidents such as line short circuits and fires, seriously affecting the stability and safety of the power system. Therefore, timely detection and handling of foreign objects in the power line area is crucial to ensure the normal operation of the power system.

[0003] At present, the identification of foreign objects in power lines mainly relies on manual inspection methods. Although manual inspection can detect most foreign objects, it has problems such as low efficiency, high cost, and difficulty in covering large areas. Although traditional image processing methods can assist manual inspection, the recognition accuracy and robustness in complex environments still need to be improved.

[0004] In view of the problems existing in the prior art, the present invention proposes a method and system for identifying foreign objects in power lines based on image processing, aiming to improve the accuracy of foreign object identification and realize real-time monitoring of power line areas. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed to provide a method and system for identifying foreign objects in power lines based on image processing, aiming to improve the accuracy of foreign object identification and realize real-time monitoring of power line areas.

[0006] In a first aspect, the present invention provides a method for identifying foreign objects in power lines based on image processing, comprising: taking an image of the power line area by a camera installed in the power line area; performing color space conversion on the image; performing image enhancement on the image; analyzing the image by using a pre-trained deep learning model, extracting features in the image, and identifying foreign objects from the image based on the features in the image; calculating the risk index of the foreign object according to a preset strategy; and generating treatment measures for the foreign object according to the risk index of the foreign object.

[0007] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing performs color space conversion on the image, including: extracting the hue value, saturation value and brightness value of each pixel in the image; calculating the maximum brightness value, minimum brightness value and average brightness value of all pixels in the image; adjusting the hue value of each pixel in the image, and the adjusted hue value of the i-th pixel is ,in, is the hue value of the i-th pixel before adjustment, is the brightness value of the i-th pixel before adjustment, is the minimum brightness value of all pixels in the image, is the maximum brightness value of all pixels in the image; the brightness value of each pixel in the image is adjusted, wherein the brightness value of the i-th pixel after adjustment is ,in, is the average brightness value of all pixels in the image, , It is the preset curve adjustment coefficient.

[0008] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing, performing color space conversion on the image, further comprises: obtaining the light intensity value of the power line area; adjusting the saturation value of each pixel in the image, and the adjusted saturation value of the i-th pixel is ,in, is the saturation value of the i-th pixel before adjustment, is the light intensity value of the power line area, is the preset initial light intensity value, is the preset weight coefficient.

[0009] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing, before performing image enhancement on the image, the method also includes: detecting the noise type of the image; when the noise type of the image is Gaussian noise, obtaining a Gaussian filter, and adjusting the kernel and standard deviation of the Gaussian filter according to the neighborhood mean and variance of each pixel, and using the adjusted Gaussian filter to process the image; when the noise type of the image is salt and pepper noise, obtaining a median filter, and adjusting the filter window size of the median filter according to the level of the salt and pepper noise, and using the adjusted median filter to process the image.

[0010] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing adjusts the kernel and standard deviation of the Gaussian filter according to the neighborhood mean and variance of each pixel, including: detecting the gray value of each pixel in the image, calculating the average gray value of all pixels in the image; calculating the adjustment item of the Gaussian filter ,in, is the number of all pixels in the image, is the gray value of the jth pixel, is the average grayscale value of all pixels in the image; calculate the standard deviation of the Gaussian filter ,in, is the first adjustment coefficient; calculate the kernel of the Gaussian filter ,in, is the second adjustment coefficient.

[0011] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing performs image enhancement on the image, including: calculating the global contrast of each pixel in the image; calculating the local contrast of each pixel in the image according to a preset local area size; adjusting the pixel value of each pixel in the image, and the adjusted pixel value of the i-th pixel is ,in, is the pixel value of the i-th pixel before adjustment, is the preset weight coefficient, is the global contrast of the i-th pixel, is the local contrast of the i-th pixel, is a fixed constant, is the average pixel value of all pixels in the image.

[0012] Optionally, in the aforementioned method for identifying foreign objects in power lines based on image processing, the activation function used by the neural network is ,in, is the input of the neural network, is the adaptive slope parameter with an initial value of 1, is an adaptive bias parameter with an initial value of 0. During the training process of the neural network, the adaptive slope parameter and the adaptive bias parameters Make adjustments so that ,in, is the loss function of the neural network, is the learning rate of the neural network.

[0013] Optionally, the aforementioned method for identifying foreign objects in power lines based on image processing calculates the risk index of the foreign object according to a preset strategy, including: obtaining the type, size and position of the foreign object; analyzing the type of security incident caused by the foreign object according to the type of the foreign object; calculating the distance between the foreign object and the core area in the power line area according to the position of the foreign object; calculating the risk index of the foreign object according to the risk degree of the security incident to the power line area, the distance between the foreign object and the core area, and the size of the foreign object. ,in, Indicates the type of security event. Indicates the risk level of the security incident to the power line area, is the size of the foreign body, is the distance between the foreign body and the core area, is the maximum reference distance.

[0014] In a second aspect, the present invention provides a power line foreign object identification system based on image processing, comprising: an image capture module, which captures an image of the power line area by means of a camera installed in the power line area; a conversion module, which performs color space conversion on the image; an image enhancement module, which performs image enhancement on the image; a recognition module, which analyzes the image using a pre-trained deep learning model, extracts features from the image, and identifies foreign objects from the image based on the features in the image; a calculation module, which calculates a risk index of the foreign object according to a preset strategy; and a processing module, which generates treatment measures for the foreign object according to the risk index of the foreign object.

[0015] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects:

[0016] According to the technical solution of the present invention, through color space conversion and image enhancement, it is possible to adapt to different environmental conditions, such as different lighting, weather and background complexity, thereby improving the robustness of the system. The risk index of foreign matter is dynamically calculated by combining the type, size, risk type and environmental parameters of the foreign matter. The risk assessment strategy is dynamically adjusted through the machine learning model, which improves the accuracy and reliability of the assessment and enables the system to better cope with various complex situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 is a flow chart of a method for identifying foreign objects in a power line based on image processing according to an embodiment of the present application;

[0019] Figure 2 is a partial flow chart of a method for identifying foreign objects in a power line based on image processing according to an embodiment of the present application;

[0020] Figure 3 Another layout flow chart of the method for identifying foreign objects in power lines based on image processing according to an embodiment of the present application;

[0021] Figure 4is another partial flow chart of a method for identifying foreign objects in a power line based on image processing according to an embodiment of the present application;

[0022] Figure 5 A partial schematic diagram of a method for identifying foreign objects in a power line based on image processing according to an embodiment of the present application;

[0023] Figure 6 4 is a block diagram of a power line foreign object recognition system based on image processing according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] Some embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0025] like Figure 1 As shown, in one embodiment of the present invention, a method for identifying foreign objects in a power line based on image processing is provided, comprising:

[0026] Step S110: photographing an image of the power line area by using a camera installed in the power line area.

[0027] Step S120, performing color space conversion on the image.

[0028] In this embodiment, by performing color space conversion on the image, the contrast and clarity of the image are improved, making the foreign matter more obvious in the image, thereby improving the accuracy of recognition.

[0029] Step S130, performing image enhancement on the image.

[0030] In this embodiment, the image is enhanced to further improve the details and clarity of the image, reduce noise interference, and enhance the recognizability of foreign objects.

[0031] Step S140, using a pre-trained deep learning model to analyze the image, extract features in the image, and identify foreign matter from the image based on the features in the image.

[0032] In this embodiment, a pre-trained deep learning model is used to analyze the image, extract features from the image, and identify foreign objects based on the features. The deep learning model has powerful feature extraction capabilities and robustness, and can accurately identify foreign objects in complex backgrounds, significantly improving the accuracy of recognition.

[0033] Step S150, calculating the risk index of the foreign matter according to a preset strategy.

[0034] Step S160: generating treatment measures for the foreign matter according to the risk index of the foreign matter.

[0035] In this embodiment, the treatment measures for foreign objects are generated according to the risk index of foreign objects. Low-risk foreign objects are monitored through automatic notification and regular review, and high-risk foreign objects are immediately alarmed and maintenance personnel are dispatched to handle them, which improves work efficiency and reduces manpower investment.

[0036] According to the technical solution of this embodiment, through color space conversion and image enhancement, it is possible to adapt to different environmental conditions, such as different lighting, weather and background complexity, thereby improving the robustness of the system. The risk index of foreign matter is dynamically calculated by combining the type, size, risk type and environmental parameters of the foreign matter. The risk assessment strategy is dynamically adjusted through the machine learning model, which improves the accuracy and reliability of the assessment and enables the system to better cope with various complex situations.

[0037] like Figure 2 As shown, another embodiment of the present invention provides a method for identifying foreign objects in a power line based on image processing. Compared with the above-mentioned embodiment, the method for identifying foreign objects in a power line based on image processing in this embodiment includes: step S120:

[0038] Step S210, extracting the hue value, saturation value and brightness value of each pixel in the image.

[0039] Step S220, calculating the maximum brightness value, the minimum brightness value and the average brightness value of all pixels in the image.

[0040] Step S230, adjusting the hue value of each pixel in the image, the adjusted hue value of the i-th pixel ,in, is the hue value of the i-th pixel before adjustment, is the brightness value of the i-th pixel before adjustment, is the minimum brightness value of all pixels in the image, is the maximum brightness value of all pixels in the image.

[0041] The formula of this embodiment adjusts the hue value of each pixel in the image so that the change of the hue value is smoother, which helps to distinguish foreign objects from the background and improve the accuracy of recognition.

[0042] Step S240, adjusting the brightness value of each pixel in the image, the brightness value of the i-th pixel after adjustment is ,in, is the average brightness value of all pixels in the image, , It is the preset curve adjustment coefficient.

[0043] The formula of this embodiment adjusts the brightness value of each pixel in the image, so that the brightness distribution of the image is more reasonable, noise interference is reduced, and the recognizability of foreign objects is enhanced.

[0044] Step S250, obtaining the light intensity value of the power line area.

[0045] Step S260, adjusting the saturation value of each pixel in the image, the saturation value of the i-th pixel after adjustment is ,in, is the saturation value of the i-th pixel before adjustment, is the light intensity value in the power line area, is the preset initial light intensity value, is the preset weight coefficient.

[0046] The formula of this embodiment adjusts the saturation value of each pixel in the image, so that the image can still maintain good quality and clarity under different lighting conditions, thereby improving the adaptability and robustness of the system.

[0047] According to the technical solution of this embodiment, by extracting the hue value, saturation value and brightness value of each pixel in the image and accurately adjusting them, this embodiment can more effectively improve the quality of the image and enhance the contrast and clarity of the image.

[0048] like Figure 3 As shown, another embodiment of the present invention provides a method for identifying foreign objects in a power line based on image processing. Compared with the above-mentioned embodiment, the method for identifying foreign objects in a power line based on image processing in this embodiment, before step S130, further includes:

[0049] Step S310: detecting the noise type of the image.

[0050] Step S320, when the noise type of the image is Gaussian noise, obtain a Gaussian filter, adjust the kernel and standard deviation of the Gaussian filter according to the neighborhood mean and variance of each pixel, and use the adjusted Gaussian filter to process the image.

[0051] Among them, the kernel and standard deviation of the Gaussian filter are adjusted as follows:

[0052] (1) Detect the grayscale value of each pixel in the image and calculate the average grayscale value of all pixels in the image.

[0053] (2) Calculate the adjustment term of the Gaussian filter ,in, is the number of all pixels in the image, is the gray value of the jth pixel, is the average gray value of all pixels in the image.

[0054] (3) Calculate the standard deviation of the Gaussian filter ,in, is the first adjustment coefficient.

[0055] (4) Calculate the kernel of the Gaussian filter ,in, is the second adjustment coefficient.

[0056] In this embodiment, through the above dynamic adjustment, the Gaussian filter can better adapt to the noise distribution in the image, effectively remove Gaussian noise, and at the same time retain the detailed information of the image to avoid excessive blurring.

[0057] Step S330, when the noise type of the image is salt and pepper noise, a median filter is obtained, and the filter window size of the median filter is adjusted according to the level of the salt and pepper noise, and the image is processed using the adjusted median filter.

[0058] When the noise type of the image is salt and pepper noise, this embodiment dynamically adjusts the filter window size of the median filter according to the level of the salt and pepper noise, thereby ensuring the denoising effect while reducing the damage to the image details.

[0059] According to the technical solution of this embodiment, through targeted denoising processing, this embodiment can significantly reduce noise interference in the image and improve the clarity and contrast of the image. This embodiment focuses on retaining the detailed information of the image, avoiding information loss caused by excessive smoothing, making the image more realistic and natural, and high-quality images help to quickly and accurately identify foreign objects, realize real-time monitoring of the power line area, and timely discover potential foreign object threats.

[0060] like Figure 4 As shown, another embodiment of the present invention provides a method for identifying foreign objects in a power line based on image processing. Compared with the above-mentioned embodiment, the method for identifying foreign objects in a power line based on image processing in this embodiment includes step S130:

[0061] Step S410, calculating the global contrast of each pixel in the image.

[0062] Step S420, calculating the local contrast of each pixel in the image according to the preset local area size.

[0063] In this embodiment, by calculating the global contrast of each pixel in the image, this embodiment can comprehensively evaluate the overall contrast of the image, providing a basis for subsequent image enhancement. According to the preset local area size, the local contrast of each pixel in the image is calculated, so that the local details of the image are more prominent.

[0064] Step S430, adjusting the pixel value of each pixel in the image, the pixel value of the i-th pixel after adjustment is ,in, is the pixel value of the i-th pixel before adjustment, is the preset weight coefficient, is the global contrast of the ith pixel, is the local contrast of the ith pixel, is a fixed constant, is the average pixel value of all pixels in the image.

[0065] According to the technical solution of this embodiment, the above formula is conducive to improving the contrast and clarity of the image, making the foreign matter more obvious in the image, and improving the accuracy of foreign matter recognition.

[0066] Another embodiment of the present invention provides a method for identifying foreign objects in a power line based on image processing. Compared with the above-mentioned embodiment, the activation function used by the neural network in the method for identifying foreign objects in a power line based on image processing in this embodiment is: ,in, is the input of the neural network, is the adaptive slope parameter with an initial value of 1, is an adaptive bias parameter with an initial value of 0. During the training of the neural network, the adaptive slope parameter and adaptive bias parameters Make adjustments so that ,in, is the loss function of the neural network, is the learning rate of the neural network.

[0067] According to the technical solution of this embodiment, by adaptively adjusting the slope parameter and the bias parameter, the AS activation function can better adapt to different input data, enhance the nonlinear expression ability of the neural network, and improve the fitting ability and generalization ability of the model.

[0068] like Figure 5 As shown, another embodiment of the present invention provides a method for identifying foreign objects in a power line based on image processing. Compared with the above-mentioned embodiment, the method for identifying foreign objects in a power line based on image processing in this embodiment includes step S150:

[0069] Step S510, obtaining the type, size and position of the foreign matter.

[0070] In this embodiment, by obtaining the type, size and location of the foreign matter, this embodiment can fully understand the characteristics of the foreign matter and provide detailed data support for subsequent risk assessment.

[0071] Step S520: Analyze the type of security incident caused by the foreign object according to the type of the foreign object.

[0072] In this embodiment, the types of safety incidents that may be caused by foreign objects, such as fire, short circuit, etc., are analyzed according to the types of foreign objects, which helps to more accurately assess the risks of foreign objects.

[0073] Step S530, calculating the distance between the foreign object and the core area in the power line area according to the position of the foreign object.

[0074] In this embodiment, the distance between the foreign object and the core area in the power line area is calculated according to the location of the foreign object, and the influence of the location of the foreign object on the risk is taken into account, so that the risk assessment is more comprehensive and accurate.

[0075] Step S540, calculating the risk index of the foreign object according to the risk level of the security incident to the power line area, the distance between the foreign object and the core area, and the size of the foreign object ,in, Indicates the type of security event. Indicates the risk level of security incidents to the power line area. is the size of the foreign body, is the distance between the foreign body and the core area, is the maximum reference distance.

[0076] According to the technical solution of this embodiment, the risk index of foreign objects is calculated comprehensively based on the risk level of the security incident to the power line area, the distance between the foreign object and the core area, and the size of the foreign object, thereby ensuring the scientificity and rationality of the risk assessment. Based on the risk index, corresponding treatment measures or solutions can be generated.

[0077] like Figure 6 As shown, in one embodiment of the present invention, a power line foreign body identification system based on image processing is provided, comprising:

[0078] The image capturing module 610 captures an image of the power line area through a camera installed in the power line area.

[0079] The conversion module 620 performs color space conversion on the image.

[0080] In this embodiment, by performing color space conversion on the image, the contrast and clarity of the image are improved, making the foreign matter more obvious in the image, thereby improving the accuracy of recognition.

[0081] The image enhancement module 630 performs image enhancement on the image.

[0082] In this embodiment, the image is enhanced to further improve the details and clarity of the image, reduce noise interference, and enhance the recognizability of foreign objects.

[0083] The recognition module 640 uses a pre-trained deep learning model to analyze the image, extract features in the image, and recognize foreign objects from the image based on the features in the image.

[0084] In this embodiment, a pre-trained deep learning model is used to analyze the image, extract features from the image, and identify foreign objects based on the features. The deep learning model has powerful feature extraction capabilities and robustness, and can accurately identify foreign objects in complex backgrounds, significantly improving the accuracy of recognition.

[0085] The calculation module 650 calculates the risk index of the foreign matter according to a preset strategy.

[0086] The processing module 660 generates processing measures for the foreign matter according to the risk index of the foreign matter.

[0087] In this embodiment, the treatment measures for foreign objects are generated according to the risk index of foreign objects. Low-risk foreign objects are monitored through automatic notification and regular review, and high-risk foreign objects are immediately alarmed and maintenance personnel are dispatched to handle them, which improves work efficiency and reduces manpower investment.

[0088] According to the technical solution of this embodiment, through color space conversion and image enhancement, it is possible to adapt to different environmental conditions, such as different lighting, weather and background complexity, thereby improving the robustness of the system. The risk index of foreign matter is dynamically calculated by combining the type, size, risk type and environmental parameters of the foreign matter. The risk assessment strategy is dynamically adjusted through the machine learning model, which improves the accuracy and reliability of the assessment and enables the system to better cope with various complex situations.

[0089] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0090] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0091] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0092] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0093] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for identifying foreign objects in power lines based on image processing, characterized in that: include: Capturing an image of the power line area by a camera installed in the power line area; Performing color space conversion on the image; performing image enhancement on the image; Analyzing the image using a pre-trained deep learning model, extracting features in the image, and identifying foreign matter from the image based on the features in the image; Calculating the risk index of the foreign matter according to a preset strategy; According to the risk index of the foreign matter, a treatment measure for the foreign matter is generated.

2. The method for identifying foreign matter in a power line based on image processing according to claim 1, characterized in that: Performing color space conversion on the image, comprising: Extracting the hue value, saturation value and lightness value of each pixel in the image; Calculate the maximum brightness value, the minimum brightness value and the average brightness value of all pixels in the image; The hue value of each pixel in the image is adjusted, and the adjusted hue value of the i-th pixel is ,in, is the hue value of the i-th pixel before adjustment, is the brightness value of the i-th pixel before adjustment, is the minimum brightness value of all pixels in the image, is the maximum brightness value of all pixels in the image; The brightness value of each pixel in the image is adjusted, wherein the brightness value of the i-th pixel after adjustment is ,in, is the average brightness value of all pixels in the image, , It is the preset curve adjustment coefficient.

3. The method for identifying foreign matter in a power line based on image processing according to claim 2, characterized in that: Performing color space conversion on the image also includes: Obtaining a light intensity value of the power line area; The saturation value of each pixel in the image is adjusted, and the saturation value of the i-th pixel after adjustment is ,in, is the saturation value of the i-th pixel before adjustment, is the light intensity value of the power line area, is the preset initial light intensity value, is the preset weight coefficient.

4. The method for identifying foreign matter in a power line based on image processing according to claim 1, characterized in that: Before performing image enhancement on the image, the method further includes: detecting a noise type of the image; When the noise type of the image is Gaussian noise, obtain a Gaussian filter, adjust the kernel and standard deviation of the Gaussian filter according to the neighborhood mean and variance of each pixel, and use the adjusted Gaussian filter to process the image; When the noise type of the image is salt and pepper noise, a median filter is obtained, and a filter window size of the median filter is adjusted according to the level of the salt and pepper noise, and the image is processed using the adjusted median filter.

5. The method for identifying foreign matter in a power line based on image processing according to claim 4, characterized in that: The kernel and standard deviation of the Gaussian filter are adjusted according to the neighborhood mean and variance of each pixel, including: Detecting the grayscale value of each pixel in the image, and calculating the average grayscale value of all pixels in the image; Calculate the adjustment term of the Gaussian filter ,in, is the number of all pixels in the image, is the gray value of the jth pixel, is the average gray value of all pixels in the image; Calculate the standard deviation of the Gaussian filter ,in, is the first adjustment coefficient; Calculate the kernel of the Gaussian filter ,in, is the second adjustment coefficient.

6. The method for identifying foreign matter in a power line based on image processing according to claim 1, characterized in that: Performing image enhancement on the image, comprising: Calculating a global contrast for each pixel in the image; Calculating the local contrast of each pixel in the image according to a preset local area size; The pixel value of each pixel in the image is adjusted, and the pixel value of the i-th pixel after adjustment is ,in, is the pixel value of the i-th pixel before adjustment, is the preset weight coefficient, is the global contrast of the i-th pixel, is the local contrast of the i-th pixel, is a fixed constant, is the average pixel value of all pixels in the image.

7. The method for identifying foreign matter in a power line based on image processing according to claim 1, characterized in that: The activation function used by the neural network is ,in, is the input of the neural network, is the adaptive slope parameter with an initial value of 1, is an adaptive bias parameter with an initial value of 0. During the training process of the neural network, the adaptive slope parameter and the adaptive bias parameters Make adjustments so that ,in, is the loss function of the neural network, is the learning rate of the neural network.

8. The method for identifying foreign matter in a power line based on image processing according to claim 1, characterized in that: According to the preset strategy, the risk index of the foreign matter is calculated, including: Obtaining the type, size and location of the foreign matter; Analyze the type of security incident caused by the foreign object according to the type of the foreign object; Calculating the distance between the foreign object and the core area in the power line area according to the position of the foreign object; Calculate the risk index of the foreign object according to the risk level of the security incident to the power line area, the distance between the foreign object and the core area, and the size of the foreign object. ,in, Indicates the type of security event. Indicates the risk level of the security incident to the power line area, is the size of the foreign body, is the distance between the foreign body and the core area, is the maximum reference distance.

9. The power line foreign body identification system based on image processing is characterized by: include: An image capturing module, which captures an image of the power line area through a camera installed in the power line area; A conversion module, performing color space conversion on the image; An image enhancement module, performing image enhancement on the image; a recognition module, analyzing the image using a pre-trained deep learning model, extracting features from the image, and identifying foreign matter from the image based on the features in the image; A calculation module, which calculates the risk index of the foreign matter according to a preset strategy; The processing module generates processing measures for the foreign matter according to the risk index of the foreign matter.

Images