Power transmission line image generation method, system and device based on physical law and medium

Abstract

The invention relates to the technical field of power transmission line image generation, and discloses a power transmission line image generation method, system and device based on a physical law, and a medium, and the method comprises the steps: obtaining a first image of a target power transmission line, and carrying out the first marking operation of the first image; presetting a first factor set, and establishing a relaxation index based on the first factor set; obtaining a stress form graph of the power transmission line according to the relaxation index; establishing a first generative adversarial network; and generating a target power transmission line image according to the first generative adversarial network. According to the method, the accuracy and authenticity of power transmission line image generation are improved, and the physical state of the power transmission line can be reflected more accurately. Factors influencing the looseness of the power transmission line are quantified through a preset factor set and a looseness index, and accurate data are provided for image generation. The image is generated by using the first generative adversarial network, so that style diversification is realized, meanwhile, the stress characteristics of the power transmission line are reserved, and the practicability and attractiveness of the image are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission line image generation, and in particular to a power transmission line image generation method, system, device and medium based on physical laws. BACKGROUND

[0002] The stable operation of the power system is crucial to the protection of social production and life, and the operation state of the power transmission line, as a key link of power transmission, is directly related to the stability of the entire power system and the reliability of power supply. In recent years, the analysis method based on the visual image of the power transmission line has gradually become the mainstream. Through the means of unmanned aerial vehicle inspection, satellite detection and other means, the image data of the power transmission line is obtained, and the computer vision technology and deep learning algorithm are used to analyze the image.

[0003] However, in the research and practice of power transmission line image recognition, there are problems of poor sample consistency and difficulty in covering various abnormal situations. This will affect the generalization ability and recognition accuracy of the model. At the same time, the environment where the power transmission line is located is complex and changeable, including different weather conditions, different temperature and humidity, etc. These factors will interfere with image recognition. Researchers summarize the above problems as a small sample problem. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a power transmission line image generation method, system, device and medium based on physical laws, which can solve the problems of poor sample consistency and difficulty in covering various abnormal situations, and improve the generalization ability and recognition accuracy of the model.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the present application provides a power transmission line image generation method based on physical laws, comprising:

[0008] obtaining a first image of a target power transmission line, and performing a first labeling operation on the first image;

[0009] The first labeling operation is used for relaxation degree labeling and weather condition labeling of the first image;

[0010] a first factor set is preset, and a relaxation degree index based on the first factor set is established;

[0011] The relaxation degree index is used to quantify the influence degree of the influencing factors in the first factor set on the relaxation degree of the power transmission line;

[0012] According to the relaxation degree index, a stress form diagram of the power transmission line is obtained;

[0013] The first generative adversarial network is established, and a generator of the first generative adversarial network is inputted with different stylized power line images and the stress state diagram;

[0014] The different stylized power line images are obtained through pre-training operations;

[0015] The target power line image is generated according to the first generative adversarial network.

[0016] As a preferred scheme of the power line image generation method based on physical laws provided in the application, the pre-training operation comprises:

[0017] Different stylized historical power line images are obtained, and the historical power line images are classified;

[0018] The classified historical power line images are preprocessed;

[0019] A second recurrent generative adversarial network is established, and the different stylized power line images are generated through the second recurrent generative adversarial network.

[0020] As a preferred scheme of the power line image generation method based on physical laws provided in the application, the establishment of the relaxation index based on the first factor set comprises:

[0021] A first factor set is determined, and the first factor set comprises a plurality of influence factors related to the relaxation degree of the power line;

[0022] A power line relaxation degree influence logic based on each influence factor is established;

[0023] The power line relaxation degree influence logic of each influence factor is quantified, and a relaxation index is established.

[0024] The preferred scheme can more accurately reflect the actual relaxation state of the power line, and provide more accurate data support for subsequent image generation. At the same time, by establishing the relaxation index, the influence degree of different influence factors on the relaxation degree of the power line can be intuitively compared, which helps relevant personnel better understand and analyze the stress state of the power line, and provides a strong basis for the maintenance and management of the power line.

[0025] As a preferred scheme of the power line image generation method based on physical laws provided in the application, the stress state diagram of the power line is obtained according to the relaxation index, which comprises:

[0026] The historical values of the influence factors corresponding to the relaxation index are obtained;

[0027] The stress change diagram of the power line under the corresponding relaxation index is drawn according to the historical values;

[0028] traversing historical values of influencing factors in the first factor set;

[0029] all the force change diagrams obtained after the traversal are recorded as force state diagrams.

[0030] As a preferred scheme of the power transmission line image generation method based on physical laws provided in the present application, the second cycle generation adversarial network comprises:

[0031] a generator that maps stylization in domain A to stylization in domain B;

[0032] a generator that maps stylization in domain B back to stylization in domain A, and

[0033] a discriminator that judges real images and generated images.

[0034] As a preferred scheme of the power transmission line image generation method based on physical laws provided in the present application, the method further comprises:

[0035] The historical values of the influencing factors comprise historical values obtained according to a first image of the target power transmission line, or virtual historical values generated according to a preset value logic.

[0036] As a preferred scheme of the power transmission line image generation method based on physical laws provided in the present application, the method further comprises:

[0037] The first factor set comprises temperature, humidity, wind speed, and wind direction.

[0038] The different stylizations comprise sunny days, rainy days, snowy days, and foggy days.

[0039] In a second aspect, the present application provides a power transmission line image generation system based on physical laws, comprising:

[0040] a labeling module configured to obtain a first image of a target power transmission line and perform a first labeling operation on the first image;

[0041] The first labeling operation is configured to perform relaxation degree labeling and weather condition labeling on the first image.

[0042] an index acquisition module configured to preset a first factor set and establish a relaxation index based on the first factor set;

[0043] The relaxation index is configured to quantify the influence of influencing factors in the first factor set on the relaxation degree of the power transmission line.

[0044] a force determination module configured to obtain a force state diagram of the power transmission line according to the relaxation index.

[0045] a network establishment module, configured to establish a first generative adversarial network, wherein an input of a generator of the first generative adversarial network is a different stylized power line image and the stress state diagram;

[0046] The different stylized power line image is obtained through a pre-training operation.

[0047] An image generation module is configured to generate a target power line image according to the first generative adversarial network.

[0048] In a third aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described above when executing the computer program.

[0049] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method described above.

[0050] Compared with the prior art, the present application has the following beneficial effects: the present application provides a power line image generation method based on physical laws, obtains a first image of a target power line, and performs a first labeling operation on the first image; a first factor set is preset, a relaxation index based on the first factor set is established, a stress state diagram of the power line is obtained according to the relaxation index, a first generative adversarial network is established, an input of a generator of the first generative adversarial network is a different stylized power line image and the stress state diagram, and a target power line image is generated according to the first generative adversarial network. This method not only improves the accuracy and authenticity of power line image generation, but also makes the generated image better reflect the actual physical state of the power line. Through the preset first factor set and the relaxation index, the present application can quantify the influence of various factors on the relaxation degree of the power line, thereby providing more accurate data support for subsequent image generation. In addition, the first generative adversarial network is used for image generation, which makes the generated image more diversified in style while retaining the stress state characteristics of the power line, improving the practicality and aesthetic value of the image. Therefore, the present application has wide application prospects in the field of power line image generation. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0052] Figure 1A method flowchart of a power transmission line image generation method based on physical laws is provided for an embodiment of the present application.

[0053] Figure 2 An internal structure diagram of an electronic device of a power transmission line image generation method based on physical laws is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0054] To make the above objectives, features and advantages of the present application more apparent, more comprehensible, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor should fall within the scope of protection of the present application.

[0055] Embodiment 1, refer to Figures 1-2 For the first embodiment of the present application, the embodiment provides a power transmission line image generation method based on physical laws, comprising:

[0056] In the prior related art, there are some problems, for example: the training of the image recognition model needs a large amount of labeled data, but in actual application, it is difficult to obtain power transmission line image data covering all cases, which limits the generalization ability of the model. In addition, the slack degree of the power transmission line is affected by many factors, such as temperature, humidity, wind speed and wind direction, etc., and the performance of these factors in the image is complex and changeable, which increases the difficulty of image recognition.

[0057] The present application provides a method that can effectively solve the above-mentioned problems, and the following will be described in detail how to realize the power transmission line image generation method based on physical laws by combining multiple embodiments;

[0058] Figure 1 A method flowchart of a power transmission line image generation method based on physical laws is shown, comprising:

[0059] S101, acquiring a first image of a target power transmission line, and performing a first labeling operation on the first image;

[0060] It should be noted that in order to generate power transmission line images according to physical laws, some historical image data about power transmission lines need to be obtained, and stable generation rules need to be found from the historical image data, so as to realize power transmission line image generation based on physical laws.

[0061] In an optional embodiment, the morphological features, colors, textures, etc. of the power transmission line in the historical image data can be analyzed in combination with environmental factors such as temperature, humidity, wind speed and wind direction, etc. to analyze the change rule of the power transmission line under different conditions.

[0062] In an optional embodiment, by learning and understanding these laws, a power line image generation model based on physical laws can be constructed. This model can automatically generate power line images that conform to actual conditions according to given physical parameters and environmental conditions, thereby greatly reducing the dependence on labeled data and improving the accuracy and generalization ability of image recognition.

[0063] In an optional embodiment, the way to obtain the image of the target power line can be obtained by a drone, or by satellite remote sensing technology, or by other image acquisition devices on the ground.

[0064] In the embodiments of the present application, the first labeling operation is used to label the relaxation degree and the weather condition of the first image.

[0065] In an optional embodiment, the first labeling operation can use a deep learning algorithm for automatic labeling. The deep learning algorithm can automatically identify and extract the key features of the power line, including the relaxation degree and the weather condition, by learning a large amount of historical image data, thereby achieving efficient labeling. This way not only improves the accuracy and efficiency of labeling, but also reduces the subjectivity and inconsistency of manual labeling.

[0066] In an optional embodiment, augmented reality (AR) technology can also be used to assist the labeling process. Through AR technology, labeling information can be directly superimposed in the actual scene, making it easier for labeling personnel to understand the state of the power line and the environmental conditions. This way not only improves the accuracy of labeling, but also improves the work efficiency and safety of labeling personnel, as they do not need to be close to the high-voltage power line to complete the labeling task.

[0067] In the embodiments of the present application, the relaxation degree labeling in this step is an operation of labeling different physical conditions, which aims to provide a data basis for the establishment of the following relaxation index. By accurately labeling the relaxation degree, the relaxation state of the power line can be quantified, providing key information for subsequent analysis and image generation.

[0068] In the embodiments of the present application, weather condition labeling is to record the weather conditions at the time of image collection, including different weather types such as sunny, rainy, snowy, and foggy.

[0069] Specifically, the basic image of the power line, i.e. the first image, is obtained using a drone, and the physical conditions at the time (temperature, humidity, wind speed, wind direction) are collected, i.e. the relaxation degree labeling. The physical conditions are collected using sensors for studying the relaxation degree changes of the power line under different physical conditions. The collected basic power line image is labeled with the relaxation degree.

[0070] It should be noted that obtaining the first image of the target power transmission line and performing the first labeling operation on the first image provide basic data and labeling information for subsequent steps. By labeling the relaxation degree and weather conditions of the first image, the establishment of the relaxation degree index, the acquisition of the stress form diagram, and the training of the generative adversarial network in the subsequent steps can be more accurate and efficient. At the same time, these labeling information also provides an important basis for understanding and analyzing the physical state of the power transmission line, which helps to further improve the accuracy and authenticity of image generation. On this basis, the power transmission line image generation method based on physical laws proposed in the present application can make full use of these labeling information and physical laws to generate power transmission line images that conform to the actual situation, providing strong support for the maintenance and management of power transmission lines.

[0071] S102, a first factor set is preset, and a relaxation degree index based on the first factor set is established;

[0072] In the embodiment of the present application, the relaxation degree index is used to quantify the influence of the influencing factors in the first factor set on the relaxation degree of the power transmission line;

[0073] In the embodiment of the present application, the first factor set includes temperature, humidity, wind speed, and wind direction;

[0074] In an optional embodiment, establishing a relaxation degree index based on the first factor set can be achieved by statistically analyzing historical data to determine the specific influence of different factors on the relaxation degree of the power transmission line in different value ranges. For example, the change in the relaxation degree of the power transmission line when the temperature rises or falls by a certain amplitude can be analyzed. Similarly, the influence of humidity, wind speed, and wind direction on the relaxation degree can also be studied. By quantifying these influences, a relaxation degree index can be established, which can comprehensively reflect the combined effect of multiple factors on the relaxation degree of the power transmission line, providing more accurate data support for subsequent image generation.

[0075] In an optional embodiment, machine learning algorithms can be used to fit historical data to obtain a mathematical relationship model between factors and relaxation degree, and then calculate the relaxation degree index based on the model. This method not only improves the accuracy of index calculation, but also enables the index to be continuously updated and optimized as new data is added, thereby better adapting to changes in actual situations.

[0076] However, in the embodiment of the present application, establishing a relaxation degree index based on the first factor set includes:

[0077] determining the first factor set, which contains several influencing factors related to the relaxation degree of the power transmission line;

[0078] establishing a power transmission line relaxation degree influence logic based on each influencing factor;

[0079] The influence logic of each influencing factor on the slackness of the power transmission line is quantified, and a slackness index is established.

[0080] It should be noted that the present application focuses on external physical factors, including temperature, humidity, wind speed, and wind direction. The slackness index is defined using physical factors to simulate the influence of these factors on the ground wire of the power transmission line and draw a force pattern diagram of the ground wire under specific conditions to estimate different fault levels.

[0081] Specifically, the establishment of the influence logic of each influencing factor on the slackness of the power transmission line includes:

[0082] The influence of temperature on the ground wire of the power transmission line is analyzed. The thermal expansion and contraction of temperature leads to a change in the length of the power transmission line, and the length change ΔL is related to the initial length L0, the temperature change ΔT, and the linear expansion coefficient α of the material. The calculation formula is: ΔL = L0αΔT, which indicates that when the temperature rises, the power transmission line will lengthen due to thermal expansion; conversely, when the temperature drops, the power transmission line will shorten due to cold contraction. This change in length will affect the mechanical properties of the power transmission line such as tension and sag, and further affect the stability and safety of power transmission;

[0083] The influence of humidity on the ground wire of the power transmission line is analyzed. When humidity increases, the surface of the power transmission line will adsorb water, resulting in an increase in the weight of the power transmission line, and the power transmission line will sag. The weight increase ΔW is related to the initial weight W0, and the total weight of the power transmission line changes to W 总 = ΔW + W0, resulting in an increase in the slackness of the power transmission line;

[0084] The influence of wind speed and wind direction on the ground wire of the power transmission line is analyzed. When the wind speed increases, the power transmission line will be subjected to wind load, resulting in an increase in the deflection of the power transmission line. Where ρ is the air density, v is the wind speed, C d is the drag coefficient, and A is the windward area of the power transmission line, then the wind load is When the wind direction changes, the deflection direction of the power transmission line changes, which is quantified by the wind direction angle θ;

[0085] In an alternative embodiment, the influence logic of each influencing factor on the slackness of the power transmission line is quantified, including:

[0086] According to the influence of temperature, humidity, wind speed, and wind direction on the ground wire of the power transmission line, the slackness index is defined. This index reflects the slack state of the power transmission line under different environmental conditions through comprehensive consideration of these four physical factors in a quantitative value;

[0087] According to the defined relaxation index, comprehensively combining various physical parameters of the transmission line conductor and ground wire, a stress pattern diagram of the transmission line conductor and ground wire is drawn, that is, the stress changes of the transmission line under different conditions are artificially drawn according to the stress conditions of the transmission line, and the shape change of the transmission line after the relaxation index defined by the physical information is changed is derived.

[0088] In an optional embodiment, according to the influence of temperature, humidity, wind speed and wind direction on the transmission line conductor and ground wire, the specific steps of defining the relaxation index can be:

[0089] Firstly, the direct and indirect effects of the four influencing factors of temperature, humidity, wind speed and wind direction on the relaxation degree of the transmission line are determined. For temperature, considering the physical phenomenon of thermal expansion and cold contraction, a relaxation coefficient based on temperature change can be set, which increases with the increase of temperature and decreases with the decrease of temperature. For humidity, since the increase of humidity will cause the adsorption of water on the surface of the transmission line, thereby increasing the weight of the transmission line, a relaxation increment proportional to humidity can be set. For wind speed and wind direction, based on the calculation formula of wind load, a relaxation adjustment coefficient proportional to the square of wind speed and related to wind direction angle can be set.

[0090] Then, the relaxation coefficients or increments of the above four influencing factors are weighted and summed to obtain a comprehensive relaxation index. The determination of the weight can be based on statistical analysis of historical data or obtained by expert scoring and other methods. In this way, the relaxation index can comprehensively reflect the combined influence of temperature, humidity, wind speed and wind direction on the relaxation degree of the transmission line.

[0091] Finally, according to the obtained relaxation index, combining the physical parameters of the transmission line (such as length, diameter, material properties, etc.), the stress pattern diagram of the transmission line under different relaxation indexes is drawn by using mechanical principles and related calculation tools. These stress pattern diagrams can intuitively show the relaxation state and shape change of the transmission line under different environmental conditions, providing important reference for subsequent image generation.

[0092] In an optional embodiment, other methods besides weighted summation can also be used to comprehensively calculate the relaxation index. For example, machine learning algorithms such as neural networks, support vector machines, etc. can be used to nonlinearly fit the four influencing factors of temperature, humidity, wind speed and wind direction, to obtain a relaxation index that can more accurately reflect the relaxation degree of the transmission line.

[0093] It should be noted that the preset first factor set and the establishment of the slackness index based on the first factor set provide key data support and basis for the training of the generated adversarial network and image generation in subsequent steps. Through accurate calculation and quantization of the slackness index, the generated power line image can be more consistent with the actual situation, improving the authenticity and accuracy of the image. At the same time, the slackness index also provides an important reference for understanding and analyzing the physical state of the power line, which helps to further optimize the image generation method and improve the performance of image recognition. On this basis, the power line image generation method based on physical laws proposed in the present application can fully utilize the slackness index and physical laws to generate high-quality power line images, providing more powerful support for the maintenance and management of power lines.

[0094] S103, obtaining a stress form diagram of the power line according to the slackness index;

[0095] In the embodiment of the present application, obtaining the stress form diagram of the power line according to the slackness index comprises:

[0096] obtaining historical values of the influence factors corresponding to the slackness index;

[0097] drawing a stress change diagram of the power line under the corresponding slackness index according to the historical values;

[0098] traversing the historical values of the influence factors in the first factor set;

[0099] all the stress change diagrams obtained after traversal are recorded as the stress form diagram.

[0100] In an optional embodiment, the historical values can be directly obtained through the historical data of the target power line, and the stress change diagram can be drawn by professional mechanical analysis software or tools. These software or tools can automatically calculate the stress of the power line according to the input slackness index and physical parameters of the power line, and generate the corresponding stress change diagram. These stress change diagrams can intuitively show the form changes of the power line under different slackness indexes, including the tension distribution of the power line, the sag change, etc., providing important reference for subsequent image generation.

[0101] It should be noted that after traversing the historical values of all influence factors in the first factor set, the present application will obtain a series of stress change diagrams. These stress change diagrams cover the stress conditions of the power line under different temperatures, humidities, wind speeds and wind directions. Integrating these stress change diagrams together, the stress form diagram of the power line can be obtained. The stress form diagram can comprehensively reflect the stress state and form change of the power line under different environmental conditions, providing key data support for subsequent image generation.

[0102] In the embodiment of the present application, obtaining the stress form diagram of the power line according to the slackness index further comprises:

[0103] The historical values of the influencing factors include historical values obtained according to the first image of the target transmission line, or virtual historical values generated according to preset value logic.

[0104] In an optional embodiment, the virtual historical values generated according to the preset value logic can be generated by simulating the variation range under different physical conditions. For example, a series of value combinations of temperature, humidity, wind speed and wind direction can be set, which cover various environmental conditions that can actually be encountered. Then, based on these virtual historical values, the mechanical principles and related calculation tools can be used to draw the force variation diagram of the transmission line under these virtual conditions.

[0105] It should be noted that this method can expand the coverage of the force pattern diagram, making it more comprehensive and accurate, and providing more abundant reference for subsequent image generation. At the same time, by comparing the force pattern diagrams generated by the actual historical values and the virtual historical values, the calculation method of the slack index and the drawing process of the force pattern diagram can be further verified and optimized, thereby improving the accuracy and reliability of the entire image generation method.

[0106] It should also be noted that obtaining the force pattern diagram of the transmission line according to the slack index can provide more abundant and diverse training samples for the training of the generative adversarial network. The force pattern diagram shows the morphological changes of the transmission line under different slack indexes, which reflect the real physical state of the transmission line. Inputting these force pattern diagrams as training samples into the generative adversarial network can make the network learn the complex relationship between the morphological changes of the transmission line and the slack index. During the training process, the generative adversarial network will continuously adjust its internal parameters to generate transmission line images that are more consistent with the actual situation. In this way, in the subsequent image generation stage, only the specific slack index needs to be input, and the generative adversarial network can quickly generate the corresponding transmission line image, greatly improving the efficiency and accuracy of image generation. At the same time, since the force pattern diagram covers the force conditions of the transmission line under different physical conditions, the generated transmission line image will also have better generalization ability and be able to adapt to more complex environmental conditions.

[0107] S104, a first generative adversarial network is established, and the generator of the first generative adversarial network inputs different stylized transmission line images and force pattern diagrams;

[0108] In the embodiments of the present application, the different stylized transmission line images are obtained through a pre-training operation.

[0109] In the embodiments of the present application, the different stylization includes sunny day, rainy day, snowy day and foggy day.

[0110] In the embodiments of the present application, the pre-training operation includes:

[0111] obtain historical transmission line images of different stylizations, and classify the historical transmission line images;

[0112] preprocess the classified historical transmission line images;

[0113] establish a second cycle generative adversarial network, and generate transmission line images of different stylizations through the second cycle generative adversarial network.

[0114] In the embodiments of the present application, the second cycle generative adversarial network comprises:

[0115] a generator for mapping the stylization in domain A to the stylization in domain B;

[0116] a generator for mapping the stylization in domain B back to the stylization in domain A, and

[0117] a discriminator for judging real images and generated images.

[0118] Specifically, the specific steps for establishing the second cycle generative adversarial network are as follows:

[0119] transmission line pictures of different styles (sunny day, rainy day, snowy day, and foggy day) are classified and placed as an original data set;

[0120] The original data set is preprocessed, including reading data, converting image data into tensors, and performing normalization processing to adjust the numerical range of image data to a specific interval, so that the data has better consistency and stability to meet the requirements of cycle generative adversarial network (CycleGAN) training;

[0121] In an optional embodiment, the preprocessed data set is input into the model for training using the cycle generative adversarial network (CycleGAN), and the model parameters are adjusted so that the model can learn the feature representation of different style images, thereby generating transmission line images of different stylizations from a single picture;

[0122] In an optional embodiment, the cycle generative adversarial network (CycleGAN) for training and generating transmission line images of different stylizations is a deep learning model for unsupervised image-to-image conversion, which includes a generator for mapping domain A to domain B, a generator for mapping domain B back to domain A, and a discriminator for judging real images and generated images.

[0123] In an optional embodiment, during the training process, the power line images of different styles are saved according to certain rules. This is convenient for managing and viewing the training results, and is also conducive to further analysis and application of the generated images. At the same time, the generator is trained using a CycleGAN, and by continuously adjusting the parameters of the generator, it can generate power line images with different styles according to the input images.

[0124] It should be noted that the single power line picture to be processed is input into the trained model, and the model will perform style transfer on the input picture according to the different style features learned during the training process using feature mapping and image generation algorithms. The model will first extract features from the input picture, then match and map these features with the learned different style features, and through a series of calculations and processing, generate a power line image of the corresponding style.

[0125] Specifically, the steps of establishing the first generative adversarial network include:

[0126] A conditional generative adversarial network (CGAN) is constructed, including a generator and a discriminator. The generator is responsible for generating power line images, and the discriminator is used to judge whether the generated images are real. A conditional embedding layer is added to the generator, and the slackness index defined by the physical information is embedded into the generator as a conditional label.

[0127] In an optional embodiment, the stylized power line image is preprocessed, including image loading, grayscale conversion, edge detection, normalization, etc.

[0128] The image is loaded and converted into a single-channel grayscale image, which helps to simplify subsequent image processing. Then, Gaussian blur is used to smooth the image. Gaussian blur can effectively reduce noise in the image by weighting and averaging each pixel point and its surrounding pixel points, making the image smoother and providing a better image basis for subsequent edge detection operations.

[0129] The Canny algorithm is used for edge detection, and a low threshold is intentionally set to detect fine edges. As a key element in the image, the power line has relatively thin lines. By setting a low threshold, the edges of the power line can be detected more accurately. After detecting the edges of the power line, they are marked red to clearly indicate the position and shape of the CGAN power line in the image, so that it pays more attention to the generation of the power line during the training process, thereby improving the accuracy and clarity of the power line in the generated image.

[0130] The processed image is converted into a tensor and normalized, and the numerical range of the image data is adjusted to a specific interval, so that the data has better consistency and stability to meet the requirements of conditional generative adversarial network (CGAN) training;

[0131] In an optional embodiment, the slackness index obtained after simulating the stress state of the power line ground wire according to physical laws and the preprocessed stylized power line image are used as input information of the generator, and the generator generates a power line image according to the information;

[0132] It should be noted that the discriminator compares and judges the image generated by the generator with the real power line image in detail, and feeds back the judgment result to the generator as the improvement direction. The discriminator will analyze from various aspects of the image, and through continuous iterative training, the generator will gradually adjust its parameters according to the feedback information of the discriminator, so that the generated image is more realistic in all aspects, provides high-quality image data for the monitoring and maintenance of the power line, and improves the safety and stability of the power system.

[0133] S105, generating a target power line image according to the first generative adversarial network.

[0134] In summary, the present application provides a power line image generation method based on physical laws, obtains a first image of a target power line, and performs a first labeling operation on the first image; a first factor set is preset, and a slackness index based on the first factor set is established; a stress state diagram of the power line is obtained according to the slackness index; a first generative adversarial network is established, and the input of the generator of the first generative adversarial network is a power line image of different styles and the stress state diagram; and a target power line image is generated according to the first generative adversarial network. This method not only improves the accuracy and authenticity of the generated power line image, but also makes the generated image better reflect the actual physical state of the power line. Through the preset first factor set and the slackness index, the present application can quantify the influence of various factors on the slackness of the power line, thereby providing more accurate data support for subsequent image generation. In addition, the first generative adversarial network is used for image generation, which makes the generated image more diverse in style while retaining the stress state characteristics of the power line, improving the practicality and aesthetic value of the image. Therefore, the present application has wide application prospects in the field of power line image generation.

[0135] In a preferred embodiment, when generating target transmission line images according to the first generative adversarial network, the specific steps can be refined as follows: first, the pre-processed stylized transmission line image and the stress pattern are input into the generator. The generator will gradually construct the image of the transmission line based on these input information, combined with its internal learned feature representation and generation rules. In this process, the generator will pay special attention to the shape, position of the transmission line and the coordination with the surrounding environment to ensure that the generated image not only conforms to the physical laws but also has good visual effects. Then, the discriminator will make a detailed judgment on the image output by the generator. It will consider from the authenticity, clarity, accuracy of the transmission line and other aspects, and feed back the judgment result to the generator. After receiving the feedback from the discriminator, the generator will further adjust its parameters and generation strategy according to this information, so as to generate more realistic and high-quality transmission line images in the subsequent iterations. Through continuous iterative training and feedback adjustment, the generator can finally generate target transmission line images that conform to the physical laws and have good visual effects. These images can provide strong support for the monitoring and maintenance of the power system, helping workers to more accurately understand the state of the transmission line, so as to take corresponding measures in time and ensure the safety and stability of the power system.

[0136] In embodiment 3, a transmission line image generation system based on physical laws is also provided, which includes:

[0137] The labeling module is used to obtain the first image of the target transmission line and perform a first labeling operation on the first image.

[0138] The first labeling operation is used to label the relaxation degree and weather conditions of the first image.

[0139] The index acquisition module is used to pre-set a first factor set and establish a relaxation index based on the first factor set.

[0140] The relaxation index is used to quantify the influence of the influencing factors in the first factor set on the relaxation degree of the transmission line.

[0141] The stress determination module is used to obtain the stress pattern of the transmission line according to the relaxation index.

[0142] The network establishment module is used to establish a first generative adversarial network, and the generator of the first generative adversarial network inputs different stylized transmission line images and stress patterns.

[0143] The different stylized transmission line images are obtained through pre-training operation.

[0144] The image generation module is used to generate target transmission line images according to the first generative adversarial network.

[0145] The above each unit module can be embedded in or independent of the processor in the electronic device in hardware form, or can be stored in the memory in the electronic device in software form, so that the processor calls and executes the operations corresponding to the above each module.

[0146] The embodiment also provides an electronic device, which can be a terminal, and an internal structure diagram of the electronic device can be as shown in Figure 2 The electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the electronic device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The computer program is executed by the processor to implement a power transmission line image generation method based on physical laws. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen. The input device of the electronic device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0147] The embodiment also provides a computer readable storage medium having a computer program stored thereon, and the computer program is executed by a processor to implement the following steps:

[0148] obtaining a first image of a target power transmission line, and performing a first labeling operation on the first image;

[0149] The first labeling operation is used to label the relaxation degree and weather conditions of the first image;

[0150] A first factor set is preset, and a relaxation degree index based on the first factor set is established;

[0151] The relaxation degree index is used to quantify the influence degree of the influencing factors in the first factor set on the relaxation degree of the power transmission line;

[0152] A stress form diagram of the power transmission line is obtained according to the relaxation degree index;

[0153] A first generative adversarial network is established, and the input of the generator of the first generative adversarial network is a power transmission line image of different styles and the stress form diagram;

[0154] The power transmission line image of different styles is obtained through a pre-training operation;

[0155] The target power transmission line image is generated according to the first generative adversarial network.

[0156] It should be noted that the above examples are only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced without departing from the spirit and scope of the present application, and all these modifications and equivalents should be included in the scope of the claims of the present application.

[0157] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take a form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. In addition, the present application can take a form of a computer program product implemented on one or more computer-usable storage media (including but not limited to a magnetic disk storage, a CD-ROM, an optical storage, etc.) containing computer-usable program codes. The solutions in the embodiments of the present application can be implemented in various computer languages.

[0158] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0159] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatus, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0160] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that carries out the functions specified in one or more flows and / or blocks.

[0161] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those skilled in the art once they learn of the basic inventive concepts. Therefore, the appended claims are intended to encompass within their scope all such variations and modifications as are included within the scope of the application.

[0162] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for generating a transmission line image based on physical laws, characterized in that: include: Acquire a first image of a target transmission line, and perform a first labeling operation on the first image; The first labeling operation is used to label the first image with a relaxation degree and a weather condition; Presetting a first factor set and establishing a slackness index based on the first factor set; The slack index is used to quantify the degree of influence of the influencing factors in the first factor set on the slack degree of the transmission line; Obtaining a force morphology diagram of the transmission line according to the slack index; Establishing a first generative adversarial network, wherein a generator input of the first generative adversarial network is a transmission line image of different stylizations and the force shape diagram; The different stylized power line images are obtained through a pre-training operation; A target power line image is generated according to the first generative adversarial network.

2. The method for generating a power line image based on physical laws according to claim 1, wherein: The pre-training operation includes: Acquiring historical transmission line images of different stylizations and classifying the historical transmission line images; Preprocessing the classified historical transmission line images; A second cyclic generative adversarial network is established to generate power line images of different styles through the second cyclic generative adversarial network.

3. The method for generating a power line image based on physical laws according to claim 2, wherein: The establishing of the slackness index based on the first factor set includes: determining a first factor set, wherein the first factor set includes a plurality of influencing factors related to the degree of slack of the transmission line; Establish the transmission line slack impact logic based on each influencing factor; The transmission line slack influence logic of each influencing factor is quantified, and a slack index is established.

4. The method for generating a power line image based on physical laws according to claim 3, wherein: The step of obtaining a force diagram of the transmission line according to the slack index includes: Get the historical values ​​of the factors affecting the corresponding slack index; Drawing a force variation diagram of the transmission line under the corresponding slack index according to the historical values; Traversing historical values ​​of influencing factors in the first factor set; All force change diagrams obtained after the traversal is completed are recorded as force shape diagrams.

5. The method for generating a power line image based on physical laws according to claim 4, wherein: The second cyclic generative adversarial network includes: A generator that maps stylization in domain A to stylization in domain B; A generator that maps the stylization of domain B back to the stylization of domain A, and; A discriminator that distinguishes between real and generated images.

6. The method for generating a power line image based on physical laws according to claim 5, wherein: The step of obtaining a force diagram of the transmission line according to the slack index further includes: The historical values ​​of the influencing factors include historical values ​​obtained according to the first image of the target transmission line, or virtual historical values ​​generated according to a preset value logic.

7. The method for generating a power line image based on physical laws according to claim 6, wherein: Also includes: The first factor set includes temperature, humidity, wind speed and wind direction; The different stylizations include sunny, rainy, snowy, and foggy.

8. A system for generating images of power lines based on physical laws, applying the method according to any one of claims 1 to 7, characterized in that: include: a labeling module, configured to acquire a first image of a target transmission line and perform a first labeling operation on the first image; The first labeling operation is used to label the first image with a relaxation degree and a weather condition; An indicator acquisition module, configured to preset a first factor set and establish a slackness indicator based on the first factor set; The slack index is used to quantify the degree of influence of the influencing factors in the first factor set on the slack degree of the transmission line; a force determination module, configured to obtain a force morphology diagram of the transmission line according to the slack index; A network establishment module, configured to establish a first generative adversarial network, wherein a generator input of the first generative adversarial network is a transmission line image of different stylizations and the force form diagram; The different stylized power line images are obtained through a pre-training operation; An image generation module is used to generate a target power transmission line image based on the first generative adversarial network.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the processor implements the steps of a method for generating a power line image based on physical laws according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a method for generating a power line image based on physical laws according to any one of claims 1 to 7 are implemented.

Images