Binocular camera power transmission channel vegetation information acquisition method based on visible light

Abstract

The invention discloses a binocular camera power transmission channel vegetation information acquisition method based on visible light, and relates to the technical field of power transmission channel information acquisition, and the method comprises the steps: collecting power transmission channel vegetation real-time data, carrying out the preprocessing of the real-time data, and transmitting the data to a data processing center; vegetation feature information is extracted through a data processing center, and a power transmission channel three-dimensional model is constructed; and comprehensively evaluating the risk of the power transmission channel based on the three-dimensional model of the power transmission channel, and performing early warning. According to the method, the accuracy of evaluating the risk of the power transmission channel is improved by constructing the three-dimensional model of the power transmission channel; by comprehensively evaluating the risk of the power transmission channel and identifying potential risk factors, decision support is provided for management and maintenance of the power transmission channel, targeted risk management measures and emergency plans are formulated, the overall performance and reliability of the power transmission channel are improved, and better effects are achieved in the aspects of accuracy, efficiency and reliability.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission channel information collection, and in particular to a method for collecting vegetation information on a power transmission channel using a binocular camera based on visible light. Background Art

[0002] The continuous development of the power industry has made the safe operation of transmission channels the key to the stability of the power system. In complex terrains such as mountains and forests, traditional monitoring methods may find it difficult to fully capture vegetation information. At this time, visible light binocular camera technology provides an innovative solution. Visible light imaging technology uses optical sensors to capture targets in the visible light band and generate images. This technology is known for its intuitive imaging effects and rich information. It can clearly display the actual situation of the transmission channel and its surrounding environment. In terms of vegetation information collection, this technology can capture key features such as the appearance, morphology and distribution of vegetation, providing important data support for vegetation identification, classification and monitoring.

[0003] As the core path for power transmission, the safe and stable operation of transmission channels is crucial to ensuring power supply. However, the growth of surrounding vegetation may pose a threat to transmission facilities, such as causing short circuits or tripping. Therefore, it is particularly important to monitor transmission channels in real time and identify and deal with safety hazards in a timely manner.

[0004] Visible light-based binocular camera technology provides a new perspective for monitoring and analysis by capturing vegetation information under the transmission channel. The growth status of vegetation not only reflects the changes in the environment around the transmission channel, but also provides important reference information for the maintenance and management of transmission facilities. By monitoring vegetation information, potential threats to transmission facilities can be discovered and dealt with in a timely manner to ensure the safe and stable operation of the transmission channel. Therefore, visible light binocular camera technology plays a vital role in improving the level of power transmission safety. It not only enhances the real-time monitoring capability of the transmission channel, but also provides a scientific basis for preventing and reducing power failures caused by vegetation through precise vegetation information collection. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is: the existing transmission channel information collection method has problems of low accuracy, low efficiency and reliability, and how to achieve comprehensive monitoring, efficient analysis and timely warning of vegetation in the transmission channel to improve the level of safety management of the transmission channel.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a method for collecting vegetation information on a power transmission channel based on a binocular camera with visible light, comprising collecting real-time data of vegetation on the power transmission channel, preprocessing the real-time data and transmitting it to a data processing center; extracting vegetation feature information and constructing a three-dimensional model of the power transmission channel through the data processing center; and comprehensively evaluating the risks of the power transmission channel and issuing early warnings based on the three-dimensional model of the power transmission channel.

[0008] As a preferred solution of the method for collecting vegetation information of a power transmission channel based on visible light binocular camera described in the present invention, the real-time data of collecting vegetation in the power transmission channel includes collecting left and right image data, color image data and infrared image data of the power transmission channel and surrounding vegetation through a binocular camera, and collecting depth information data of the power transmission channel and the surrounding environment through a lidar sensor.

[0009] As a preferred solution of the method for collecting vegetation information of a power transmission channel based on visible light binocular camera described in the present invention, the real-time data of collecting vegetation in the power transmission channel includes collecting left and right image data, color image data and infrared image data of the power transmission channel and surrounding vegetation through a binocular camera, and collecting depth information data of the power transmission channel and the surrounding environment through a lidar sensor.

[0010] As a preferred solution of the method for collecting vegetation information in a power transmission channel based on visible light binocular camera described in the present invention, the extracting vegetation feature information includes extracting vegetation feature information on an HDR image through a data processing center based on data collected by the data processing center, and extracting feature information through a convolutional neural network, which is expressed as:

[0011]

[0012] Among them, F l (x, y) is the feature output of the pixel point at the coordinate (x, y) of the image collected by the binocular camera after the l-th layer of convolution operation, q is half the size of the convolution kernel, W l (m, n) is the weight matrix of the convolutional neural network, m and n are the row and column indexes in the weight matrix, I is the image collected by the binocular camera, b l is the bias vector of the lth layer; based on the linear classifier, the classification decision function is calculated, which is expressed as:

[0013] Y(g)=sign(w·g+b)

[0014] Among them, Y is the classification decision function, g is the feature vector, and w is the weight vector; Gaussian difference pyramid is established to calculate the stereo matching SIFT feature points, which is expressed as:

[0015] D(x,y,σ)=(G(x,y,lσ)-G(x,y,σ))*I(x,y)

[0016] Where D is the image blurred by the Gaussian kernel function G at the image coordinates (x, y), σ is the standard deviation of the Gaussian kernel, l is the proportional constant, and * is the convolution operation; the disparity d(x, y) is calculated through the stereo matching results, which is expressed as:

[0017] d(x,y)=x L -x R

[0018] Where d(x, y) is the disparity, and d(x, y) represents the point x in the left image. L With point x in the left image R The difference in the horizontal direction; through the disparity, the depth E(x, y) is calculated, expressed as:

[0019]

[0020] Among them, E(x, y) is the depth, E(x, y) represents the actual world distance at the image coordinate (x, y), b is the focal length of the camera, and B is the baseline distance of the binocular camera. The RANSAC algorithm is used to remove the mismatched points and calculate the feature vector fusion formula, which is expressed as:

[0021] V fusion =α·V IR +β·V RGB

[0022] Among them, V fusion For fusion V IR and V RGB The characteristic vector of IR is the feature vector of the infrared image, V RGB is the feature vector of the visible light image, α is the fusion weight of the feature vector of the infrared image, and β is the fusion weight of the feature vector of the visible light image.

[0023] As a preferred solution of the method for collecting vegetation information of a power transmission channel based on a binocular camera with visible light described in the present invention, the construction of a three-dimensional model of the power transmission channel includes extracting vegetation feature information and the spatial positions of stereo matching points on an HDR image to construct a three-dimensional model of the power transmission channel.

[0024] As a preferred solution of the method for collecting vegetation information of power transmission channels based on binocular cameras with visible light described in the present invention, the comprehensive assessment of power transmission channel risk includes inputting a three-dimensional model of the power transmission channel data feature vector v, and calculating the vegetation classification probability, which is expressed as:

[0025] P(h|v)=softmax(W·v+b)

[0026] Among them, P is the vegetation classification probability, h is the vegetation classification result; the vegetation coverage rate is calculated, expressed as:

[0027]

[0028] Among them, CR is the vegetation coverage rate, A vegetation is the total area covered by vegetation, A total is the total area; calculate the vegetation height, expressed as:

[0029] H=Z max -Z min

[0030] Among them, H is the vegetation height, Z max , Z min are the three-dimensional coordinate height values ​​of the top and base of the vegetation respectively; the vegetation density is calculated and expressed as:

[0031]

[0032] Where T is the vegetation density, N vegetation is the number of vegetation per unit area; the distance between vegetation and power transmission equipment is calculated and expressed as:

[0033]

[0034] Among them, u is the distance between vegetation and power transmission equipment, U is the vegetation point at the coordinates (x1, y1, z1), and Q is the power transmission equipment point at the coordinates (x2, y2, z2); when the distance u between vegetation and power transmission equipment is less than the set safety threshold u safe When the vegetation is at a certain height, it poses a potential threat to the power transmission equipment. The risk index S is calculated by the vegetation parameters and the distance from the power transmission equipment, which is expressed as:

[0035]

[0036] Among them, δ is the risk weight parameter of vegetation height, ε is the risk weight parameter of vegetation density, γ is the risk weight parameter of the distance between vegetation and power transmission equipment, and H max is the maximum vegetation height, T max is the maximum value of vegetation density; through time series data, the vegetation growth rate J is calculated, which is expressed as:

[0037]

[0038] Among them, H(t1) and H(t2) are the vegetation heights at time t1 and time t2, respectively.

[0039] As a preferred solution of the method for collecting vegetation information of power transmission channels based on binocular cameras with visible light described in the present invention, the early warning includes early warning of the state of the power transmission channel based on a comprehensive assessment of the risk of the power transmission channel, combined with the risk index and the vegetation growth rate; when the risk index S is greater than or equal to the transmission channel risk index limit value S max When the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel, there is a major risk in the transmission channel, an alarm signal is issued, and the transmission channel needs to be maintained; when the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel, max When an abnormal situation occurs, there is a risk in the transmission channel, an early warning signal is issued, and the transmission channel needs to be maintained; the maintenance process of the transmission channel in the abnormal state is recorded, the data is counted to the data processing center and an abnormal data log is generated.

[0040] Another object of the present invention is to provide a visible light-based binocular camera transmission channel vegetation information acquisition system, which can extract vegetation feature information and construct a three-dimensional model of the transmission channel through a data processing center, thereby solving the problem of low accuracy of current transmission channel information acquisition methods.

[0041] As a preferred solution of the vegetation information acquisition system for a power transmission channel based on a binocular camera with visible light described in the present invention, it includes: a data processing module, a three-dimensional model feature module, and a comprehensive evaluation module; the data processing module is used to collect real-time vegetation data of the power transmission channel, pre-process the real-time data and transmit it to a data processing center; the three-dimensional model feature module is used to extract vegetation feature information and construct a three-dimensional model of the power transmission channel through the data processing center; the comprehensive evaluation module is used to comprehensively evaluate the risks of the power transmission channel based on the three-dimensional model of the power transmission channel and issue an early warning.

[0042] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a method for collecting vegetation information on a power transmission channel using a binocular camera based on visible light.

[0043] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a method for collecting vegetation information in a power transmission channel using a binocular camera based on visible light.

[0044] Beneficial effects of the present invention: The method for collecting vegetation information on a power transmission channel based on visible light binocular cameras provided by the present invention collects high-quality color images and infrared images by installing a binocular camera system, and collects depth information by using a lidar sensor, which helps to more comprehensively understand the growth status and distribution of vegetation. Real-time preprocessing includes image denoising, contrast enhancement and deformity correction, and the use of HDR synthesis technology to improve data quality and ensure the accuracy of analysis; non-contact measurement is performed by binocular cameras, which avoids direct contact with vegetation by traditional manual inspections and reduces damage and interference to vegetation; through a data processing center, vegetation feature information is extracted and a three-dimensional model of the power transmission channel is constructed, which can accurately reflect the growth status and distribution of vegetation and improve the accuracy of assessing the risks of the power transmission channel; by comprehensively assessing the risks of the power transmission channel and identifying potential risk factors, decision support is provided for the management and maintenance of the power transmission channel, and targeted risk management measures and emergency plans are formulated, which helps to improve the overall performance and reliability of the power transmission channel. The present invention achieves better results in terms of accuracy, efficiency and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0046] Figure 1 An overall flow chart of a method for collecting vegetation information in a power transmission channel using a binocular camera based on visible light provided in the first embodiment of the present invention;

[0047] Figure 2 A module schematic diagram of a vegetation information acquisition system for a power transmission channel using a binocular camera based on visible light provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0049] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a method for collecting vegetation information in a power transmission channel using a binocular camera based on visible light, comprising:

[0050] S1: Collect real-time vegetation data of the transmission channel, pre-process the real-time data and transmit it to the data processing center.

[0051] Furthermore, collecting real-time data of vegetation in the power transmission channel includes collecting left and right image data, color image data and infrared image data of the power transmission channel and surrounding vegetation through a binocular camera, and collecting depth information data of the power transmission channel and surrounding environment through a lidar sensor.

[0052] It should be noted that collecting real-time data on vegetation in the power transmission channel includes collecting left and right image data, color image data and infrared image data of the power transmission channel and surrounding vegetation through a binocular camera, and collecting depth information data of the power transmission channel and the surrounding environment through a lidar sensor.

[0053] It should also be noted that the binocular camera photographs the transmission channel area from two slightly offset angles. Due to the different shooting angles, the position of the same object in the two images will be different. This difference is called parallax. By calculating the parallax of each corresponding point in the left and right images, and combining the internal parameters of the camera (such as focal length) and external parameters (such as the distance and relative position between cameras), the three-dimensional coordinates of each point in the scene can be calculated, thereby achieving three-dimensional reconstruction. Ensure that the two images are taken at the same timestamp to reduce the error caused by time difference. During the acquisition process, pay close attention to weather changes and the surrounding environment to ensure that the camera is stable, the light is sufficient, and avoid obstructions that affect the shooting effect. Also avoid the impact of sudden changes in light on image quality, and regularly check the captured images. Perform quality checks to ensure that the image is clear and free of blur or distortion. Automatically adjust the camera exposure time according to the lighting conditions of the transmission channel to ensure image quality. In strong or weak light environments, automatic exposure control or manual exposure parameter setting is required. Noise is filtered out of the captured image to improve image clarity. Contrast enhancement is performed on the captured image to improve the contrast of the image and enhance the distinction between vegetation and background. Due to the camera lens, the image may be geometrically distorted. By using the camera calibration parameters, the image distortion can be corrected to ensure the geometric accuracy of the image. HDR images can provide richer details and a wider brightness range, which helps to more accurately extract vegetation information in environments with large lighting changes.

[0054] S2: Through the data processing center, vegetation feature information is extracted and a three-dimensional model of the transmission channel is constructed.

[0055] Furthermore, extracting vegetation feature information includes extracting vegetation feature information on the HDR image through the data processing center based on the data collected by the data processing center, and extracting feature information through a convolutional neural network, which is expressed as:

[0056]

[0057] Among them, F l (x, y) is the feature output of the pixel point at the coordinate (x, y) of the image collected by the binocular camera after the l-th layer of convolution operation, q is half the size of the convolution kernel, W l (m, n) is the weight matrix of the convolutional neural network, m and n are the row and column indexes in the weight matrix, I is the image collected by the binocular camera, b l is the bias vector of the lth layer; based on the linear classifier, the classification decision function is calculated, which is expressed as:

[0058] Y(g)=sign(w·g+b)

[0059] Among them, Y is the classification decision function, g is the feature vector, and w is the weight vector; Gaussian difference pyramid is established to calculate the stereo matching SIFT feature points, which is expressed as:

[0060] D(x,y,σ)=(G(x,y,lσ)-G(x,y,σ))*I(x,y)

[0061] Where D is the image blurred by the Gaussian kernel function G at the image coordinates (x, y), σ is the standard deviation of the Gaussian kernel, l is the proportional constant, and * is the convolution operation; the disparity d(x, y) is calculated through the stereo matching results, which is expressed as:

[0062] d(x,y)=x L -x R

[0063] Where d(x, y) is the disparity, and d(x, y) represents the point x in the left image. L With point x in the left image R The difference in the horizontal direction; through the disparity, the depth E(x, y) is calculated, expressed as:

[0064]

[0065] Among them, E(x, y) is the depth, E(x, y) represents the actual world distance at the image coordinate (x, y), b is the focal length of the camera, and B is the baseline distance of the binocular camera. The RANSAC algorithm is used to remove the mismatched points and calculate the feature vector fusion formula, which is expressed as:

[0066] V fusion =α·V IR +β·V RGB

[0067] Among them, V fusion For fusion V IR and V RGB The characteristic vector of IR is the feature vector of the infrared image, VRGB is the feature vector of the visible light image, α is the fusion weight of the feature vector of the infrared image, and β is the fusion weight of the feature vector of the visible light image.

[0068] It should be noted that constructing the three-dimensional model of the power transmission channel includes extracting vegetation feature information and spatial positions of stereo matching points based on the HDR image to construct the three-dimensional model of the power transmission channel.

[0069] It should also be noted that the feature extraction process uses a convolutional neural network to automatically learn high-level feature representations of vegetation, and uses labeled vegetation image data to train classifiers to distinguish between vegetation and non-vegetation areas. These two methods improve the robustness and accuracy of features, and detect SIFT feature points in the left and right images respectively. These feature points have characteristics such as scale invariance and rotation invariance. A descriptor is generated for each feature point to search for matching points in the other image, and a disparity map is calculated based on the matching results. The disparity map represents the difference in distance from each point in the scene to the camera, and the RANSAC algorithm is used to eliminate mismatched points. The feature vector fusion formula is combined with infrared and visible light multimodal data to fuse data of different modalities. For example, the temperature information in the infrared image can be combined with the color and texture features in the visible light image. By combining the three characteristics of the transmission channel, a richer feature vector is formed, thus obtaining more comprehensive and accurate vegetation information on the transmission channel. By constructing a three-dimensional model of the transmission channel, an in-depth understanding of the vegetation growth status is achieved. The establishment of this three-dimensional model makes the management process more data-based and refined. For example, by extracting the appearance characteristics and temperature characteristics of the vegetation, the health status of the vegetation can be effectively monitored, problems can be discovered and dealt with in a timely manner, thereby ensuring the safe operation of the transmission channel. The calculation of the vegetation height characteristic function helps to assess the potential risks that vegetation may cause to the transmission line. The introduction of characteristic function weights improves the pertinence and flexibility of data analysis, making the decision-making process more scientific and reasonable. By analyzing the weight distribution of different characteristic functions, managers can more accurately grasp the key factors of vegetation information and formulate more effective management strategies.

[0070] S3: Based on the three-dimensional model of the transmission channel, comprehensively assess the risks of the transmission channel and issue early warning.

[0071] Furthermore, the comprehensive assessment of transmission channel risk includes calculating the vegetation classification probability based on the transmission channel three-dimensional model, inputting the transmission channel three-dimensional model data feature vector v, and expressing it as:

[0072] P(h|v)=softmax(W·v+b)

[0073] Among them, P is the vegetation classification probability, h is the vegetation classification result; the vegetation coverage rate is calculated, expressed as:

[0074]

[0075] Among them, CR is the vegetation coverage rate, A vegetation is the total area covered by vegetation, A total is the total area; calculate the vegetation height, expressed as:

[0076] H=Z max -Z min

[0077] Among them, H is the vegetation height, Z max , Z min are the three-dimensional coordinate height values ​​of the top and base of the vegetation respectively; the vegetation density is calculated and expressed as:

[0078]

[0079] Where T is the vegetation density, N vegetaion is the number of vegetation per unit area; the distance between vegetation and power transmission equipment is calculated and expressed as:

[0080]

[0081] Among them, u is the distance between vegetation and power transmission equipment, U is the vegetation point at the coordinates (x1, y1, z1), and Q is the power transmission equipment point at the coordinates (x2, y2, z2); when the distance u between vegetation and power transmission equipment is less than the set safety threshold u safe When the vegetation is at a certain height, it poses a potential threat to the power transmission equipment. The risk index S is calculated by the vegetation parameters and the distance from the power transmission equipment, which is expressed as:

[0082]

[0083] Among them, δ is the risk weight parameter of vegetation height, ε is the risk weight parameter of vegetation density, γ is the risk weight parameter of the distance between vegetation and power transmission equipment, and H max is the maximum vegetation height, T max is the maximum value of vegetation density; through time series data, the vegetation growth rate J is calculated, which is expressed as:

[0084]

[0085] Among them, H(t1) and H(t2) are the vegetation heights at time t1 and time t2, respectively.

[0086] It should be noted that the early warning includes early warning of the transmission channel status based on a comprehensive assessment of the transmission channel risk, combined with the risk index and vegetation growth rate; when the risk index S is greater than or equal to the transmission channel risk index limit value S maxWhen the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel, there is a major risk in the transmission channel, an alarm signal is issued, and the transmission channel needs to be maintained; when the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel, max When an abnormal situation occurs, there is a risk in the transmission channel, an early warning signal is issued, and the transmission channel needs to be maintained; the maintenance process of the transmission channel in the abnormal state is recorded, the data is counted to the data processing center and an abnormal data log is generated.

[0087] It should also be noted that by inputting the characteristic vector of the three-dimensional model data of the transmission channel and calculating the vegetation classification probability, vegetation coverage, vegetation height, vegetation density and the distance between vegetation and transmission equipment, comprehensive monitoring and analysis of the vegetation status of the transmission channel can be achieved. For example, by calculating the vegetation classification probability, different types of vegetation can be identified more accurately, which is crucial for formulating targeted maintenance strategies. The calculation of vegetation coverage and vegetation density helps to evaluate the impact of vegetation on transmission lines, while the measurement of vegetation height and distance from transmission equipment provides a quantitative basis for judging whether vegetation poses a potential threat. The calculation of risk index S takes into account multiple factors such as vegetation height, density and distance from transmission equipment, providing a comprehensive indicator for risk assessment of transmission channels. The use of this method makes the decision-making process more data-based and objective, thereby improving the efficiency and effectiveness of transmission channel management. For example, by analyzing the risk index, managers can promptly identify high-risk areas and take corresponding preventive measures. The introduction of vegetation growth rate J provides important information for monitoring vegetation growth dynamics, which is of great significance for predicting and preventing safety hazards caused by excessive vegetation growth. Combining risk index and vegetation growth rate to conduct The system can provide early warning to ensure the safety of the transmission channel in a timely manner. When the risk index S or the vegetation growth rate J exceeds the set limit, the system can automatically send out an alarm signal to prompt the manager to perform necessary maintenance work. This early warning mechanism significantly improves the operation and maintenance response speed and emergency handling capabilities of the transmission channel. The maintenance process of the abnormal transmission channel is recorded, and the data is statistically sent to the data processing center to generate an abnormal data log, which not only enhances the data management capabilities, but also promotes the consistency of information sharing and analysis. This centralized data processing method greatly improves the power system managers' level of cognition of the transmission channel status and provides strong technical support for the safe and stable operation of the transmission line. In this way, we can better understand the laws of vegetation growth in the transmission channel, so as to formulate a more scientific and reasonable maintenance plan.

[0088] Example 2 is an embodiment of the present invention, which provides a method for collecting vegetation information in a power transmission channel using a binocular camera based on visible light. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0089] First, a binocular camera system, including visible light and infrared cameras, and a lidar sensor are installed along the transmission channel to ensure that all equipment is correctly calibrated and connected to the data processing center. The binocular camera system is started to collect color images, infrared images, and depth information data of the transmission channel area in real time. At the same time, the lidar sensor is used to collect depth information of the transmission channel and the surrounding environment. In the data processing center, image denoising, contrast enhancement, and deformity correction are performed on the real-time data. Using HDR synthesis technology, multiple images with different exposures are fused into an HDR image, and the pre-processed data is transmitted to the data processing center for storage and backup. Next, on the HDR image, a convolutional neural network is used to extract vegetation feature information, calculate vegetation appearance feature function, temperature feature function, and height feature function, and construct a three-dimensional model of the transmission channel based on the extracted vegetation feature information. Calculate vegetation classification probability, coverage, height, density, and distance from the transmission equipment. Finally, calculate the risk index S, and calculate the vegetation growth rate J in combination with time series data. When the risk index S is greater than or equal to the transmission channel risk index limit value S max Or the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel max When an abnormal situation occurs, an early warning signal is issued, the maintenance process of the transmission channel in abnormal state is recorded, and the data is counted to the data processing center and an abnormal data log is generated.

[0090] Refer to Table 1 to record and analyze the experimental data.

[0091] Table 1 Experimental data record table

[0092]

[0093] By comparing the vegetation classification probability, vegetation coverage rate, vegetation height, vegetation density, distance between vegetation and transmission equipment, risk index S and vegetation growth rate J of different transmission channels, the beneficial effects of our invention can be seen. For example, the vegetation classification probability and vegetation coverage rate of transmission channel 3 are relatively high, indicating that the vegetation growth in this area is relatively vigorous, and the corresponding risk index is also high, which indicates that there may be greater safety hazards in this area and priority attention and maintenance are required. It can be seen from Table 1 that the vegetation height and vegetation density of transmission channel 3 are also relatively high, and the distance between vegetation and transmission equipment is relatively close. These factors together lead to a higher risk index. At the same time, the vegetation growth rate is also high, indicating that the vegetation grows faster. If it is not handled in time, it may quickly become a major threat to the safety of the transmission channel. Through this comprehensive evaluation, the present invention can accurately identify the transmission channels that need priority maintenance. Channels, such as transmission channel 3, whose vegetation information shows a high risk index and a high vegetation growth rate, indicating that the maintenance demand in this area is more urgent. The present invention is superior to the traditional transmission channel management method because it not only takes into account the current state of the vegetation, but also predicts the future growth trend of the vegetation, thereby providing a more comprehensive and forward-looking risk assessment. Compared with the prior art, the present invention provides a more accurate, data-driven transmission channel vegetation information collection and risk assessment method. By integrating a variety of data acquisition devices such as binocular camera systems and lidar sensors and using advanced image processing and machine learning algorithms, it can more accurately identify the potential risks and maintenance needs of the transmission channel, which not only improves the efficiency of transmission channel management, but also helps to ensure the safe and stable operation of the power system and promote the power industry to develop towards intelligent and refined management. Therefore, our invention is creative.

[0094] Example 3, reference Figure 2 , is an embodiment of the present invention, and provides a vegetation information acquisition system for a power transmission channel based on a binocular camera of visible light, including a data processing module, a three-dimensional model feature module, and a comprehensive evaluation module.

[0095] The data processing module is used to collect real-time vegetation data on the transmission channel, pre-process the real-time data and transmit it to the data processing center; the three-dimensional model feature module is used to extract vegetation feature information through the data processing center and construct a three-dimensional model of the transmission channel; the comprehensive assessment module is used to comprehensively assess the risks of the transmission channel based on the three-dimensional model of the transmission channel and issue an early warning.

[0096] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0097] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0098] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0099] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0100] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for collecting vegetation information in a power transmission channel based on a binocular camera with visible light, characterized in that: include: Collect real-time vegetation data of power transmission channels, pre-process the real-time data and transmit it to the data processing center; Through the data processing center, vegetation feature information is extracted and a three-dimensional model of the power transmission channel is constructed; Based on the three-dimensional model of the transmission channel, the risks of the transmission channel are comprehensively assessed and early warning is issued.

2. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 1, characterized in that: The real-time data collection of vegetation in the power transmission channel includes collecting left and right image data, color image data and infrared image data of the power transmission channel and surrounding vegetation through a binocular camera, and collecting depth information data of the power transmission channel and the surrounding environment through a lidar sensor.

3. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 2, characterized in that: The transmission to the data processing center includes image denoising, contrast enhancement and deformity correction of real-time data, using HDR synthesis technology to fuse multiple images with different exposures into one HDR image, and transmitting the pre-processed data to the data processing center for data storage and data backup.

4. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 3, characterized in that: The extracting of vegetation feature information includes extracting vegetation feature information on the HDR image through a data processing center based on data collected by the data processing center, and extracting feature information through a convolutional neural network, which is expressed as: Among them, F l (x, y) is the feature output of the pixel point at the coordinate (x, y) of the image collected by the binocular camera after the l-th layer of convolution operation, q is half the size of the convolution kernel, W l (m,n) is the weight matrix of the convolutional neural network, m and n are the row and column indexes in the weight matrix, I is the image collected by the binocular camera, and b l is the bias vector of the lth layer; Based on the linear classifier, the classification decision function is calculated, which is expressed as: Y(g)=sign(w·g+b) Among them, Y is the classification decision function, g is the feature vector, and w is the weight vector; Establish a Gaussian difference pyramid and calculate the stereo matching SIFT feature points, which can be expressed as: D(x,y,σ)=(G(x,y,lσ)-G(x,y,σ))*I(x,y) Where D is the image blurred using the Gaussian kernel function G at the image coordinates (x, y), σ is the standard deviation of the Gaussian kernel, l is the proportional constant, and * is the convolution operation; Through the stereo matching results, the disparity d(x,y) is calculated, which is expressed as: d(x,y)=x L -x R Where d(x,y) is the disparity, and d(x,y) represents the point x in the left image. L With point x in the left image R Differences in horizontal direction; Through parallax, the depth E(x,y) is calculated, which is expressed as: Where E(x,y) is the depth, E(x,y) represents the actual world distance at the image coordinate (x,y), b is the focal length of the camera, and B is the baseline distance of the binocular camera; The RANSAC algorithm is used to remove mismatched points and calculate the feature vector fusion formula, which is expressed as: V fusion =α·V IR +β·V RGB Among them, V fusion For fusion V IR and V RGB The characteristic vector of IR is the feature vector of the infrared image, V RGB is the feature vector of the visible light image, α is the fusion weight of the feature vector of the infrared image, and β is the fusion weight of the feature vector of the visible light image.

5. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 4, characterized in that: The construction of the three-dimensional model of the power transmission channel includes extracting vegetation feature information and spatial positions of stereo matching points on the HDR image to construct the three-dimensional model of the power transmission channel.

6. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 5, characterized in that: The comprehensive assessment of transmission channel risk includes inputting a three-dimensional model of the transmission channel data feature vector v and calculating the vegetation classification probability based on the three-dimensional model of the transmission channel, which is expressed as: P(h|v)=softmax(W·v+b) Among them, P is the vegetation classification probability, and h is the vegetation classification result; Calculate the vegetation coverage, expressed as: Among them, CR is the vegetation coverage rate, A vegetation is the total area covered by vegetation, A total is the total area; Calculate the vegetation height, expressed as: H=Z max -Z min Among them, H is the vegetation height, Z max , Z min They are the three-dimensional coordinate height values ​​of the top and base of the vegetation respectively; Calculate vegetation density, expressed as: Where T is the vegetation density, N vegetation is the amount of vegetation per unit area; Calculate the distance between vegetation and power transmission equipment, expressed as: Among them, u is the distance between vegetation and power transmission equipment, U is the vegetation point at the coordinates (x1, y1, z1), and Q is the power transmission equipment point at the coordinates (x2, y2, z2); When the distance u between vegetation and power transmission equipment is less than the set safety threshold u safe When vegetation is used, it poses a potential threat to power transmission equipment; The risk index S is calculated by vegetation parameters and the distance from the power transmission equipment, which is expressed as: Among them, δ is the risk weight parameter of vegetation height, ε is the risk weight parameter of vegetation density, γ is the risk weight parameter of the distance between vegetation and power transmission equipment, and H max is the maximum vegetation height, T max is the maximum value of vegetation density; Through time series data, the vegetation growth rate J is calculated and expressed as: Among them, H(t1) and H(t2) are the vegetation heights at time t1 and time t2, respectively.

7. The method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in claim 6, characterized in that: The early warning includes early warning of the transmission channel status based on a comprehensive assessment of the transmission channel risk, combined with the risk index and the vegetation growth rate; When the risk index S is greater than or equal to the transmission channel risk index limit value S max When a transmission channel is at a significant risk, an alarm signal is issued and maintenance of the transmission channel is required; When the vegetation growth rate J is greater than or equal to the vegetation growth rate limit value J of the transmission channel max When the transmission channel is at risk, an early warning signal is issued and the transmission channel needs to be maintained; The maintenance process of the transmission channel in abnormal state is recorded, the data is counted to the data processing center and an abnormal data log is generated.

8. A system using the method for collecting vegetation information in a power transmission channel based on a binocular camera using visible light as claimed in any one of claims 1 to 7, characterized in that: Including data processing module, 3D model feature module, and comprehensive evaluation module; The data processing module is used to collect real-time data of vegetation in the power transmission channel, pre-process the real-time data and transmit it to the data processing center; The three-dimensional model feature module is used to extract vegetation feature information through the data processing center and construct a three-dimensional model of the power transmission channel; The comprehensive assessment module is used to comprehensively assess the risks of the power transmission channel based on the three-dimensional model of the power transmission channel and to issue an early warning.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for collecting vegetation information in a power transmission channel based on a visible light binocular camera according to any one of claims 1 to 7 are implemented.

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 the method for collecting vegetation information on a power transmission channel based on a visible light binocular camera according to any one of claims 1 to 7 are implemented.

Images