Cable tunnel characteristic data transmission method and device and computer equipment

Abstract

The invention relates to a cable tunnel characteristic data transmission method and device and computer equipment. The method comprises the steps of obtaining point cloud data of a cable tunnel, performing model fitting according to fitting model information and the point cloud data to obtain a current model used for representing a plane model or a curved surface model corresponding to the cable tunnel, obtaining distance information from the point cloud data to the current model, and obtaining a target point of the cable tunnel according to the distance information and inner point threshold value information. Determining a current inner point set and a current inner point number, if the current inner point number is greater than the inner point number recorded last time, updating the inner point number and the inner point set recorded last time, and returning to execute model fitting to obtain a new current target model and determining a new current inner point set and a new current inner point number; and until the maximum number of iterations is reached, determining the current inner point set as compressed data, and transmitting the compressed data to the edge node closest to the cable tunnel. By adopting the method, the delay of data transmission is reduced, and the transmission efficiency of data transmission is improved.

Description

Technical Field

[0001] The present application relates to the technical field of the Internet of Things, and particularly to a method, device, computer device, computer-readable storage medium, and computer program product for transmitting cable tunnel feature data. Background Art

[0002] With the wide application of Internet of Things technology, the monitoring requirements for infrastructure such as cable tunnels are increasing day by day. The amount of data collected in cable tunnels is huge. With the increase in the number of sensors in cable tunnels, the collected data has also grown rapidly.

[0003] However, the current traditional data transmission method has the problem of high data transmission delay. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, computer device, computer-readable storage medium, and computer program product for transmitting cable tunnel feature data that can reduce data transmission delay.

[0005] In a first aspect, the present application provides a method for transmitting cable tunnel feature data, including:

[0006] Obtaining cable tunnel point cloud data of a cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0007] Performing model fitting according to preset fitting model information and the cable tunnel point cloud data to obtain a current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel;

[0008] Obtaining distance information from the cable tunnel point cloud data to the current target model;

[0009] Determining a current inlier set and a current inlier number corresponding to the current target model according to the distance information and preset inlier threshold information;

[0010] If the current inlier number is greater than the previously recorded inlier number, updating the previously recorded inlier number and inlier set, and returning to execute model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain a new current target model and determining a new current inlier set and a new current inlier number until the set maximum number of iterations is reached, and determining the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0011] Transmitting the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0012] In one embodiment, after acquiring the point cloud data of the cable tunnel and before performing model fitting according to the preset fitting model information and the point cloud data of the cable tunnel, the method includes:

[0013] Obtain the neighborhood corresponding to each data point in the point cloud data of the cable tunnel;

[0014] According to each data point and its corresponding neighborhood, obtain the neighborhood distance corresponding to each data point;

[0015] Based on the neighborhood distance, obtain the global mean of the point cloud data of the cable tunnel, and according to the neighborhood distance and the global mean, obtain the global standard deviation of the point cloud data of the cable tunnel;

[0016] Use the global mean, the global standard deviation, and the neighborhood distance to remove the abnormal data points in the point cloud data of the cable tunnel, and obtain the cleaned point cloud data of the cable tunnel.

[0017] In one embodiment, using the global mean, the global standard deviation, and the neighborhood distance to remove the abnormal data points in the point cloud data of the cable tunnel and obtain the cleaned point cloud data of the cable tunnel includes:

[0018] Based on the global mean and the global standard deviation, determine the abnormal neighborhood distance threshold corresponding to the point cloud data of the cable tunnel;

[0019] Determine the data points with neighborhood distances greater than the abnormal neighborhood distance threshold as abnormal data points;

[0020] Remove the abnormal data points in the point cloud data of the cable tunnel to obtain the cleaned point cloud data of the cable tunnel.

[0021] In one embodiment, after obtaining the cleaned point cloud data of the cable tunnel and before performing model fitting according to the preset fitting model information and the point cloud data of the cable tunnel, the method further includes:

[0022] Generate a corresponding feature matrix according to the cleaned point cloud data of the cable tunnel;

[0023] Generate a covariance matrix based on the feature matrix, and perform eigenvalue decomposition on the covariance matrix to obtain feature information; the feature information includes eigenvalues and eigenvectors;

[0024] Sort the feature information according to the magnitudes of the eigenvalues to obtain a sorted eigenvector matrix;

[0025] Determine the target eigenvectors that meet the preset sorting sequence numbers from the sorted eigenvector matrix, and construct a target eigenvector matrix according to the target eigenvectors;

[0026] Based on the target eigenvector matrix and the feature matrix, determine the cable tunnel feature data.

[0027] In an exemplary embodiment, model fitting is performed according to preset fitting model information and cable tunnel point cloud data to obtain a current target model, including:

[0028] Based on the preset fitting model information and cable tunnel feature data, a plurality of sample sets are randomly generated;

[0029] Model fitting is performed on any one of the sample sets to obtain the current target model.

[0030] In a second aspect, the present application further provides a method for transmitting cable tunnel feature data, which is applied to an edge node closest to the cable tunnel, including:

[0031] Receiving compressed cable tunnel feature data, and generating a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained by the method according to any one of claims 1 to 5;

[0032] Performing singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix;

[0033] Truncating the singular value matrix according to preset threshold information to obtain a truncated singular value matrix;

[0034] Performing data restoration on the truncated singular value matrix to obtain cable tunnel feature data.

[0035] In a third aspect, the present application further provides a device for transmitting cable tunnel feature data, including:

[0036] A point cloud data acquisition module, configured to acquire cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0037] A model fitting module, configured to perform model fitting according to preset fitting model information and cable tunnel point cloud data to obtain a current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel;

[0038] A distance information acquisition module, configured to acquire distance information from the cable tunnel point cloud data to the current target model;

[0039] An inlier acquisition module, configured to determine a current inlier set and a current inlier number corresponding to the current target model according to the distance information and preset inlier threshold information;

[0040] A data compression module, which is used to update the previously recorded number of inlier points and the inlier point set if the current number of inlier points is greater than the previously recorded number of inlier points, and return to execute model fitting based on the preset fitting model information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier point set and a new current number of inlier points until the set maximum number of iterations is reached, and determine the current inlier point set as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0041] A data transmission module, which is used to transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0042] In a fourth aspect, the present application further provides a device for transmitting cable tunnel feature data, which is applied to the edge node closest to the cable tunnel, and includes:

[0043] A data receiving module, which is used to receive the compressed cable tunnel feature data and generate a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained by the method according to any one of claims 1 to 5;

[0044] A decomposition module, which is used to perform singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix;

[0045] A truncation module, which is used to truncate the singular value matrix according to the preset threshold information to obtain a truncated singular value matrix;

[0046] A data restoration module, which is used to restore the data of the truncated singular value matrix to obtain the cable tunnel feature data.

[0047] In a fifth aspect, the present application further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0048] Obtain the cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0049] Perform model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain a current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel;

[0050] Obtain the distance information from the cable tunnel point cloud data to the current target model;

[0051] Determine the current inlier point set and the current number of inlier points corresponding to the current target model according to the distance information and the preset inlier threshold information;

[0052] If the current number of inlier points is greater than the previously recorded number of inlier points, update the previously recorded number of inlier points and the set of inlier points, and return to perform model fitting based on the pre-set fitting model information and the cable tunnel point cloud data to obtain a new current target model, determine a new current set of inlier points and a new current number of inlier points, until the set maximum number of iterations is reached, and determine the current set of inlier points as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0053] Transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0054] In a sixth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0055] Obtain the cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0056] Perform model fitting based on the pre-set fitting model information and the cable tunnel point cloud data to obtain a current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel;

[0057] Obtain the distance information from the cable tunnel point cloud data to the current target model;

[0058] Determine the current set of inlier points and the current number of inlier points corresponding to the current target model according to the distance information and the pre-set inlier threshold information;

[0059] If the current number of inlier points is greater than the previously recorded number of inlier points, update the previously recorded number of inlier points and the set of inlier points, and return to perform model fitting based on the pre-set fitting model information and the cable tunnel point cloud data to obtain a new current target model, determine a new current set of inlier points and a new current number of inlier points, until the set maximum number of iterations is reached, and determine the current set of inlier points as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0060] Transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0061] In a seventh aspect, the present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0062] Obtain the cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0063] Perform model fitting based on the pre-set fitting model information and the cable tunnel point cloud data to obtain the current target model; the current target model is used to represent the plane model or the surface model corresponding to the cable tunnel;

[0064] Obtain the distance information from the cable tunnel point cloud data to the current target model;

[0065] Determine the current inlier set and the current inlier number corresponding to the current target model according to the distance information and the pre-set inlier threshold information;

[0066] If the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and the inlier set, and return to perform model fitting according to the pre-set fitting model information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0067] Transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0068] For the above cable tunnel feature data transmission method, device, computer device, computer-readable storage medium and computer program product, obtain the cable tunnel point cloud data of the cable tunnel, where the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data. Perform model fitting according to the pre-set fitting model information and the cable tunnel point cloud data to obtain the current target model used to represent the plane model or the surface model corresponding to the cable tunnel. Obtain the distance information from the cable tunnel point cloud data to the current target model. Determine the current inlier set and the current inlier number corresponding to the current target model according to the distance information and the pre-set inlier threshold information. If the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and the inlier set, and return to perform model fitting according to the pre-set model fitting information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel, and transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel. Use the cable tunnel point cloud data for model fitting, use the distance information between the cable tunnel point cloud data and the fitted model to compress the cable tunnel point cloud data to obtain the compressed cable tunnel feature data. By compressing the point cloud data with a large amount of data, select important and effective data for transmission, reduce the latency of data transmission, and improve the transmission efficiency of data transmission. Description of the Drawings

[0069] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0070] Figure 1 It is an application environment diagram of the cable tunnel feature data transmission method in an embodiment;

[0071] Figure 2 It is a schematic flowchart of the cable tunnel feature data transmission method in an embodiment;

[0072] Figure 3 It is a schematic flowchart of the cable tunnel feature data transmission method in another embodiment;

[0073] Figure 4 It is a specific schematic flowchart of the cable tunnel feature data transmission method in an embodiment;

[0074] Figure 5 It is a slope graph for selecting the number of principal components in another embodiment;

[0075] Figure 6 It is a structural block diagram of the cable tunnel feature data transmission device in an embodiment;

[0076] Figure 7 It is a structural block diagram of the cable tunnel feature data transmission device in another embodiment;

[0077] Figure 8 It is an internal structure diagram of a computer device in an embodiment;

[0078] Figure 9 It is an internal structure diagram of a computer device in another embodiment. Detailed implementation manners

[0079] In order to make the objectives, technical solutions and advantages of the present application clearer, the following further details the present application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0080] The cable tunnel feature data transmission method provided by the embodiments of the present application can be applied to, for example, Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The terminal 102 receives the point cloud data collected by the sensors in the cable tunnel and can upload it to the server 104 for data processing, or the terminal 102 can perform data processing by itself. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed on the cloud or other network servers. The terminal 102 or the server 104 obtains the point cloud data of the cable tunnel, where the point cloud data of the cable tunnel includes at least one of the pipe point cloud data, the cable point cloud data, and the obstacle point cloud data. Model fitting is performed according to the preset fitting model information and the point cloud data of the cable tunnel to obtain the current target model. The current target model is used to represent the plane model or the surface model corresponding to the cable tunnel. The distance information from the point cloud data of the cable tunnel to the current target model is obtained. According to the distance information and the preset inlier threshold information, the current inlier set and the current inlier number corresponding to the current target model are determined. If the current inlier number is greater than the previously recorded inlier number, the previously recorded inlier number and the inlier set are updated, and return to execute model fitting according to the preset fitting model information and the point cloud data of the cable tunnel to obtain a new current target model and determine a new current inlier set and a new current inlier number until the set maximum number of iterations is reached. The current inlier set is determined as the compressed cable tunnel feature data corresponding to the cable tunnel, and the compressed cable tunnel feature data is transmitted to the edge node closest to the cable tunnel. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, projection devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, a smart glasses, etc. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0081] In an exemplary embodiment, as Figure 2 shown, a method for transmitting cable tunnel feature data is provided. Taking the method applied to Figure 1 the terminal 102 in

[0082] as an example, it includes the following steps S201 to S206. Among them:

[0082] Step S201, obtain the point cloud data of the cable tunnel; the point cloud data of the cable tunnel includes at least one of the pipe point cloud data, the cable point cloud data, and the obstacle point cloud data.

[0083] Among them, the point cloud data can be understood as a set of points, which is a data format that uses a set of points in three-dimensional space to represent the three-dimensional shape and features of an object or a scene. It is usually composed of a large number of points, and each point has three coordinate values (X, Y, Z), representing the position of the point in three-dimensional space.

[0084] Optionally, the terminal 102 obtains various types of point cloud data collected by sensors installed in the cable tunnel, including at least one of the pipe point cloud data, cable point cloud data, and obstacle point cloud data in the cable tunnel. Obtaining a large amount of point cloud data in the cable tunnel lays a data foundation for subsequent data processing and data transmission, and collecting such data can help understand the internal state information of the cable tunnel.

[0085] Step S202: Perform model fitting based on the preset fitting model information and the cable tunnel point cloud data to obtain the current target model; the current target model is used to represent the plane model or surface model corresponding to the cable tunnel.

[0086] Among them, the fitting model information can be understood as the minimum number of data points required to form a plane model or a surface model, and the current target model can be understood as the model that can best represent the characteristics of the cable tunnel.

[0087] Exemplarily, the terminal 102 performs model fitting based on the preset fitting model information and the cable tunnel point cloud data to obtain the current target model for representing the plane model or surface model corresponding to the cable tunnel, laying a selection basis for subsequent selection of effective point cloud data.

[0088] Step S203: Obtain the distance information from the cable tunnel point cloud data to the current target model.

[0089] Step S204: Determine the current inlier set and the current inlier number corresponding to the current target model according to the distance information and the preset inlier threshold information.

[0090] Among them, the distance information can be understood as the distance from each data point to the current target model, the inlier can be understood as the data point equivalent to being inside the current target model, that is, the typical data that can better represent the cable tunnel, and the inlier threshold information can be understood as the numerical information that the data points inside the model need to meet.

[0091] Optionally, the terminal 102 obtains the distance information from the cable tunnel point cloud data to the current target model, compares each distance information with the preset inlier threshold information, classifies the data points with the distance information less than or equal to the inlier threshold information into the current inlier set, determines the corresponding current inlier number, and saves it. By comparing the distance information with the inlier threshold, the inliers that meet the standards can be quickly screened out, providing a data basis for subsequent data update and data iteration.

[0092] Step S205, if the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and inlier set, and return to execute model fitting according to the preset fitting model information and cable tunnel point cloud data to obtain a new current target model, determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel.

[0093] Step S206, transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0094] Among them, the compressed cable tunnel feature data can be understood as the data set that best represents the cable tunnel in the current iteration cycle.

[0095] Exemplarily, if the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and inlier set, and return to execute model fitting according to the preset fitting model information and cable tunnel point cloud data to obtain a new current target model, determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and no longer continue model fitting. Determine the currently stored current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel, and use the combination of MEC (Mobile Edge Computing) technology and IoT (Internet of Things) technology to transmit the compressed cable tunnel feature data:

[0096] 1. The MEC node deployed at the data source location collects the already compressed data. According to Shannon's theorem, the uplink rate at which the sensor r offloads the computing task to the MEC server q is:

[0097]

[0098] Where is the transmit power from the MEC server q to the sensor r; is the Gaussian white noise; is the uplink channel transmission bandwidth; is the channel gain of the transmission line sensor r transferred to the MEC server q;

[0099] 2. The MEC server q offloads the computing task to the cloud server e, with a rate as follows:

[0100]

[0101] where is the uplink channel transmission bandwidth of the cloud server e;

[0102] 3. The task is divided into two parts of arbitrary size and can be executed in parallel on the sensor device and the mobile edge server. The task q is executed on the sensor r, and the size of the task to be executed by the sensor r at time slot t is , and the offloading ratio of the task is . Given that the CPU clock frequency of the sensor r is f, the local time consumption is:

[0103]

[0104] 4. The total power consumption of local execution is related to the effective switching capacitance g of the chip structure, and the total power consumption is:

[0105]

[0106] Since the computing power of the MEC server is relatively strong, the execution time of offloading the computing task to the MEC server can be ignored;

[0107] 5. By combining the Alternating Direction Method of Multipliers (ADMM) and nonlinear fractional programming, an energy consumption minimization algorithm is obtained. The algorithm steps are as follows:

[0108] *1: Initialization: , , , , int i = 1, j = 1, l = 1;

[0109] *2: Reset the offloading rate and power of the sensor;

[0110] *3: int j = 1, Update the ADMM parameters;

[0111] *4: j++, while , go to *3;

[0112] *5: Iterate ;

[0113] *6: l++, while , go to *3;

[0114] *7: Output ;

[0115] *8: i++, while , go to *2;

[0116] According to Shannon's theorem, by reasonably allocating the transmission power, bandwidth, and channel gain, the uplink rate from the sensor to server 104 is optimized, ensuring the efficiency and stability of data transmission.

[0117] In the above cable tunnel feature data transmission method, the cable tunnel point cloud data of the cable tunnel is obtained. The cable tunnel point cloud data includes at least one of the pipe point cloud data, cable point cloud data, and obstacle point cloud data. Model fitting is performed according to the preset fitting model information and the cable tunnel point cloud data to obtain the current target model for characterizing the corresponding plane model or surface model of the cable tunnel. The distance information from the cable tunnel point cloud data to the current target model is obtained. According to the distance information and the preset inlier threshold information, the current inlier set and the current inlier number corresponding to the current target model are determined. If the current inlier number is greater than the previously recorded inlier number, the previously recorded inlier number and the inlier set are updated, and return to execute model fitting according to the preset model fitting information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier set and a new current inlier number until the set maximum number of iterations is reached. The current inlier set is determined as the compressed cable tunnel feature data corresponding to the cable tunnel, and the cable compressed tunnel feature data is transmitted to the edge node closest to the cable tunnel. Model fitting is performed using the cable tunnel point cloud data, and data compression is performed on the cable tunnel point cloud data using the distance information between the cable tunnel point cloud data and the fitted model to obtain the compressed cable tunnel feature data. By performing data compression on the point cloud data with a large amount of data, important and effective data is selected for transmission, reducing the latency of data transmission and improving the transmission efficiency of data transmission.

[0118] In one embodiment, after obtaining the cable tunnel point cloud data of the cable tunnel and before performing model fitting according to the preset fitting model information and the cable tunnel point cloud data, the method includes:

[0119] Obtain the neighborhood corresponding to each data point in the cable tunnel point cloud data; obtain the neighborhood distance corresponding to each data point according to each data point and its corresponding neighborhood; obtain the global mean of the cable tunnel point cloud data based on the neighborhood distance, and obtain the global standard deviation of the cable tunnel point cloud data according to the neighborhood distance and the global mean; use the global mean, global standard deviation, and neighborhood distance to remove the abnormal data points in the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data.

[0120] Optionally, 1. Define a neighborhood. For each point, select its neighborhood ;

[0121] 2. Calculate the distance l from a point to each point within its neighborhood ij . If the neighborhood of point p is defined as the k nearest neighbor points, calculate the distances from point p to these k points, which are respectively l p1 , l p2 , , l pk , where l pi represents the distance from point p to its i-th neighborhood point;

[0122] 3. Calculate the global mean and variance of all points and their neighborhood distances;

[0123] 3.1. Global mean calculation. Suppose there are n points in the point cloud data. For each point , its neighborhood has points ( is the pre-set neighborhood size). The distance from point to the point ( ) within its neighborhood is .

[0124] 3.2. Calculate the global mean , and its formula is as follows:

[0125]

[0126] 3.3. Calculate the global variance , and its formula is as follows:

[0127]

[0128] Among them, the global standard deviation is the square root of the global variance , that is .

[0129] Using the global mean , the global standard deviation and the neighborhood distance , remove the abnormal data points in the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data. By calculating the distances between points and neighborhood points and applying the global mean and variance, abnormal data points can be effectively detected. These abnormal points are usually caused by noise or incorrect measurements. By removing these points, the quality and reliability of the data are improved.

[0130] In one embodiment, global mean, global standard deviation, and neighborhood distance are used to remove abnormal data points from the cable tunnel point cloud data, and the cleaned cable tunnel point cloud data is obtained, including: determining an abnormal neighborhood distance threshold corresponding to the cable tunnel point cloud data based on the global mean and global standard deviation; determining data points with neighborhood distance greater than the abnormal neighborhood distance threshold as abnormal data points; removing the abnormal data points from the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data.

[0131] Exemplarily, data points that satisfy or are determined as abnormal data points, and the abnormal data points are removed from the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data. The noise and errors are removed from the cleaned point cloud data, improving the overall quality of the data. This helps with subsequent analysis and applications, such as 3D modeling, condition monitoring, etc., ensuring the accuracy and effectiveness of the obtained results.

[0132] In an exemplary embodiment, after obtaining the cleaned cable tunnel point cloud data and before performing model fitting based on the pre-set fitting model information and the cable tunnel point cloud data, the method further includes:

[0133] Generating a corresponding feature matrix based on the cleaned cable tunnel point cloud data; generating a covariance matrix based on the feature matrix and performing eigenvalue decomposition on the covariance matrix to obtain feature information; the feature information includes eigenvalues and eigenvectors; sorting the feature information according to the magnitudes of the eigenvalues to obtain a sorted eigenvector matrix; determining target eigenvectors that satisfy a preset sorting sequence number from the sorted eigenvector matrix, and constructing a target eigenvector matrix according to the target eigenvectors; determining the cable tunnel feature data based on the target eigenvector matrix and the feature matrix.

[0134] Optionally, the terminal 102 generates a corresponding original feature matrix based on the cleaned cable tunnel point cloud data, calculates the mean of each column of data in the original feature matrix, subtracts the corresponding mean from the data in each column for centering to obtain the corresponding feature matrix, then generates a covariance matrix based on the feature matrix and performs eigenvalue decomposition on the covariance matrix to obtain feature information including eigenvalues and eigenvectors, sorts the feature information according to the magnitudes of the eigenvalues to obtain a sorted eigenvector matrix, determines target eigenvectors that satisfy a preset sorting sequence number from the sorted eigenvector matrix, constructs a target eigenvector matrix according to the target eigenvectors, and finally determines the cable tunnel feature data based on the target eigenvector matrix and the feature matrix. By performing efficient feature extraction (i.e., feature dimensionality reduction) in the above manner, the usability and effectiveness of the cable tunnel feature data are improved, thereby improving the fitting effect of subsequent model fitting, and at the same time accelerating the compression speed of data compression.

[0135] In one embodiment, model fitting is performed based on preset fitting model information and cable tunnel point cloud data to obtain a current target model, including:

[0136] Based on the preset fitting model information and cable tunnel feature data, multiple sample sets are randomly generated; model fitting is performed on any one of the sample sets to obtain the current target model.

[0137] Exemplarily, the terminal 102 randomly divides the cable tunnel feature data based on the number of the least data points required for forming the model included in the preset fitting model information to generate multiple sample sets, and then performs model fitting on any one of the sample sets to obtain the current target model. The random division enables the same data to be utilized multiple times, and the method of performing model fitting with different combinations improves the data utilization efficiency.

[0138] In an exemplary embodiment, as Figure 3 shown, a method for transmitting cable tunnel feature data is provided. Taking the method applied to the edge node 106 in Figure 1 as an example for illustration, it includes the following steps S301 to S304. Among them:

[0139] Step S301: Receive the compressed cable tunnel feature data, and generate a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained according to the method of any one of claims 1 to 5.

[0140] Optionally, the edge node 106 receives the compressed cable tunnel feature data sent by the terminal 102, organizes the cleaned feature data into a matrix A with a dimension of m×n, where m is the number of feature points and n is the dimension of the feature, and constructs the corresponding compressed feature matrix, laying a data foundation for the subsequent restoration of the compressed data.

[0141] Step S302: Perform singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix.

[0142] Exemplarily, the edge node 106 performs SVD:

[0143] Perform singular value decomposition on the feature matrix A, expressed as:

[0144] A = USV T

[0145] Wherein:

[0146] U is an m×m left singular vector matrix.

[0147] S is an m×n diagonal matrix, and the diagonal elements are singular values, which are usually arranged in descending order.

[0148] V T is the transpose of an n×n right singular vector matrix.

[0149] The magnitudes of the singular values provide an indication of the main characteristics of the data, laying a data foundation for subsequent truncation of the singular value matrix to select an important feature matrix.

[0150] Step S303: Truncate the singular value matrix according to preset threshold information to obtain a truncated singular value matrix.

[0151] Optionally, the edge node 106 selects the top k singular values and corresponding singular vectors according to a given threshold or feature importance, and constructs a truncated singular value matrix based on the selected singular vectors. Truncating the singular value matrix based on the magnitudes of the singular values can retain the main information of the data to the greatest extent, helping to completely restore the received compressed data.

[0152] Step S304: Restore the data of the truncated singular value matrix to obtain cable tunnel feature data.

[0153] Exemplarily, the edge node 106 restores the feature matrix:

[0154] Construct a restored feature matrix using the selected singular values and corresponding singular vectors :

[0155]

[0156] Where:

[0157] U k is the column in U corresponding to the top k singular values.

[0158] S k is a diagonal matrix composed of the top k singular values extracted from S.

[0159] is the row in V T corresponding to the top k singular values. Reconstructing based on partial features, the constructed feature matrix A′ can well reproduce the structure of the original matrix with a reasonable k value, thereby achieving a high data restoration quality. At the same time, this method avoids processing a large amount of data, speeds up the data processing speed, and ensures the timeliness of controlling the real-time situation of the cable tunnel.

[0160] In one embodiment, as Figure 4 shown, a specific implementation of a cable tunnel feature data transmission method is provided:

[0161] Step 1. Eliminate outliers retained in the point cloud mapping using the sparse outlier method, significantly improving the detection accuracy of the algorithm:

[0162] Step 1.1. Define the neighborhood. For each point, select its neighborhood ;

[0163] Step 1.2. Calculate the distance l from the point to each point in its neighborhood ij , if the neighborhood of point p is defined as the k nearest neighbor points, calculate the distances from this point p to these k points respectively, which are l p1 , l p2 , , l pk , where l pi represents the distance from point p to its i-th neighborhood point;

[0164] Step 1.3. Calculate the global mean and variance of the distances between all points and their neighborhoods;

[0165] Step 1.3.1. Global mean calculation. Assume there are n points in the point cloud data. For each point , its neighborhood has points ( is the pre-set neighborhood size), the distance from point to the point ( ) in its neighborhood is .

[0166] Step 1.3.2. Calculate the global mean , and its formula is as follows:

[0167]

[0168] Step 1.3.3. Calculate the global variance , and its formula is as follows:

[0169]

[0170] Among them, the global standard deviation is the square root of the global variance , that is .

[0171] Step 1.4. Remove outliers and remove the outlier points that meet the or conditions.

[0172] Step 2. Use the PCA (Principal Component Analysis) method to reduce the dimensionality of the feature data and extract the main feature data:

[0173] Step 2.1. Perform dimensionality reduction on the sample data. Let \(x\) represent the total number of samples and \(y\) represent the number of indicators in each sample. Then the original feature data can be calculated as follows:

[0174]

[0175] where: \(A\) ij is the \(j\)-th feature of the \(i\)-th sample. The sample mean of the feature data is: , and the sample variance of the feature data is: , then the following matrix is formed:

[0176]

[0177] PCA first preprocesses the data to obtain the principal components and performs principal component analysis: , first calculate the average value of each column , and then center the values in each column by subtracting the average value, that is:

[0178]

[0179] Step 2.2. Calculate the standard deviation matrix \(H\), that is:

[0180]

[0181] where is the eigenvalue in the sample.

[0182] For the \(y\) dimensions, let represent the symmetric covariance matrix, that is:

[0183]

[0184] where \(A\) ij The covariance between is:

[0185]

[0186] where and are the sample values;

[0187] When processing the covariance matrix, orthogonal eigenvectors can be extracted, and the principal components can be selected according to the contribution rate of each component. The criteria for selecting the principal components can include the eigenvalue criterion, the explained variance ratio criterion, and the minimum commonality criterion, etc. That is, the covariance matrix is:

[0188]

[0189] If the covariance is positive, it means that the data variables are positively correlated. If the covariance is negative, the data variables are negatively correlated;

[0190] Step 2.4 Extract the common variation among variables, determine the optimal projection direction using eigenvalue sorting, thereby select the number of principal components to retain the eigenvectors with larger variances, discard the noise, and clarify the selected number of principal components through a slope graph. The slope graph is as Figure 5 shown.

[0191] Step 3. Use the RANSAC algorithm to clean the point cloud dataset to obtain representative feature data. Repeat the calculation and randomly select data subsets to complete the selection of the optimal model and the points that conform to the optimal model:

[0192] Step 3.1. Assume that the proportion of "valid data" in the overall data is , and the formula is as follows:

[0193]

[0194] where represents valid data points, and represents invalid data points.

[0195] Step 3.2. When calculating a model with N sample points each time, and the minimum number of points required to solve the model is K, then the probability that at least one invalid data point exists among the selected points is ; therefore, the probability H that at least one success occurs in M samplings, that is, K valid data points can be sampled to calculate the correct model, is as follows:

[0196]

[0197] Step 3.3. Use the RANSAC algorithm to select the rough and unwanted invalid points on the surface of the point cloud in the form of the cable tunnel surface to reduce the influence caused by factors such as fluctuations during the scanning process or poor scanner calibration;

[0198] Step 3.4. After setting the initial model parameters, select the number of algorithm iterations and determine the smoothness of the point cloud processing. The algorithm will iteratively calculate the model parameters that are most suitable for the cable tunnel point cloud, select the invalid points for deletion, complete the removal of the mixed points, retain the representative cable tunnel data, and at the same time remove unnecessary noise points;

[0199] Step 3.5. The data cleaning of the RANSAC algorithm identifies the structures and features inside the cable tunnel, such as pipes, cables, obstacles, etc.

[0200] Step 4. Use the combination of MEC (Mobile Edge Computing) technology and IoT (Internet of Things) technology to transmit the compressed cable tunnel feature data:

[0201] Step 4.1. The MEC node deployed at the data source location collects the already compressed data. According to Shannon's theorem, the uplink rate at which the sensor r offloads the computing task to the MEC server q is:

[0202]

[0203] Where is the transmission power from the MEC server q to the sensor r; is the Gaussian white noise; is the uplink channel transmission bandwidth; is the channel gain of the transmission line sensor r transferred to the MEC server q;

[0204] Step 4.2. The MEC server q offloads the computing task to the cloud server e, and the rate is as follows:

[0205]

[0206] Where is the uplink channel transmission bandwidth of the cloud server e;

[0207] Step 4.3. Divide the task into two parts of arbitrary size, which can be executed in parallel on the sensor device and the mobile edge server. The task q is executed on the sensor r. The size of the task to be executed by the sensor r at time slot t is and the offloading ratio of the task is . If the CPU clock frequency of the sensor r is f, then the local time consumption is:

[0208]

[0209] Step 4.4. The total power consumption of local execution is related to the effective switching capacitance g of the chip structure, and the total power consumption is:

[0210]

[0211] Since the computing power of the MEC server is relatively strong, the execution time of offloading the computing task to the MEC server can be ignored;

[0212] Step 4.5. By combining the alternating direction method of multipliers (ADMM) and nonlinear fractional programming, an energy consumption minimization algorithm is obtained. The algorithm steps are as follows:

[0213] *1: Initialization: , , , , int i = 1, j = 1, l = 1;

[0214] *2: Reset the unloading rate and power of the sensor;

[0215] *3: int j = 1, update the ADMM parameters;

[0216] *4: j++, while , go to *3;

[0217] *5: Iteration ;

[0218] *6: l++, while , go to *3;

[0219] *7: Output ;

[0220] *8: i++, while , go to *2;

[0221] Step 5. Use SVD (Singular Value Decomposition) to restore the cable tunnel feature data cleaned by the RANSAC algorithm:

[0222] Step 5.1. Solve linear equations, and gradually solve for the compressed elements starting from the first equation. Solve a set of linear equations to restore its compressed elements. is a sub - matrix of

[0223]

[0224] where is the compressed element to be solved, is an element in the matrix .

[0225] Step 5.2. Restore the compressed elements. The row vectors of

[0226]

[0227] where is the compressed element to be solved, is the matrix the elements in

[0228] Step 5.3, Restoration and After restoring the compressed elements of the normalized reconstructed data matrix .

[0229] Compared with the prior art, the present application has the following advantages:

[0230] 1. Through efficient data cleaning, compression, and reconstruction, combined with the RANSAC algorithm, the cable data collected is compressed, reducing the amount of transmitted data and thus reducing the latency of data transmission.

[0231] 2. An efficient data transmission scheme based on MEC and IoT (combining edge computing and Internet of Things technologies for data transmission), and an energy efficiency optimization offloading and computing task allocation method (performing energy efficiency optimization through the ADMM algorithm).

[0232] 3. The above improvements overcome problems such as insufficient bandwidth, high latency, and data loss faced in traditional data processing and transmission, providing strong technical support for the intelligent management and maintenance of cable tunnels.

[0233] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown sequentially according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless otherwise clearly stated in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same time, but can be executed at different times, and the execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turns with at least a part of other steps or steps or stages in other steps.

[0234] Based on the same inventive concept, the embodiments of the present application also provide a cable tunnel feature data transmission device for implementing the cable tunnel feature data transmission method described above. The solution provided by this device for solving problems is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the cable tunnel feature data transmission device provided below can refer to the limitations on the cable tunnel feature data transmission method in the above text, and will not be repeated here.

[0235] In an exemplary embodiment, as Figure 6As shown in the figure, a cable tunnel feature data transmission device is provided, including: a point cloud data acquisition module 601, a model fitting module 602, a distance information acquisition module 603, an inlier acquisition module 604, a data compression module 605, and a data transmission module 606, where:

[0236] The point cloud data acquisition module 601 is used to acquire the cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data;

[0237] The model fitting module 602 is used to perform model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain the current target model; the current target model is used to represent the plane model or surface model corresponding to the cable tunnel;

[0238] The distance information acquisition module 603 is used to acquire the distance information from the cable tunnel point cloud data to the current target model;

[0239] The inlier acquisition module 604 is used to determine the current inlier set and the current inlier number corresponding to the current target model according to the distance information and the preset inlier threshold information;

[0240] The data compression module 605 is used to, if the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and inlier set, return to execute model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel;

[0241] The data transmission module 606 is used to transmit the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

[0242] In one embodiment, after acquiring the cable tunnel point cloud data of the cable tunnel and before performing model fitting according to the preset fitting model information and the cable tunnel point cloud data, the cable tunnel feature data transmission device further includes a data cleaning module, which is used to acquire the neighborhood corresponding to each data point in the cable tunnel point cloud data; acquire the neighborhood distance corresponding to each data point according to each data point and its corresponding neighborhood; obtain the global mean of the cable tunnel point cloud data based on the neighborhood distance, and obtain the global standard deviation of the cable tunnel point cloud data according to the neighborhood distance and the global mean; use the global mean, global standard deviation, and neighborhood distance to remove the abnormal data points in the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data.

[0243] In one embodiment, the data cleaning module is further configured to determine an abnormal neighborhood distance threshold corresponding to the cable tunnel point cloud data based on the global mean and the global standard deviation; determine data points with a neighborhood distance greater than the abnormal neighborhood distance threshold as abnormal data points; and remove the abnormal data points from the cable tunnel point cloud data to obtain the cleaned cable tunnel point cloud data.

[0244] In an exemplary embodiment, after obtaining the cleaned cable tunnel point cloud data and before performing model fitting based on the preset fitting model information and the cable tunnel point cloud data, the cable tunnel feature data transmission device further includes a data dimensionality reduction module, configured to generate a corresponding feature matrix according to the cleaned cable tunnel point cloud data; generate a covariance matrix based on the feature matrix, perform eigenvalue decomposition on the covariance matrix to obtain feature information; the feature information includes eigenvalues and eigenvectors; sort the feature information according to the magnitudes of the eigenvalues to obtain a sorted eigenvector matrix; determine target eigenvectors that meet a preset sorting sequence number from the sorted eigenvector matrix, and construct a target eigenvector matrix according to the target eigenvectors; and determine the cable tunnel feature data based on the target eigenvector matrix and the feature matrix.

[0245] In one embodiment, the model fitting module 602 is further configured to randomly generate a plurality of sample sets based on the preset fitting model information and the cable tunnel feature data; and perform model fitting on any one of the sample sets to obtain the current target model.

[0246] In an exemplary embodiment, as Figure 7 shown, there is provided a cable tunnel feature data transmission device, which is applied to an edge node closest to the cable tunnel, and includes: a data receiving module 701, a decomposition module 702, a truncation module 703, and a data restoration module 704, where:

[0247] The data receiving module 701 is configured to receive the compressed cable tunnel feature data and generate a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained according to the method of any one of claims 1 to 5.

[0248] The decomposition module 702 is configured to perform singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix.

[0249] The truncation module 703 is configured to truncate the singular value matrix according to the preset threshold information to obtain a truncated singular value matrix.

[0250] The data restoration module 704 is configured to perform data restoration on the truncated singular value matrix to obtain the cable tunnel feature data.

[0251] Each module in the above cable tunnel feature data transmission device can be implemented in whole or in part by software, hardware, or a combination thereof. Each of the above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to each of the above modules.

[0252] In an exemplary embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 8 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store cable tunnel point cloud data, distance information, inlier sets, the number of inliers, and compressed cable tunnel feature data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a method for transmitting cable tunnel feature data.

[0253] In an exemplary embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 9As shown in the figure. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit, and an input device. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface, the display unit, and the input device are connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a method for transmitting cable tunnel feature data. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the outer shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0254] Those skilled in the art can understand that Figure 8 and Figure 9 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0255] In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, it implements the method for transmitting cable tunnel feature data in the above embodiment.

[0256] In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, it implements the method for transmitting cable tunnel feature data in the above embodiment.

[0257] In an embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, it implements the method for transmitting cable tunnel feature data in the above embodiment.

[0258] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0259] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, artificial intelligence (AI) processors, etc., and are not limited thereto.

[0260] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this application.

[0261] The above-described embodiments merely represent several implementation manners of this application. The description is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of this application. It should be noted that for those of ordinary skill in the art, without departing from the concept of this application, several modifications and improvements can still be made, and these all belong to the protection scope of this application. Therefore, the protection scope of this application should be subject to the appended claims.

Claims

1. A method for transmitting characteristic data of a cable tunnel, characterized in that, The method includes: Obtaining point cloud data of a cable tunnel; the point cloud data of the cable tunnel includes at least one of point cloud data of pipelines, point cloud data of cables, and point cloud data of obstacles; Performing model fitting based on preset fitting model information and the point cloud data of the cable tunnel to obtain a current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel; Obtaining distance information from the point cloud data of the cable tunnel to the current target model; Determining a current inlier set and a current inlier number corresponding to the current target model according to the distance information and preset inlier threshold information; If the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and inlier set, and return to execute model fitting based on preset fitting model information and the point cloud data of the cable tunnel to obtain a new current target model and determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel; Transmitting the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

2. The method according to claim 1, characterized in that After obtaining the point cloud data of the cable tunnel and before performing model fitting based on preset fitting model information and the point cloud data of the cable tunnel, the method includes: Obtaining the neighborhood corresponding to each data point in the point cloud data of the cable tunnel; Obtaining the neighborhood distance corresponding to each data point according to each data point and its corresponding neighborhood; Obtaining the global mean of the point cloud data of the cable tunnel based on the neighborhood distance, and obtaining the global standard deviation of the point cloud data of the cable tunnel according to the neighborhood distance and the global mean; Using the global mean, the global standard deviation, and the neighborhood distance to remove abnormal data points in the point cloud data of the cable tunnel, and obtaining the cleaned point cloud data of the cable tunnel.

3. The method according to claim 2, wherein The using the global mean, the global standard deviation, and the neighborhood distance to remove abnormal data points in the point cloud data of the cable tunnel and obtaining the cleaned point cloud data of the cable tunnel includes: Determining an abnormal neighborhood distance threshold corresponding to the point cloud data of the cable tunnel based on the global mean and the global standard deviation; Determining the data points with a neighborhood distance greater than the abnormal neighborhood distance threshold as abnormal data points; Removing the abnormal data points in the point cloud data of the cable tunnel to obtain the cleaned point cloud data of the cable tunnel.

4. The method according to claim 2 or 3, characterized in that, After obtaining the cleaned point cloud data of the cable tunnel and before performing model fitting based on preset fitting model information and the point cloud data of the cable tunnel, the method further includes: Generating a corresponding feature matrix according to the cleaned point cloud data of the cable tunnel; Generating a covariance matrix based on the feature matrix, and performing eigenvalue decomposition on the covariance matrix to obtain feature information; the feature information includes eigenvalues and eigenvectors; Sorting the feature information according to the magnitudes of the eigenvalues to obtain a sorted eigenvector matrix; Determine a target feature vector that meets a preset sorting serial number from the sorted feature vector matrix, and construct a target feature vector matrix based on the target feature vector; Based on the target feature vector matrix and the feature matrix, determine the cable tunnel feature data.

5. The method according to claim 4, characterized in that, The model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain the current target model includes: Based on the preset fitting model information and the cable tunnel feature data, randomly generate a plurality of sample sets; Perform model fitting on any one of the sample sets to obtain the current target model.

6. A method for transmitting characteristic data of a cable tunnel, characterized in that, Applied to the edge node closest to the cable tunnel, the method includes: Receive the compressed cable tunnel feature data, and generate a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained according to the method described in any one of claims 1 to 5; Perform singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix; Truncate the singular value matrix according to the preset threshold information to obtain a truncated singular value matrix; Restore the data of the truncated singular value matrix to obtain the cable tunnel feature data.

7. A cable tunnel feature data transmission device, characterized in that, The device includes: A point cloud data acquisition module for acquiring cable tunnel point cloud data of the cable tunnel; the cable tunnel point cloud data includes at least one of pipeline point cloud data, cable point cloud data, and obstacle point cloud data; A model fitting module for performing model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain the current target model; the current target model is used to represent a plane model or a surface model corresponding to the cable tunnel; A distance information acquisition module for acquiring the distance information from the cable tunnel point cloud data to the current target model; An inlier acquisition module for determining a current inlier set and a current inlier number corresponding to the current target model according to the distance information and the preset inlier threshold information; A data compression module, if the current inlier number is greater than the previously recorded inlier number, update the previously recorded inlier number and inlier set, return to execute model fitting according to the preset fitting model information and the cable tunnel point cloud data to obtain a new current target model and determine a new current inlier set and a new current inlier number, until the set maximum number of iterations is reached, and determine the current inlier set as the compressed cable tunnel feature data corresponding to the cable tunnel; A data transmission module for transmitting the compressed cable tunnel feature data to the edge node closest to the cable tunnel.

8. A cable tunnel feature data transmission device, characterized in that Applied to the edge node closest to the cable tunnel, the device includes: A data receiving module for receiving the compressed cable tunnel feature data and generating a corresponding compressed feature matrix according to the compressed cable tunnel feature data; the compressed cable tunnel feature data is obtained according to the method described in any one of claims 1 to 5; A decomposition module for performing singular value decomposition on the compressed feature matrix to obtain a singular value matrix corresponding to the compressed feature matrix; A truncation module, configured to truncate the singular value matrix according to preset threshold information to obtain a truncated singular value matrix; A data restoration module, configured to restore data of the truncated singular value matrix to obtain cable tunnel feature data.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 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 according to any one of claims 1 to 6 are implemented.

Images