CN111382767B - Power line point identification method, device, equipment, computer equipment and storage medium - Google Patents
Power line point identification method, device, equipment, computer equipment and storage medium Download PDFInfo
- Publication number
- CN111382767B CN111382767B CN201811637018.7A CN201811637018A CN111382767B CN 111382767 B CN111382767 B CN 111382767B CN 201811637018 A CN201811637018 A CN 201811637018A CN 111382767 B CN111382767 B CN 111382767B
- Authority
- CN
- China
- Prior art keywords
- power line
- data
- point cloud
- point
- cloud data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a power line point identification method, a device, equipment, computer equipment and a storage medium. The method comprises the following steps: the method comprises the steps of identifying target point cloud data by acquiring the target point cloud data and adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to acquire first power line point data; then, original point cloud data is obtained, and the first power line point data and the original point cloud data are processed by adopting a nearest neighbor search method to obtain second power line point data; and summarizing the first power line point data and the second power line point data to obtain third power line point data. By adopting the method, the noise can be effectively reduced, the extraction precision can be improved, and the continuity of the power line points can be improved.
Description
Technical Field
The present application relates to the field of power transmission line technologies, and in particular, to a method, an apparatus, a device, a computer device, and a storage medium for identifying a power line point.
Background
With the development of power transmission line technology, a power line extraction technology is developed to identify and distinguish power line points and non-power line points.
The existing power line point extraction technology comprises the steps of preprocessing power point cloud data, projecting the power point cloud data to a two-dimensional plane, extracting a power line by applying an iterative Hough change method to the plane data, or dividing power line LiDAR data into a plurality of ellipsoid small blocks to extract a power line point set, fitting point data on the same sub-conductor by a point aggregation method, and iteratively detecting the power line and carrying out three-dimensional modeling by a simplified catenary equation parameter solving process.
However, the conventional method has a problem of low accuracy.
Disclosure of Invention
In view of the above, it is necessary to provide a power line point identification method, apparatus, device, computer device and storage medium for solving the above technical problems.
A power line point identification method, the method comprising:
acquiring target point cloud data, and identifying the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to obtain first power line point data;
acquiring original point cloud data, and processing the first power line point data and the original point cloud data by adopting a nearest neighbor search method to acquire second power line point data;
and summarizing the first power line point data and the second power line point data to obtain third power line point data.
In one embodiment, the obtaining target point cloud data, identifying the target point cloud data by using a surface curvature estimation algorithm and a local dimension multi-scale algorithm, and obtaining first power line point data includes:
identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data;
and identifying residual point cloud data by adopting a local dimension multi-scale algorithm to obtain fifth power line point data, and obtaining the first power line point data according to the fourth power line point data and the fifth power line point data.
In one embodiment, the identifying the target point cloud data by using a curved surface curvature estimation algorithm to obtain fourth power line point data and remaining point cloud data includes:
selecting a reference point in the target point cloud data, and calculating point data of the reference point in the target point cloud data in a first preset range;
performing tangent plane fitting on the point data in the first preset range to obtain a plane equation;
performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadric surface equation;
and determining the fourth power line point data according to the quadric surface equation.
In one embodiment, the performing tangential plane fitting on the point data in the first preset range to obtain a plane equation includes:
constructing a local coordinate system at a reference point in the target point cloud data;
and approximating the point data on a tangent plane constructed by the local coordinate system, determining a plane normal vector and obtaining the plane equation according to a characteristic vector estimation method.
In one embodiment, the performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadric surface equation includes:
converting the local coordinate system to an original coordinate system, and acquiring a spherical equation of point data in a first preset range under the original coordinate system;
and obtaining a plane normal vector, and obtaining the quadric surface equation according to the plane normal vector and a singular value decomposition method in a linear least square method.
In one embodiment, the identifying remaining point cloud data by using a local dimension multi-scale algorithm to obtain fifth power line point data, and obtaining the first power line point data according to the fourth power line point data and the fifth power line point data includes:
acquiring the residual point cloud data, and calculating the corresponding characteristic values of the points in the residual point cloud data in a second preset range;
constructing local dimension multi-scale features according to the feature values;
and determining the fifth power line point data according to the local dimension multi-scale feature.
In one embodiment, the obtaining of the original point cloud data, and processing the first power line point data and the original point cloud data by using a nearest neighbor search method, and the obtaining of the second power line point data includes:
mixing the first power line point data into the original point cloud data, and selecting points in a third preset range of the first power line point data;
and summarizing the points in the third preset range to obtain second power line point data.
In one embodiment, the obtaining target point cloud data, identifying the point cloud data by using a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm, and before obtaining the first power line point data, the method includes:
acquiring original point cloud data, and performing rarefaction processing on the original point cloud data to obtain target point cloud data.
A power line point identification apparatus, the apparatus comprising:
the first power line point data acquisition module is used for acquiring target point cloud data, identifying the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm and acquiring first power line point data;
the second power line point data acquisition module is used for acquiring original point cloud data, and processing the first power line point data and the original point cloud data by adopting a nearest neighbor search method to acquire second power line point data;
and the third power line point data acquisition module is used for summarizing the first power line point data and the second power line point data to obtain third power line point data.
A power line point identification device, the device at least comprises a power line point identification device.
A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method as described in any one of the above when executing the computer program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method as set forth in any one of the preceding claims.
According to the power line point identification method, the device, the equipment, the computer equipment and the storage medium, the target point cloud data is obtained, and the target point cloud data is identified by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to obtain first power line point data; then, original point cloud data is obtained, and the first power line point data and the original point cloud data are processed by adopting a nearest neighbor search method to obtain second power line point data; and summarizing the first power line point data and the second power line point data to obtain third power line point data. By the method, the noise can be effectively reduced, the extraction precision can be improved, and the continuity of the power line points can be improved.
Drawings
Fig. 1 is a diagram illustrating an exemplary embodiment of a power line point identification method;
FIG. 2 is a flow chart illustrating a method for identifying power line points in one embodiment;
FIG. 3 is a schematic flowchart of step S1 in one embodiment;
FIG. 4 is a flowchart illustrating step S11 in another embodiment;
FIG. 5 is a flowchart illustrating step S112 in another embodiment;
FIG. 6 is a schematic illustration of a reference point coordinate system in another embodiment;
FIG. 7 is a diagram illustrating the distribution of feature vectors in another embodiment;
FIG. 8 is a flowchart illustrating step S113 in another embodiment;
FIG. 9 is a flowchart illustrating step S12 in another embodiment;
FIG. 10 is a schematic diagram of the spatial shape of different point cloud data according to another embodiment;
FIG. 11 (a) is a schematic diagram of a 220KV power line corridor raw point cloud scene;
FIG. 11 (b) is a schematic view of a 500KV power line corridor raw point cloud scene;
FIG. 11 (c) is a schematic diagram of a power line corridor raw point cloud scene at 750KV voltage;
FIG. 12 (a) is a schematic representation of an estimated curvature of a 220KV line corridor raw point cloud;
FIG. 12 (b) is a schematic representation of the curvature of the original point cloud of the power line corridor at 500KV voltage after estimation;
FIG. 12 (c) is a schematic representation of the curvature estimation of the original point cloud of the power line corridor at 750KV voltage;
FIG. 13 (a) is a schematic representation of a sorted cloud of raw points in a power line corridor at 220 KV;
FIG. 13 (b) is a schematic representation of a 500KV power line corridor raw point cloud after classification;
FIG. 13 (c) is a schematic representation of a 750KV line corridor raw point cloud after classification;
fig. 14 is a block diagram showing the structure of a power line point identification apparatus according to an embodiment;
FIG. 15 is a diagram showing an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad application.
The power line point identification method provided by the application can be applied to the application environment shown in fig. 1. Wherein the terminal 102 communicates with the server 104 via a network. The method comprises the steps that a terminal 102 obtains target point cloud data and transmits the target point cloud data to a server 104, and the server 104 identifies the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to obtain first power line point data; then, original point cloud data is obtained, and the first power line point data and the original point cloud data are processed by adopting a nearest neighbor search method to obtain second power line point data; and summarizing the first power line point data and the second power line point data to obtain third power line point data. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, a power line point identification method is provided, which is described by taking the method as an example applied to the server in fig. 1, and includes the following steps:
step S1: obtaining target point cloud data, and identifying the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to obtain first power line point data.
Specifically, the target point cloud data refers to point cloud data with equivalent interval uniform distribution obtained by preprocessing the laser point cloud data.
The first power line point cloud data is point data obtained by sequentially performing a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm on the target point cloud data.
Step S2: and acquiring original point cloud data, and processing the first power line point data and the original point cloud data by adopting a nearest neighbor search method to acquire second power line point data.
Specifically, the original point cloud data is laser point cloud data acquired by a laser radar. In addition, the unprocessed laser point cloud data is discrete three-dimensional point cloud data. The laser point cloud data comprises power line point cloud data and non-power line point cloud data, wherein the non-power line point cloud data comprises surface points, vegetation points, tower pole points, building points and the like.
The nearest neighbor searching method is to search a point in a certain distance range from a certain point as the center of a circle. The method has the advantages of simplicity and high processing speed.
The second power line point data is point data obtained by processing the first power line point data and the original point cloud data by using a nearest neighbor search method.
And step S3: and summarizing the first power line point data and the second power line point data to obtain third power line point data.
Specifically, the third power line point data refers to point data obtained by mixing the first power line point data and the second power line point data.
According to the power line point identification method, target point cloud data is obtained, and a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm are adopted to identify the point cloud data, so that first power line point data is obtained; then, original point cloud data is obtained, and the first power line point data and the original point cloud data are processed by adopting a nearest neighbor search method to obtain second power line point data; and summarizing the first power line point data and the second power line point data to obtain third power line point data. By the method, the noise can be effectively reduced, the extraction precision can be improved, and the continuity of the power line point can be improved.
In one embodiment, in conjunction with fig. 3, the step S1 includes:
step S11: and identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data.
Specifically, the point cloud category is identified by adopting a curved surface curvature estimation algorithm, namely, the power line points are preliminarily identified by using the characteristic that the unknown coefficient of the quadric surface equation cannot be calculated according to the point set of linear distribution, and the power line lead points are further extracted. Because point cloud data is a point set in discrete distribution and has no topological information, the curvature of the point set can be only approximately estimated
The fourth power line point data refers to point data obtained by performing curved surface curvature estimation algorithm identification on the target point cloud data, and the remaining point cloud data refers to point data left by subtracting the fourth power electricity data from the target point cloud data.
Step S12: and identifying residual point cloud data by adopting a local dimension multi-scale algorithm to obtain fifth power line point data, and obtaining the first power line point data according to the fourth power line point data and the fifth power line point data.
Specifically, the local dimension multi-scale algorithm refers to local dimension characteristics of the point cloud under different scales. That is, the geometrical characteristics of the point cloud are studied through "local dimensions", so that the point cloud appears as a line, a plane, or distributed in the whole area at a given position and scale. And for the same part of point cloud, adopting different scales, and respectively presenting different dimensional characteristics on the internal points of the ball. Dimension characteristics under different scales can be combined to serve as a basis for distinguishing different object classes.
And the fifth power line point data refers to power line point data obtained by performing a local dimension multi-scale algorithm on the residual point cloud data.
In one embodiment, in conjunction with fig. 4, the step S11 includes:
step S111: and selecting a reference point in the target point cloud data, and calculating the point data of the reference point in the target point cloud data in a first preset range.
Specifically, the reference point refers to any point in the target point cloud data. Because the target point cloud data is discrete point cloud data, when the discrete point cloud data is represented by a curved surface, points in a range adjacent to any point p need to be calculated, and plane fitting is carried out on the points in the first preset range. The first preset range refers to a range of fixed values from the reference point p, wherein the fixed values are set according to actual needs, such as 3cm, 4cm, 5cm and the like.
Step S112: and performing tangent plane fitting on the point data in the first preset range to obtain a plane equation.
Specifically, the principle of the plane fitting is to approximate points in the vicinity of any point on a local tangent plane, that is, to approximate point data in the first preset range of the reference point on a local tangent plane. Then, the plane equation is fitted by the least square method.
Step S113: and performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadric surface equation.
Specifically, the point data in the first preset range is subjected to surface fitting, a proper equation needs to be selected for fitting to obtain a quadric surface equation, and the point data in the first preset range is subjected to surface fitting by adopting a spherical equation.
Step S114: and determining fourth power line point data according to the quadric surface equation.
Specifically, if a quadric equation of the point data in the first preset range is obtained according to the surface fitting, it is indicated that the point data in the first preset range has curvature characteristics, that is, the point data in the first preset range is power line point data.
In one embodiment, in conjunction with fig. 5, the step S112 includes:
step S1121: constructing a local coordinate system at a reference point in the target point cloud data;
step S1122: and approximating the point data on a tangent plane constructed by the local coordinate system, determining a plane normal vector and obtaining the plane equation according to a characteristic vector estimation method.
Specifically, in connection with fig. 6, a local coordinate system (u, v, h) is constructed at a point p, which is a parametric representation in the local coordinate system, the point p being the origin of coordinates, the h-axis being the normal vector N direction at the point p, u and v being orthogonal and on the tangent plane τ.
As can be seen from higher mathematical space analytic geometry knowledge, if the plane equation is a · x + b · y + c · z + d =0, the plane normal vector is (a, b, c), and the plane fitting is solved according to the eigenvector estimation method.
In connection with fig. 7, the eigenvalues and eigenvectors are computed by means of a covariance matrix. Among them, one of the most intuitive interpretations of the eigenvector of the covariance matrix is that it always points in the direction where the data variance is largest. Assuming all data points are within an ellipsoid, v1 is the first eigenvector and λ 1 is its corresponding eigenvalue, pointing in the direction of maximum variance of the data. v2 is perpendicular to v1 and is the eigenvector with the largest data variance in this direction. v3 is perpendicular to both v1 and v2 and is the eigenvector with the largest variance of the data in this direction.
When solving the eigenvalue λ, the eigenvalues are generally sorted (in ascending order), the minimum eigenvalue (first eigenvalue) is the eigenvalue corresponding to the normal direction, and the eigenvector of the eigenvalue represents the normal direction. For three-dimensional space point cloud, after solving the covariance matrix by using Eigen, the eigenvalues and eigenvectors are arranged in an ascending order, the obtained vector is the eigenvector in the normal direction, normalization processing is generally required for eigenvectors v1, v2 and v3, two of the eigenvectors are generally normalized respectively, and the other eigenvector is obtained by cross multiplication.
The obtained normal vectors v1, v2, v3 correspond to the plane equation coefficients (a, b, c), and finally the plane equation is obtained:
a·x+b·y+c·z+d＝0。
in one embodiment, in conjunction with fig. 8, the step S113 includes:
step S1131: converting the local coordinate system to an original coordinate system, and acquiring a spherical equation of point data in a first preset range under the original coordinate system;
step S1132: and obtaining a plane normal vector, and obtaining the quadric surface equation according to the plane normal vector and a singular value decomposition method in a linear least square method.
Specifically, the tangent plane is constructed under a local coordinate system (u, v, h), which is converted to the original coordinate system (x, y, z), i.e., the unknowns (u, v, h) in the plane equation are replaced by (x, y, z).
After the tangent plane equation is obtained, surface fitting can be further carried out, and the unknown coefficient of the surface equation is calculated.
The fitting is performed by using a spherical approximation curved surface. The common spherical equation is:
(x-x 0 ) 2 +(y-y 0 ) 2 +(z-z 0 ) 2 ＝R 2
wherein x is 0 ,y 0 ,z 0 R is a spherical parameter, o = (x) 0 ,y 0 ,z 0 ) Is the center of the sphere and R is the radius.
The spherical equation is generally of the form:
F(x,y,z)＝x2+y2+z2+ax+by+cz+d＝0
the coefficient relationship of the two expression modes is
The general form of the spherical equation contains constant terms
And the coefficients a, b, c and d are independent, and the Singular Value Decomposition (SVD) method in the linear least square method can be adopted to solve the spherical fitting.
The objective function of the spherical fitting is an equation, that is, the solved quadric equation is:
Ax＝b
wherein
x＝(a,b,c,d) T ，
In one embodiment, in conjunction with fig. 9, the step S12 includes:
step S121: acquiring the residual point cloud data, and calculating the corresponding characteristic values of points in the residual point cloud data in a second preset range;
step S122: constructing local dimension multi-scale features according to the feature values;
step S123: and determining fifth power line point data according to the local dimension multi-scale features.
Specifically, the remaining point cloud data refers to the point cloud data of the target point minus the first power line point data, and the remaining point cloud data.
For residual point cloud data, firstly, calculating a characteristic value and a characteristic vector corresponding to each point neighborhood radius, and then defining dimension characteristics according to characteristic values lambda 1, lambda 2 and lambda 3.
The local dimension multi-scale feature construction process is as follows: the scale of the point set is divided according to the diameter of a ball taking an object point as a center, the adjacent ball of each point is calculated according to a certain scale, and the size of the scale can be artificially set according to the size of an actual classified entity. And PCA calculations are performed for each of the nearby balls to find directions within the nearby balls. If λ 1 > λ 2 ≈ λ 3, it means that the point set is mainly distributed in one direction.
With reference to fig. 10, most of the vegetation appears as spherical objects in the point cloud space, and the eigenvalues in the x, y and z directions are relatively close to each other; the power line is represented as a linear target in the point cloud space, and the characteristic value in only one direction is large. Therefore, only selecting the linear object can eliminate the rest points.
In one embodiment, the step S2 includes:
step S21: mixing the first power line point data into the original point cloud data, and selecting points in a third preset range of the first power line point data;
step S22: and summarizing the points in the third preset range to obtain second power line point data.
Specifically, the third preset range refers to a range of fixed values from any point of the first power line point data, where the point in the third preset range includes the first power line point data, and the fixed value is set according to actual needs, such as 3cm, 4cm, 5cm, and the like.
Through the steps, the power line points can be classified from the point cloud, but the point cloud thinning pretreatment is firstly carried out before the power line classification in order to apply the curvature characteristics. This reduces the point cloud accuracy while reducing the data volume and highlighting the curvature characteristic of the power line. In order to ensure the quality of point cloud data, the extracted power line points are merged into the original point cloud, and the precision of the point cloud is restored. It should be noted that the category attribute of the power line is preserved while merging.
Because the marked power line points obtained by extraction are subjected to spatial distance thinning and merged into the point cloud with relatively high original density, the marked power line points are adjacent to the real power line points which are not classified. To solve this problem, the power line points can be completely classified from the original point cloud by dividing the points within a certain distance range close to the marked power line points into power lines through simple nearest neighbor search.
In one embodiment, the step S1 is preceded by:
and step S4: and acquiring original point cloud data, and performing rarefying treatment on the original point cloud data to obtain target point cloud data.
Specifically, the original discrete three-dimensional point cloud data is thinned, and the data accuracy and quality are ensured, and meanwhile, the data volume of the original point cloud is reduced as much as possible. According to the method, spatial Subsampling (Spatial Subsampling) is adopted, high-density and high-redundancy point clouds can be sampled according to a certain interval distance, equivalent interval distances among points in the point clouds after sparse sampling are uniformly distributed, redundant point clouds are effectively removed, and original object feature points of the point clouds can be kept as far as possible.
After the data is thinned, the quantity of the data is greatly reduced, the basic shape characteristics of the original graph or curve can be basically reflected, and the space and the time can be saved for further processing.
In one embodiment, three different voltage class (220 kV, 500kV and 750 kV) transmission lines were selected for the experiments herein. The case of airborne LiDAR data used is shown in table 1.
TABLE 1 Point cloud data parameters of Experimental area
Referring to fig. 11 (a) - (c), a 220kV transmission line comprises 2 steps of data, 3-base towers in total, and a tower height of about 48m, each step comprises 6 phases of transmission lines, and each phase has 2 split conductors. The 500kV power transmission line comprises 3-gear data, 4 base-pole towers in total, the height of each tower is about 50m, each transmission line comprises 3 phases, and each phase is provided with 4 split conductors. The 750kV transmission line comprises 2 grades of data, 3 base towers in total, the height of the tower is about 52m, each grade comprises 3 phases, and each phase is provided with 6 classified conductors. In addition, the top layer of the 3 transmission lines comprises two ground wires.
Firstly, carrying out spatial distance thinning pretreatment on the original point cloud so as to enable the foot points of the point cloud to be uniform in interval. Therefore, the laser point cloud collected by the airborne LiDAR is a point set in discrete irregular distribution, the density distribution of the point cloud is not uniform, and the curved surface curvature features of each foot point adjacent point set directly constructed on the original point cloud are distributed in a scattered manner, so that the curvature features cannot be directly utilized for classification. After the space thinning, the point cloud resampling is performed on the original data, the intervals between the points are uniform, and a homogeneous environment can be provided for curvature feature extraction. In addition, the data volume can be reduced after thinning, and the efficiency of data processing can be improved. The experimental data were first subjected to spatial distance thinning with the distance radius set at 0.8-1.0m. Then, the curvature was estimated and the curvature radius was set to 1.6 to 2.0m. The curvature estimation results are shown in fig. 12 (a) - (c), where the gray white portion is a point without curvature value and the blue portion is a point with curvature value.
It is not near every point that the quadric surface can be successfully constructed and the curvature value can be calculated. The power transmission line corridor ground thing is mostly ground and vegetation, and most vegetation trees or ground target go up the nearly close point set of point and be mostly plane or spherical point set, then can be normal construct quadric surface, therefore mostly be blue in the picture. In addition, the gray-white part in the figure is mostly a power line point, which is determined by the linear characteristic of the power line. The neighboring point set of points on the power line is still a linear point set, so that a quadric surface cannot be constructed, and the curvature cannot be calculated. Even if a quadratic surface can be constructed with adjacent power lines to calculate the curvature, the minimum distance between power lines on the same layer or adjacent power lines above and below is more than 6 m. By setting a proper curvature estimation radius, the power line points can be identified according to the curvature characteristics, but the power line points contain a small part of other ground object noise points.
For high-voltage transmission lines, the objects in the corridor of the transmission line are mostly vegetations. Cloth filtering is used herein to separate ground points from non-ground points. The non-ground points mainly comprise vegetation points and tower points, the positions of the towers are positioned according to the prior tower coordinates, and cylindrical space classification tower points are constructed. As fig. 13 (a) - (c) show the final classification result of the corridor, although the colors of the diagram are limited by the application document, the power line and the non-power line can be distinguished from each other.
It should be understood that, although the steps in the flowchart of fig. 2 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a portion of the steps in fig. 2 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 14, there is provided a power line point identifying apparatus including: a first power line point data acquisition module, a second power line point data acquisition module, and a third power line point data acquisition module, wherein:
the first power line point data acquisition module 10 is configured to acquire target point cloud data, and identify the target point cloud data by using a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to acquire first power line point data;
a second power line point data obtaining module 20, configured to obtain original point cloud data, and process the first power line point data and the original point cloud data by using a nearest neighbor search method to obtain second power line point data;
and the third power line point data obtaining module 30 is configured to sum the first power line point data and the second power line point data to obtain third power line point data.
In one embodiment, the first power line point data obtaining module 10 includes:
a fourth power line point data obtaining module 101, configured to identify the target point cloud data by using a curved surface curvature estimation algorithm, so as to obtain fourth power line point data and remaining point cloud data;
the fifth power line point data obtaining module 102 is configured to identify remaining point cloud data by using a local dimension multi-scale algorithm to obtain fifth power line point data, and obtain first power line point data according to the fourth power line point data and the fifth power line point data.
In one embodiment, the fourth power line point data obtaining module 101 includes:
a reference point calculation module 1011, configured to select a reference point in the target point cloud data, and calculate point data of the reference point in the target point cloud data within a first preset range;
a plane equation obtaining module 1012, configured to perform tangent plane fitting on the point data in the first preset range to obtain a plane equation;
a quadric surface obtaining module 1013, configured to perform surface fitting on the point data within the first preset range according to the plane equation to obtain a quadric surface equation;
the first determining module 1014 is configured to determine the fourth power line point data according to the quadric equation.
In one embodiment, the plane equation obtaining module 1012 includes:
a local coordinate system constructing module 1012a, configured to construct a local coordinate system at a reference point in the target point cloud data;
a second determining module 1012b, configured to approximate the point data on a tangent plane constructed by the local coordinate system, determine a plane normal vector, and obtain the plane equation according to a feature vector estimation method.
In one embodiment, the quadric surface acquisition module 1013 comprises:
the spherical equation obtaining module 1013a is configured to convert the local coordinate system to an original coordinate system and obtain a spherical equation of point data in a first preset range under the original coordinate system;
and a third determining module 1013b, configured to obtain a plane normal vector, and obtain the quadric equation according to the plane normal vector and a singular value decomposition method in a linear least square method.
In one embodiment, the fifth power line point data obtaining module 102 includes:
a feature value calculation module 1021, configured to obtain the remaining point cloud data, and calculate a feature value corresponding to a point in the remaining point cloud data within a second preset range;
a local dimension multi-scale feature construction module 1022, configured to construct a local dimension multi-scale feature according to the feature value;
a fourth determining module 1023, configured to determine the fifth power line point data according to the local-dimension multi-scale feature.
In one embodiment, the second power line point data obtaining module 20 includes:
a data mixing module 201, configured to mix the first power line point data into the original point cloud data, and select a point in a third preset range of the first power line point data;
and a fifth determining module 202, configured to summarize the points within the third preset range to obtain second power line point data.
In one embodiment, the first power line point data obtaining module 10 previously includes:
and the rarefying processing module 40 is configured to acquire original point cloud data and perform rarefying processing on the original point cloud data to obtain target point cloud data.
In one embodiment, a power line point identification apparatus includes at least one power line point identification device.
For the specific definition of the power line point identification apparatus, reference may be made to the above definition of the power line point identification method, and details are not repeated here. Each module in the above-described power line point identification apparatus may be wholly or partially implemented by software, hardware, and a combination thereof. The modules can be embedded in a hardware form or independent of a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 14. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile 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 an operating system and computer programs in the non-volatile storage medium. The database of the computer device is for storing power line identification data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a power line point identification method.
Those skilled in the art will appreciate that the architecture shown in fig. 15 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In an embodiment, a computer device is provided, comprising a memory in which a computer program is stored and a processor which, when executing the computer program, carries out the steps as described in the above method.
In an embodiment, a computer-readable storage medium is provided, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method as described above.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is specific and detailed, but not to be understood as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (9)
1. A power line point identification method, the method comprising:
acquiring target point cloud data, and identifying the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm to obtain first power line point data; identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data; identifying residual point cloud data by adopting a local dimension multi-scale algorithm to obtain fifth power line point data, and obtaining the first power line point data according to the fourth power line point data and the fifth power line point data; the identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data comprises the following steps: selecting a reference point in the target point cloud data, and calculating point data of the reference point in the target point cloud data in a first preset range; performing tangential plane fitting on the point data in the first preset range to obtain a plane equation; performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadric surface equation; determining the fourth power line point data according to the quadric surface equation; the step of identifying the remaining point cloud data by adopting a local dimension multi-scale algorithm to obtain fifth power line point data comprises the following steps: acquiring the residual point cloud data, and calculating the corresponding characteristic values of points in the residual point cloud data in a second preset range; constructing local dimension multi-scale features according to the feature values; determining the fifth power line point data according to the local dimension multi-scale feature;
acquiring original point cloud data, and processing the first power line point data and the original point cloud data by adopting a nearest neighbor search method to acquire second power line point data;
and summarizing the first power line point data and the second power line point data to obtain third power line point data.
2. The method of claim 1, wherein performing tangent plane fitting on the point data in the first preset range to obtain a plane equation comprises:
constructing a local coordinate system at a reference point in the target point cloud data;
and fitting the point data on a tangent plane constructed by the local coordinate system, determining a plane normal vector and obtaining the plane equation according to a characteristic vector estimation method.
3. The method of claim 2, wherein performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadratic surface equation comprises:
converting the local coordinate system to an original coordinate system, and acquiring a spherical equation of point data in a first preset range under the original coordinate system;
and obtaining a plane normal vector, and obtaining the quadric surface equation according to the plane normal vector and a singular value decomposition method in a linear least square method.
4. The method of claim 1, wherein the obtaining of the original point cloud data, the processing of the first power line point data and the original point cloud data using a nearest neighbor search method, and the obtaining of the second power line point data comprises:
mixing the first power line point data into the original point cloud data, and selecting points in a third preset range of the first power line point data;
and summarizing the points in the third preset range to obtain second power line point data.
5. The method of claim 1, wherein obtaining the target point cloud data, identifying the point cloud data using a surface curvature estimation algorithm and a local dimension multi-scale algorithm, and prior to obtaining the first power line point data comprises:
acquiring original point cloud data, and performing rarefaction processing on the original point cloud data to obtain target point cloud data.
6. A power line point identification apparatus, characterized in that the apparatus comprises:
the first power line point data acquisition module is used for acquiring target point cloud data, identifying the target point cloud data by adopting a curved surface curvature estimation algorithm and a local dimension multi-scale algorithm and acquiring first power line point data; identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data; identifying residual point cloud data by adopting a local dimension multi-scale algorithm to obtain fifth power line point data, and obtaining the first power line point data according to the fourth power line point data and the fifth power line point data; the identifying the target point cloud data by adopting a curved surface curvature estimation algorithm to obtain fourth power line point data and residual point cloud data comprises the following steps: selecting a reference point in the target point cloud data, and calculating point data of the reference point in the target point cloud data in a first preset range; performing tangent plane fitting on the point data in the first preset range to obtain a plane equation; performing surface fitting on the point data in the first preset range according to the plane equation to obtain a quadric surface equation; determining the fourth power line point data according to the quadric surface equation; the identifying the remaining point cloud data by using the local dimension multi-scale algorithm to obtain fifth power line point data comprises: acquiring the residual point cloud data, and calculating the corresponding characteristic values of the points in the residual point cloud data in a second preset range; constructing local dimension multi-scale features according to the feature values; determining the fifth power line point data according to the local dimension multi-scale feature;
the second power line point data acquisition module is used for acquiring original point cloud data, and processing the first power line point data and the original point cloud data by adopting a nearest neighbor search method to acquire second power line point data;
and the third power line point data acquisition module is used for summarizing the first power line point data and the second power line point data to obtain third power line point data.
7. A power line point identification device, characterized in that it comprises at least the apparatus of claim 6.
8. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program performs the steps of the method according to any of claims 1 to 5.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811637018.7A CN111382767B (en) | 2018-12-29 | 2018-12-29 | Power line point identification method, device, equipment, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811637018.7A CN111382767B (en) | 2018-12-29 | 2018-12-29 | Power line point identification method, device, equipment, computer equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111382767A CN111382767A (en) | 2020-07-07 |
| CN111382767B true CN111382767B (en) | 2023-04-18 |
Family
ID=71220890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811637018.7A Active CN111382767B (en) | 2018-12-29 | 2018-12-29 | Power line point identification method, device, equipment, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111382767B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112801200B (en) * | 2021-02-07 | 2024-02-20 | 文远鄂行(湖北)出行科技有限公司 | Data packet screening method, device, equipment and storage medium |
| CN113609632B (en) * | 2021-10-08 | 2021-12-21 | 天津云圣智能科技有限责任公司 | Method and device for determining power line compensation point and server |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104020475A (en) * | 2014-06-20 | 2014-09-03 | 西安煤航信息产业有限公司 | Power line extracting and modeling method based on airborne LiDAR data |
| CN104463872A (en) * | 2014-12-10 | 2015-03-25 | 武汉大学 | Classification method based on vehicle-mounted LiDAR point cloud data |
| CN106709946A (en) * | 2016-12-16 | 2017-05-24 | 武汉大学 | Multiple-divided-conductor automatic extraction and fine modeling method based on LiDAR point clouds |
| CN107292935A (en) * | 2017-05-05 | 2017-10-24 | 深圳市建设综合勘察设计院有限公司 | Compression method, storage device and the laser radar of airborne high-density laser point cloud |
-
2018
- 2018-12-29 CN CN201811637018.7A patent/CN111382767B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104020475A (en) * | 2014-06-20 | 2014-09-03 | 西安煤航信息产业有限公司 | Power line extracting and modeling method based on airborne LiDAR data |
| CN104463872A (en) * | 2014-12-10 | 2015-03-25 | 武汉大学 | Classification method based on vehicle-mounted LiDAR point cloud data |
| CN106709946A (en) * | 2016-12-16 | 2017-05-24 | 武汉大学 | Multiple-divided-conductor automatic extraction and fine modeling method based on LiDAR point clouds |
| CN107292935A (en) * | 2017-05-05 | 2017-10-24 | 深圳市建设综合勘察设计院有限公司 | Compression method, storage device and the laser radar of airborne high-density laser point cloud |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111382767A (en) | 2020-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111523414A (en) | Face recognition method and device, computer equipment and storage medium | |
| CN107292935B (en) | Compression method of airborne high-density laser point cloud, storage device and laser radar | |
| US20180081035A1 (en) | Method and device for filtering point cloud data | |
| CN109711416B (en) | Target identification method and device, computer equipment and storage medium | |
| CN106886980B (en) | Point cloud density enhancement method based on three-dimensional laser radar target identification | |
| CN109215129B (en) | Local feature description method based on three-dimensional point cloud | |
| CN110544298B (en) | Substation modeling method, device, computer equipment and storage medium | |
| El‐Sayed et al. | Plane detection in 3D point cloud using octree‐balanced density down‐sampling and iterative adaptive plane extraction | |
| CN111260668B (en) | Power line extraction method, system and terminal | |
| CN111382767B (en) | Power line point identification method, device, equipment, computer equipment and storage medium | |
| CN111179274A (en) | Map ground segmentation method, map ground segmentation device, computer equipment and storage medium | |
| CN113537180B (en) | Tree obstacle identification method and device, computer equipment and storage medium | |
| CN116543091B (en) | Visualization method, system, computer equipment and storage medium for power transmission line | |
| CN113538264B (en) | Denoising method and device for point cloud data and storage medium | |
| WO2022133770A1 (en) | Method for generating point cloud normal vector, apparatus, computer device, and storage medium | |
| CN112241952A (en) | Method and device for recognizing brain central line, computer equipment and storage medium | |
| CN116310849B (en) | Tree point cloud monomerization extraction method based on three-dimensional morphological characteristics | |
| CN115457492A (en) | Target detection method and device, computer equipment and storage medium | |
| CN110544308A (en) | Transformer substation modeling method and device, computer equipment and storage medium | |
| CN111458691B (en) | Building information extraction method and device and computer equipment | |
| CN113436223B (en) | Point cloud data segmentation method and device, computer equipment and storage medium | |
| CN112393735A (en) | Positioning method and device, storage medium and electronic device | |
| CN113192174A (en) | Mapping method and device and computer storage medium | |
| CN112232248A (en) | Method and device for extracting plane features of multi-line LiDAR point cloud data | |
| Li et al. | Nonlinear von Mises–Fisher filtering based on isotropic deterministic sampling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 6/F, Building C, State Grid Office Area, Future Science and Technology City, Beiqijia Town, Changping District, Beijing 102213 Applicant after: State Grid Power Space Technology Co.,Ltd. Address before: 6/F, Building C, State Grid Office Area, Future Science and Technology City, Beiqijia Town, Changping District, Beijing 102213 Applicant before: SGCC GENERAL AVIATION Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |