CN112991303A - Automatic extraction method of electric tower insulator string based on three-dimensional point cloud - Google Patents

Automatic extraction method of electric tower insulator string based on three-dimensional point cloud Download PDF

Info

Publication number
CN112991303A
CN112991303A CN202110300702.1A CN202110300702A CN112991303A CN 112991303 A CN112991303 A CN 112991303A CN 202110300702 A CN202110300702 A CN 202110300702A CN 112991303 A CN112991303 A CN 112991303A
Authority
CN
China
Prior art keywords
point cloud
point
layer
dimensional
tower
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.)
Granted
Application number
CN202110300702.1A
Other languages
Chinese (zh)
Other versions
CN112991303B (en
Inventor
黄玉春
欧阳凝晖
彭祖铭
王浩宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University WHU
Original Assignee
Wuhan University WHU
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuhan University WHU filed Critical Wuhan University WHU
Priority to CN202110300702.1A priority Critical patent/CN112991303B/en
Publication of CN112991303A publication Critical patent/CN112991303A/en
Application granted granted Critical
Publication of CN112991303B publication Critical patent/CN112991303B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
    • G06T3/06—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20132—Image cropping
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30184—Infrastructure
    • 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

Abstract

The invention discloses an automatic extraction method of an electric tower insulator string based on three-dimensional point cloud. Firstly, preprocessing an original three-dimensional point cloud according to a roughly set electric tower center coordinate, wherein the point cloud is cut so as to reduce the data volume and reduce the interference of irrelevant point clouds. And calculating a rotation deviation angle kappa relative to a coordinate axis XY of the preprocessed point cloud, and then rotating the point cloud according to the kappa angle to enable the Y coordinate axis to be consistent with the direction of the tower of the electric tower. And layering the rotated point clouds according to a certain height, judging the type of each layer of wire insulator string, and then processing the vertical strings of the top layer, the non-top layer and the horizontal strings of the non-top layer by adopting different methods to respectively obtain top layer suspension points, vertical insulator strings and horizontal insulator strings. In each link of point cloud processing, the three-dimensional problem is converted into the problem of two-dimensional image processing by a projection method, so that the operation efficiency is improved, and the method has higher robustness and accuracy for various different types of electric towers.

Description

Automatic extraction method of electric tower insulator string based on three-dimensional point cloud
Technical Field
The invention belongs to the field of power line inspection, and particularly relates to an automatic extraction method of an electric tower insulator string based on three-dimensional point cloud.
Background
Along with the wide use of unmanned aerial vehicle in the power line is patrolled and examined, changed outdoor electric power in the past and patrolled and examined and rely on manpower walking, climbing, the problem that the manpower occupies a lot of and overall efficiency is not high. The automatic inspection at the present stage mainly adopts a flight path flight mode, assists in closing a loop in the process and ensures the inspection precision. And unmanned aerial vehicle's automation is patrolled and examined, need according to the spatial position information of power line part, plans corresponding flight path to shoot to the part that the automatic alignment needs to be patrolled and examined.
The electric tower insulator string is an important content of power line inspection, and the three-dimensional coordinate information of the insulator string is an important basis for unmanned aerial vehicle inspection line planning. Because the number of the electric towers in the electric transmission line is large, the distribution is wide, and the terrain is complex, an automatic method needs to be adopted to extract the insulator strings in the electric towers, so that the related three-dimensional information can be obtained.
Disclosure of Invention
Aiming at the existing background requirements, the invention provides an automatic extraction method of an insulator string of an electric tower based on three-dimensional point cloud, so that the technical problems that the number of electric towers is large, the distribution is wide, the terrain is complex, and the insulator string is difficult to extract are solved.
In order to achieve the purpose, according to one aspect of the invention, the invention provides an automatic extraction method of an insulator string of an electric tower based on three-dimensional point cloud, and the greatest innovation point of the method is that a projection method is largely used, three-dimensional point cloud processing is converted into two-dimensional image processing content, the problem of insulator string extraction is simplified, and the operation efficiency is greatly improved. The method comprises the following steps:
step 1: the method comprises the steps of obtaining three-dimensional point cloud data of an electric tower and the center of the electric tower, further cutting the three-dimensional point cloud data of the electric tower in the horizontal XY direction and the vertical Z direction to obtain a cut point cloud;
preferably, the three-dimensional point cloud data of the electric tower in the step 1 is as follows:
data0(x,y,z)＝{data0 k(xk,yk,zk),k∈[1,K]}
wherein, the data0(x, y, z) is the electric tower three-dimensional point cloud data, data0 k(xk,yk,zk) Is the pixel, x, of the kth point in the three-dimensional point cloud data of the electric towerkIs the X-direction coordinate, y of the kth point in the three-dimensional point cloud data of the electric towerkIs the Y-direction coordinate, z of the kth point in the three-dimensional point cloud data of the electric towerkThe Z-direction coordinate of the kth point in the electric tower three-dimensional point cloud data is shown, and K is the number of three-dimensional points in the electric tower three-dimensional point cloud data;
the horizontal XY direction cutting in the step 1 comprises the following steps:
according to the electric tower center, i.e. center0(x0,y0,z0)；
Three-dimensional point cloud data of electric tower0Search for abs (x) in (x, y, z)k-x0)＜threshold0And abs (y)k-y0)＜threshold0The point cloud after horizontal clipping is constructed into data1(x,y,z)，threshold0Is a threshold parameter;
step 1, the vertical Z direction cutting is as follows:
combining the point cloud (data) cut in the horizontal direction1(x, y, z), search for abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The minimum value of Z-direction coordinates in the point cloud in the range is defined as Zmin(ii) a Search for satisfied abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The maximum value of Z-direction coordinates in the point cloud in the range is defined as Zmax，threshold1Is a threshold parameter;
defining the interlayer spacing as z _ interval 1.5;
the number of layers is calculated as:
z_num＝ceil[(zmax-zmin)/z_interval]
wherein ceil represents an rounding-up function, and Z _ num represents the number of layers;
defining the point cloud density of each layer as:
z_density＝{z_densityl,l∈[1,z_num]}
and initialize z _ diversityl0, and layer 1 is the highest layer;
traversal data1(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval]Calculating the z _ now of the layer to which the kth point belongs, and counting the number of the points of each layer;
from layer 1, find that z _ dense is satisfieda+s＞10，s＝[1,6]The layer a of (a) calculates a retention value Z _ up in the Z direction from Z _ up ═ Z _ max- (a-1) × Z _ interval;
point cloud data1(x, y, z) satisfies zk≥z_min+threshold2And z iskKeeping points less than or equal to z _ up, updating the highest point z _ max to z _ up of the point cloud, and updating the lowest point z _ min to z _ min + threshold2，threshold2Is a threshold parameter.
Finally, the point cloud data after cutting is obtained2(x,y,z)
Step 2: for the clipped point cloud data2(x, y, z) acquiring certain range of point cloud data at the topmost layer3(x, y, z), calculating the spin-yaw angle kappa, and then point cloud data2Rotating (x, Y, z) to make the direction of the tower of the electric tower consistent with the direction of the coordinate axis Y, and obtaining point cloud data4(x,y,z)；
Preferably, step 2 comprises:
step 2.1: from the point cloud data2Starting from the highest point (x, y, z) downwards, setting a layering interval z _ interval1The number of layers z _ num is calculated as 0.11＝ceil[(zmax-zmin)/z_interval1]Defining the density z _ density of each layer point cloud1＝{z_density1 k,k∈[1,z_num1]And initialize z _ dense1 k0, and the 1 st layer is the highest layer. Traversal point cloud data2(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval1]And calculating the layer to which the k-th point belongs, and counting the number of points in each layer. From layer 1, find that z _ d is satisfiedensity1 kLayer k more than 100, calculating the maximum value Z of the top point cloud in the Z directiontopMax, point-to-point cloud data2(x, y, z) Retention satisfies zk≥ztopMax-3 and zk≤ztopTaking the point of max as the top-level point cloud data3(x,y,z)。
Step 2.2: data of top point cloud3(x, y, z) downwards projecting to obtain XY projection graph (the projections mentioned in the invention are all binary projections, the specific way of generating XY projection at the position is as follows, and the other situations are similar3Setting the two-dimensional grid interval to grid _ interval 0.1 according to the maximum value and the minimum value of XY coordinates of (x, y, z), and according to row ceil [ (x _ max-x _ min)/grid _ interval]Calculating the number of XY projection drawing lines, col ═ ceil [ (y _ max-y _ min)/grid _ interval]Calculating the number of XY projection diagram rows and traversing the point cloud data3(x, y, z) according to row _ now ═ ceil [ (x _ max-x) xk)/grid_interval]Calculating the line number of the projection drawing where the current point falls according to col _ now ═ ceil [ (y _ max-y)k)/grid_interval]Calculating the column number of the current point falling on the projection drawing, if a pixel has a point cloud falling in the range, the pixel value is 255, otherwise, the pixel value is 0), performing morphological filling on the XY projection, performing corrosion operation, and extracting the maximum outline. The yaw angle kappa is calculated from the first principal direction of the maximum profile using the PCA method.
Step 2.3: according to the calculated rotating deflection angle kappa, the point cloud data after cutting is processed2(x, y, z) at the center of the electric tower0(x0,y0,z0) The rotation center is the Z axis of the coordinate system, the rotation angle is calculated according to the deflection angle kappa, the rotation is carried out, the direction of the tower is consistent with the direction of the Y axis of the coordinate system, and the point cloud data after rotation is obtained4(x,y,z)。
And step 3: rotating point cloud data4(x, y, Z) and carrying out projection to obtain a ZX projection diagram, layering according to the wire height to obtain the approximate position Z _ layer of Z where the wires on the two sides of each layer of the tower intersectkK is the number of the layers obtained by layering;
preferably, step 3 comprises:
step 3.1: the projection results in a ZX projection. And finding out the column of the center of the tower and the column of the left and right boundaries of the tower according to the condition that the number of the points of the column of the electric tower is greater than that of the points of the column of the electric wire, and setting the pixel value to be 0 to remove the electric tower.
Step 3.2: and taking the lowest point of the ZX image as a seed point, growing to obtain the interference of residual tree barriers on the ground and the like, and setting the pixel value of the part of the point as 0, so that the rest part of the image basically only comprises the electric wire.
Step 3.3: and (3) performing edge extraction on the ZX projection image, then performing Hough line detection on the edge image, and screening according to the vertical distance of the slope of the line, thereby obtaining the position of each layer of electric wire in the row of the tower center column.
And 4, step 4: cutting to obtain point clouds of each layer based on each layer of position Z obtained by layering, wherein the uppermost layer is used as a top layer, and the types of insulator strings in the non-top layer are judged to be vertical string layers and horizontal string layers;
preferably, step 4 comprises:
step 4.1: the top layer is regarded as the top layer and does not participate in type judgment. And for the non-top layer, respectively taking small-range point clouds above and below the calculated Z position, and projecting downwards to obtain an XY projection diagram.
Step 4.2: and counting the continuous conditions of the pixel points in a certain row range from top to bottom according to the row of the center of the tower, wherein if the continuous conditions indicate that the layer has the tower, the layer belongs to a horizontal string layer, and otherwise, the layer belongs to a vertical string layer.
And 5: processing the point clouds on each layer of the electric tower by adopting different methods based on the types of the point clouds, and finally obtaining suspension points on the top layer, obtaining corresponding vertical insulator strings on the vertical string layer and obtaining corresponding horizontal insulator strings on the horizontal layer;
preferably, step 5 comprises:
step 5.1: the uppermost layer of the electric tower layering result is regarded as the top layer, and the suspension point is extracted by processing
Step 5.2: for the part of the electric tower with the non-top layer and the type judged as the vertical string layer, the horizontal insulator string is extracted by processing
Step 5.3: for the part of the electric tower with the non-top layer and the type judged as the transverse string layer, the vertical insulator string is extracted by processing
Preferably, step 5.1 comprises:
step 5.1.1: and (4) projecting the top-layer point cloud to obtain a ZX projection diagram, counting the number of each row of points, finding out an area boundary row with the maximum density, and calculating the X coordinate of the hanging point according to the mean value of the boundary row.
Step 5.1.2: starting from the boundary columns found in the step 5.1.1 to two sides respectively, until the columns with small enough width are found, and calculating the Z coordinate of the hanging point according to the initial row of the columns.
Step 5.1.3: and (4) projecting the point cloud to obtain an XY projection image. And moving the wire to the row where the hanging point is located from the upper end and the lower end, thereby calculating the Y coordinate of the hanging point.
Step 5.1.4: and (3) rotating the calculated three-dimensional coordinates of the hanging points by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the hanging points in the original coordinate system.
Preferably, step 5.2 comprises:
step 5.2.1: and (4) performing downward projection on the point cloud to obtain an XY projection diagram, counting the number of the electric wires according to the interval existing between different electric wires on the same layer, and obtaining the range of each electric wire Y.
Step 5.2.2: and (3) obtaining a ZX projection image by the point cloud projection of each wire, filling, refining and deburring the ZX projection, searching a three-fork point for the refined image, wherein the point is the intersection point of the insulator string and the wire, and calculating the ZX coordinate of the intersection point.
Step 5.2.3: and taking a certain range of point cloud by using the ray corresponding to the intersection point ZX, and taking the average value of the cloud point Y of the point cloud as the Y coordinate of the intersection point, thereby obtaining the three-dimensional coordinate of the intersection point.
Step 5.2.4: and (3) upwards taking a certain range of point cloud according to the three-dimensional coordinates of the intersection point to obtain ZY projection, finding a row with a certain width, namely the position of a suspension point Z from the row where the intersection point is located, and keeping the XY coordinates of the suspension point consistent with the XY coordinates of the intersection point, thereby obtaining the three-dimensional coordinates of the suspension point.
Step 5.2.5: and (3) rotating by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical string in the original coordinate system according to the three-dimensional coordinates of the vertical string intersection point and the suspension point calculated by each wire.
Preferably, step 5.3 comprises:
step 5.3.1: and (4) downwards projecting the point cloud to obtain an XY projection image, and cutting and removing the part outside a certain range from top to bottom of the central line of the tower in the XY projection.
Step 5.3.2: analyzing the connected domain of the rest part of the cutting scissors, solving the angular points at the two sides of the two connected domains, namely 4 angular points at the inner side of the H-shaped pattern, and slightly expanding the coordinates of the angular points to obtain XY coordinates of the hanging points.
Step 5.3.3: and (3) cutting according to the XY coordinates of the suspension points and the position of the center X of the tower to obtain a part of point cloud which only comprises the tower and does not comprise the electric wire and the insulator string, projecting to obtain a ZY projection diagram, and finding out lines with jumping of 255 points as the Z coordinates of the suspension points of the transverse string.
Step 5.3.4: the layer of point cloud projection obtains a ZX projection diagram, and the highest row of the electric wire, the lowest row of the electric wire and the highest row of the suspended object are traversed and recorded row by row according to the principle that the left suspension point is found to the left and the right suspension point is found to the right; and (4) taking the difference between the wire highest behavior reference condition, the wire lowest row and the suspension object highest row as a judgment condition, and carrying out classification discussion and fitting to obtain a ZX value.
Step 5.3.5: and cutting the layer of point cloud according to the ZX coordinates of the cross-string intersection points, taking a certain range of point cloud by using rays corresponding to the intersection points ZX, dividing into two clusters according to the Y coordinates of the cluster cloud, and taking the average value of each cluster of point cloud Y as the Y coordinate of each cross-string intersection point.
Step 5.3.6: and (3) rotating the three-dimensional coordinates of each transverse string intersection point and each suspension point calculated according to the layer point cloud by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical strings in the original coordinate system.
Step 6: in consideration of the problem of data defect, in order to enhance robustness, a series of post-processing is carried out on the horizontal strings, and the possibly existing horizontal insulator strings which are not fully extracted are completed.
Preferably, step 6 comprises:
step 6.1: and for step 5.3.4, the intersection point possibly cannot be found in the ZX projection, and the method is completed by adopting a post-processing method. If the layer has the horizontal string extracted, the length of the horizontal string in the three-dimensional space is calculated. And calculating the length of the insulator string in the ZX projection according to the direction vector of the insulator string. From this length the intersection ZX coordinate is directly calculated.
Step 6.2: the method is almost the same as the step 6.1, only when the horizontal string is not extracted from the same layer, if the electric tower has the extracted horizontal string, the length of the insulator string extracted from the electric tower is taken as the length of the horizontal string in the three-dimensional space, and the intersecting point ZX coordinate is completely supplemented.
Step 6.3: the method is almost the same as the step 6.1, except that when the electric tower does not extract the transverse string, a predefined length is taken as the length of the transverse string in the three-dimensional space, and the intersection point ZX coordinate is supplemented.
The method has the advantages that the three-dimensional point cloud problem is converted into the two-dimensional image processing problem through the projection method, the operation efficiency is improved, and the robustness and the accuracy are high for various types of electric towers.
Drawings
FIG. 1: is a method flow diagram provided by the embodiment of the invention;
FIG. 2: is a schematic diagram of the calculated kappa angle provided by the embodiment of the invention;
FIG. 3: is a result graph of calculating the center Y of the tower provided by the embodiment of the invention;
FIG. 4: the residual wiring diagram after eliminating the interference of the electric tower and the tree barrier provided by the embodiment of the invention;
FIG. 5: is a top layer hanging point X coordinate result chart provided by the embodiment of the invention;
FIG. 6: is a top hanging point Z coordinate result graph provided by the embodiment of the invention;
FIG. 7: is a vertical string ZX projection detailed diagram provided by the embodiment of the invention;
FIG. 8: the ZY projection diagram for calculating the vertical string Y coordinate provided by the embodiment of the invention;
FIG. 9: is a result graph of XY coordinates of the horizontal string suspension point provided by the embodiment of the invention;
FIG. 10: is a Z coordinate result graph of the horizontal string suspension point provided by the embodiment of the invention;
FIG. 11: is a ZX coordinate result chart of the cross intersection point provided by the embodiment of the invention;
FIG. 12: the invention provides a ZX projection incomplete situation diagram of a transverse cross point.
FIG. 13: the method is a graph of vertical string extraction results provided by the embodiment of the invention;
FIG. 14: the vertical string extraction result graph provided by the embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
In the present examples, "first", "second", etc. are used for distinguishing different objects, and are not necessarily used for describing a particular order or sequence.
Fig. 1 is a schematic flow chart of a method according to a first embodiment of the present invention, which specifically includes the following steps:
(1) point cloud data preprocessing: on one hand, the acquired three-dimensional point cloud has a large number of surface points, so that the method has no effect on the extraction of the insulator string and can slow down the operation efficiency; on the other hand, interference such as tree barriers exists, so that preprocessing of the input three-dimensional point cloud is very necessary. The pretreatment mainly comprises cutting in the XY direction and cutting in the Z direction.
(2) Point cloud rotation: in order to accelerate the operation efficiency, the invention uses a large amount of projection methods to project the three-dimensional point cloud to the two-dimensional plane. In order to make projection fast and effective, the point cloud needs to be rotated so that the direction of the electric tower is more standard, and here, the direction of the rotated electric tower is consistent with the direction of the Y coordinate axis.
(3) Electric tower layering: because the number, the types and the distribution of the insulator strings are different from one another in different electric towers, the whole electric tower is projected, and all point clouds are mixed together and cannot be effectively distinguished. Therefore, the layering of the electric tower can reduce the number of point clouds processed each time, and can also obtain information such as the number of layers of the electric tower and the like so as to obtain the number and type information of the insulator strings, which is beneficial to the subsequent extraction of the insulator strings.
(4) And (3) judging the type layer by layer: due to different insulator strings, the distribution forms of the insulator strings are different in space, and the surrounding point cloud characteristics are different. It is difficult to extract insulator strings of various types by one method, and it is much easier to extract insulator strings of different types by different methods. Therefore, after the electric tower is layered, the type of each layer of insulator string can be judged.
(5) Extracting the insulator strings layer by layer: and processing the point clouds on each layer of the electric tower and the categories of the point clouds by adopting different methods, obtaining suspension points on the top layer, obtaining corresponding vertical insulator strings on the vertical string layer, and obtaining corresponding horizontal insulator strings on the horizontal layer.
(6) And (3) transverse skewer extraction post-treatment: considering the problem of data defect, the horizontal string feature may not be obvious enough, and the method for extracting the horizontal string in step (5) may fail. In order to enhance robustness, a series of post-treatments are carried out on the transverse strings to complete the possibly existing horizontal insulator strings which are not fully extracted;
in the embodiment of the present invention, step (1) may be implemented by:
(1.1) horizontal cutting:
according to the electric tower center coordinate (x)c,yc,zc) Will satisfy abs (x)i-xc)＜threx，abs(yi-yc)＜threyThe point of (2) is retained.
(1.2) cutting in the vertical direction:
and acquiring a Z value of the lowest point in the XY small range of the electric tower center, adding a certain height to the Z value to serve as a cutting threshold, and removing the point cloud below the height. And certain intervals are reserved between other electric transmission line wires and the electric tower, so that the electric transmission line wires are layered at certain intervals from the highest point downwards, the positions with certain quantity of point clouds on certain continuous layers are found, the point clouds higher than the positions are removed, and the interference from other electric transmission line wires is eliminated.
In the embodiment of the present invention, step (2) may be implemented by:
(2.1) acquiring top-level point cloud:
the kappa angle calculated by the method is the direction of the first principal component calculated by a PCA method through XY projection of a tower top view of the electric tower. The reason for taking the top point cloud is that the top wire is generally thin, and the wire is easily separated from the tower. In addition, because the shapes of the electric towers are different, the method for directly taking the point clouds in a certain range from the highest point is not robust enough, so that the method for obtaining the top layer point clouds comprises the steps of starting from the highest point downwards, layering at certain intervals, calculating the number of the point clouds in each layer, and taking the point clouds in a certain range below the Z where the layers with enough number are located as the top layer point clouds.
(2.2) calculating the kappa angle:
and (4) projecting the top point cloud to obtain an XY projection diagram, filling and corroding, extracting the maximum outline, and calculating an angle between the top point cloud and the Y axis to serve as a rotation deviation angle kappa. As shown in fig. 2, the XY projection plan view of the top layer of the electric tower, the filled image, and the etched image are shown.
(2.3) rotating the point cloud:
and rotating the cut point cloud according to the calculated deflection angle, wherein the selected center is the approximate center of the tower, and the rotating shaft is the Z axis, so that the direction of the tower is consistent with the direction of the Y axis of the coordinate system.
In the embodiment of the present invention, step (3) may be implemented by:
(3.1) digging out the electric tower
Since the initial tower center may not be accurate enough, the tower center column and the columns at the left and right tower boundaries are found and the pixel value is set to 0 to remove the tower, according to the number of points in the column where the tower is located is greater than the number of points in the column where the electric wire is located. The column of the electric tower center calculated on the ZX projection is shown in fig. 3.
(3.2) removing tree barriers:
after the electric tower is dug, the electric wire is basically separated from the tree barrier, the lowest point of the ZX image is taken as a seed point, the interference of residual tree barrier on the ground and the like is obtained by growing, the pixel value of the part of the point is set to be 0, and therefore the rest part of the image basically only contains the electric wire. Fig. 4 is a ZX projection of a power tower removed and a tree barrier removed.
(3.3) electric tower layering:
and (3) performing edge extraction on the ZX projection image, then performing Hough line detection on the edge image, and screening according to the vertical distance of the slope of the line, thereby obtaining the position of each layer of electric wire in the row of the tower center column.
In the embodiment of the present invention, step (4) may be implemented by:
(4.1) generating XY projections layer by layer:
and the uppermost layer of the electric tower layering result is regarded as the top layer and does not participate in type judgment. And for the non-top layer, respectively taking small-range point clouds above and below the calculated Z position, and projecting downwards to obtain an XY projection diagram.
(4.2) judging the type layer by layer:
since the column of the center of the tower in the ZX projection is calculated in step (3.1), it corresponds to the row of the center of the tower in the known XY projection. And counting the continuous conditions of the pixel points in a certain row range from top to bottom according to the row of the center of the tower, wherein if the continuous conditions indicate that the layer has the tower, the layer belongs to a horizontal string layer, and otherwise, the layer belongs to a vertical string layer.
In the embodiment of the present invention, step (5) may be implemented by:
(5.1) top layer treatment:
the uppermost layer of the electric tower layering result is regarded as the top layer, and the suspension point is extracted by processing
(5.2) vertical series layer treatment:
for the part of the electric tower with the non-top layer and the type judged as the vertical string layer, the horizontal insulator string is extracted by processing
(5.3) transverse tandem layer treatment:
for the part of the electric tower with the non-top layer and the type judged as the transverse string layer, the vertical insulator string is extracted by processing
In the embodiment of the present invention, step (5.1) may be implemented by:
(5.1.1) calculating the X coordinate of the suspension point:
and (4) projecting the top-layer point cloud to obtain a ZX projection diagram, counting the number of each row of points, finding out an area boundary row with the maximum density, and calculating the X coordinate of the hanging point according to the mean value of the boundary row. Fig. 5 shows the calculation result of the top X coordinate.
(5.1.2) calculating the Z coordinate of the suspension point:
starting from the boundary column found in the step (5.1.1) to two sides respectively, until a column with a small enough width is found, and calculating the Z coordinate of the hanging point according to the initial row of the column. Fig. 6 shows the calculation result of the top X coordinate.
(5.1.3) calculating the Y coordinate of the suspension point:
and (4) projecting the point cloud to obtain an XY projection image. And moving the wire to the row where the hanging point is located from the upper end and the lower end, thereby calculating the Y coordinate of the hanging point.
(5.1.4) calculating the original three-dimensional coordinates of the suspension point:
and (3) rotating the calculated three-dimensional coordinates of the hanging points by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the hanging points in the original coordinate system.
In the embodiment of the present invention, step (5.2) may be implemented by:
(5.2.1) counting the number and the range of the wires in each layer:
firstly, the point cloud of the layer is projected downwards to obtain an XY projection diagram, the number of the electric wires is counted according to the interval existing between different electric wires of the same layer, and the range of each electric wire row, namely the range of Y is obtained.
(5.2.2) calculating the ZX coordinates of the vertical cross points:
and (3) obtaining a ZX projection image by the point cloud projection of each wire, filling, refining and deburring the ZX projection, searching a three-fork point for the refined image, wherein the point is the intersection point of the insulator string and the wire, and calculating the ZX coordinate of the intersection point. Fig. 7 is a vertical cross ZX diagram.
(5.2.3) calculating the Y coordinate of the vertical cross point:
and taking a certain range of point cloud by using the ray corresponding to the intersection point ZX, and taking the average value of the cloud point Y of the point cloud as the Y coordinate of the intersection point, thereby obtaining the three-dimensional coordinate of the intersection point.
(5.2.4) calculating the coordinates of the suspension points of the vertical strings:
and (3) upwards taking a certain range of point cloud according to the three-dimensional coordinates of the intersection point to obtain ZY projection, finding a row with a certain width, namely the position of a suspension point Z from the row where the intersection point is located, and keeping the XY coordinates of the suspension point consistent with the XY coordinates of the intersection point, thereby obtaining the three-dimensional coordinates of the suspension point. FIG. 8 is a Z diagram of a vertical string of hanging points
(5.2.5) calculating the original three-dimensional coordinates of the vertical string:
and (3) according to the three-dimensional coordinates of the intersection point and the suspension point of the vertical string calculated by each wire, rotating by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical string in the original coordinate system.
In the embodiment of the present invention, step (5.3) may be implemented by:
(5.3.1) cutting the center part of the XY projection tower:
and (4) downwards projecting the point cloud to obtain an XY projection image, and cutting and removing the part outside a certain range from top to bottom of the central line of the tower in the XY projection.
(5.3.2) calculating the XY coordinates of the suspension point:
analyzing the connected domain of the rest part of the cutting scissors, solving the angular points at two sides of the two connected domains (namely 4 angular points at the inner side of the H-shaped pattern), and slightly expanding the coordinates of the angular points to obtain XY coordinates of the hanging points. Fig. 9 is an XY diagram of the horizontal series of suspension points.
(5.3.3) calculating the Z coordinate of the suspension point:
and (3) cutting according to the XY coordinates of the suspension points and the position of the center X of the tower to obtain a part of point cloud which only comprises the tower and does not comprise the electric wire and the insulator string, projecting to obtain a ZY projection diagram, and finding out lines with jumping of 255 points as the Z coordinates of the suspension points of the transverse string. FIG. 10 is a cross-string hanging point Z diagram
(5.3.4) calculate the intersection ZX coordinate:
the layer of point cloud projection obtains a ZX projection diagram, and the highest row of the electric wire, the lowest row of the electric wire and the highest row of the suspended object are traversed and recorded row by row according to the principle that the left suspension point is found to the left and the right suspension point is found to the right; and (4) taking the difference between the wire highest behavior reference condition, the wire lowest row and the suspension object highest row as a judgment condition, and carrying out classification discussion and fitting to obtain a ZX value. FIG. 11 is a diagram showing a horizontal intersection ZX.
(5.3.5) calculating the intersection Y coordinate:
and cutting the layer of point cloud according to the ZX coordinates of the cross-string intersection points, taking a certain range of point cloud by using rays corresponding to the intersection points ZX, dividing into two clusters according to the Y coordinates of the cluster cloud, and taking the average value of each cluster of point cloud Y as the Y coordinate of each cross-string intersection point.
(5.3.6) calculating the original three-dimensional coordinates of the horizontal string:
and (3) rotating the three-dimensional coordinates of each horizontal string intersection point and each suspension point calculated according to the layer point cloud by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical strings in the original coordinate system.
In the embodiment of the present invention, step (6) may be implemented by:
(6.1) horizontal-series post-treatment 1:
and (5.3.4) if the intersection point can not be found in the ZX projection, completing the process by adopting a post-processing method. If the layer has the horizontal string extracted, the length of the horizontal string in the three-dimensional space is calculated. And calculating the length of the insulator string in the ZX projection according to the direction vector of the insulator string. From this length the intersection ZX coordinate is directly calculated.
(6.2) horizontal-series post-treatment 2:
and (4) the processing method is almost the same as the processing method in the step (6.1), and if the electric tower has the extracted transverse string when the transverse string is not extracted in the same layer, the length of the insulator string extracted by the electric tower is taken as the length of the transverse string in the three-dimensional space, and the intersection point ZX coordinate is completely supplemented.
(6.3) horizontal string post-treatment 3:
the method is basically the same as the processing method in the step (6.1), except that when the electric tower does not extract the transverse string, a predefined length is taken as the length of the transverse string in the three-dimensional space, and the intersection point ZX coordinate is supplemented.
The second embodiment of the invention is a three-dimensional point cloud-based automatic extraction method for an electric tower insulator string, which comprises the following specific steps:
step 1: the method comprises the steps of obtaining three-dimensional point cloud data of an electric tower and the center of the electric tower, further cutting the three-dimensional point cloud data of the electric tower in the horizontal XY direction and the vertical Z direction to obtain a cut point cloud;
step 1, the three-dimensional point cloud data of the electric tower are as follows:
data0(x,y,z)＝{data0 k(xk,yk,zk),k∈[1,K]}
wherein, the data0(x, y, z) is the electric tower three-dimensional point cloud data, data0 k(xk,yk,zk) Is the pixel, x, of the kth point in the three-dimensional point cloud data of the electric towerkIs the X-direction coordinate, y of the kth point in the three-dimensional point cloud data of the electric towerkIs the Y-direction coordinate, z of the kth point in the three-dimensional point cloud data of the electric towerkThe Z-direction coordinate of the kth point in the electric tower three-dimensional point cloud data is shown, and K is the number of three-dimensional points in the electric tower three-dimensional point cloud data;
the horizontal XY direction cutting in the step 1 comprises the following steps:
according to the electric tower center, i.e. center0(x0,y0,z0)；
Three-dimensional point cloud data of electric tower0Search for abs (x) in (x, y, z)k-x0)＜threshold0And abs (y)k-y0)＜threshold0The point cloud after horizontal clipping is constructed into data1(x,y,z)，threshold0Is a threshold parameter;
step 1, the vertical Z direction cutting is as follows:
combining the point cloud (data) cut in the horizontal direction1(x, y, z), search for abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The minimum value of Z-direction coordinates in the point cloud in the range is defined as Zmin(ii) a Search for satisfied abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The maximum value of Z-direction coordinates in the point cloud in the range is defined as Zmax，threshold1Is a threshold parameter;
defining the interlayer spacing as z _ interval 1.5;
the number of layers is calculated as:
z_num＝ceil[(zmax-zmin)/z_interval]
wherein ceil represents an rounding-up function, and Z _ num represents the number of layers;
defining the point cloud density of each layer as:
z_density＝{z_densityl,l∈[1,z_num]}
and initialize z _ diversityl0, and layer 1 is the highest layer;
traversal data1(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval]Calculating the z _ now of the layer to which the kth point belongs, and counting the number of the points of each layer;
from layer 1, find that z _ dense is satisfieda+s＞10，s＝[1,6]The layer a of (a) calculates a retention value Z _ up in the Z direction from Z _ up ═ Z _ max- (a-1) × Z _ interval;
point cloud data1(x, y, z) satisfies zk≥z_min+threshold2And z iskKeeping points less than or equal to z _ up, updating the highest point z _ max to z _ up of the point cloud, and updating the lowest point z _ min to z _ min + threshold2，threshold2Is a threshold parameter.
Finally, the point cloud data after cutting is obtained2(x,y,z)
Step 2: for the clipped point cloud data2(x, y, z) acquiring certain range of point cloud data at the topmost layer3(x, y, z), calculating the spin-yaw angle kappa, and then point cloud data2Rotating (x, Y, z) to make the direction of the tower of the electric tower consistent with the direction of the coordinate axis Y, and obtaining point cloud data4(x,y,z)；
Preferably, step 2 comprises:
step 2.1: from the point cloud data2Starting from the highest point (x, y, z) downwards, setting a layering interval z _ interval1The number of layers z _ num is calculated as 0.11＝ceil[(zmax-zmin)/z_interval1]Defining the density z _ density of each layer point cloud1＝{z_density1 k,k∈[1,z_num1]And initialize z _ dense1 k0, and the 1 st layer is the highest layer. Traversal point cloud data2(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval1]And calculating the layer to which the k-th point belongs, and counting the number of points in each layer. From layer 1, find that z _ dense is satisfied1 kLayer k more than 100, calculating the maximum value Z of the top point cloud in the Z directiontopMax, point-to-point cloud data2(x, y, z) Retention satisfies zk≥ztopMax-3 and zk≤ztopTaking the point of max as the top-level point cloud data3(x,y,z)。
Step 2.2: data of top point cloud3(x, y, z) downwards projecting to obtain XY projection graph (the projections mentioned in the invention are all binary projections, the specific way of generating XY projection at the position is as follows, and the other situations are similar3Setting the two-dimensional grid interval to grid _ interval 0.1 according to the maximum value and the minimum value of XY coordinates of (x, y, z), and according to row ceil [ (x _ max-x _ min)/grid _ interval]Calculating the number of XY projection drawing lines, col ═ ceil [ (y _ max-y _ min)/grid _ interval]Calculating the number of XY projection diagram rows and traversing the point cloud data3(x, y, z) according to row _ now ═ ceil [ (x _ max-x) xk)/grid_interval]Calculating the line number of the projection drawing where the current point falls according to col _ now ═ ceil [ (y _ max-y)k)/grid_interval]Calculating the column number of the current point falling on the projection drawing, if a pixel has a point cloud falling in the range, the pixel value is 255, otherwise, the pixel value is 0), performing morphological filling on the XY projection, performing corrosion operation, and extracting the maximum outline. The yaw angle kappa is calculated from the first principal direction of the maximum profile using the PCA method.
Step 2.3: according to the calculated rotating deflection angle kappa, the point cloud data after cutting is processed2(x, y, z) at the center of the electric tower0(x0,y0,z0) The rotation center is the Z axis of the coordinate system, the rotation angle is calculated according to the deflection angle kappa, the rotation is carried out, the direction of the tower is consistent with the direction of the Y axis of the coordinate system, and the point cloud data after rotation is obtained4(x,y,z)。
And step 3: rotating point cloud data4(x, y, Z) and carrying out projection to obtain a ZX projection diagram, layering according to the wire height to obtain the approximate position Z _ layer of Z where the wires on the two sides of each layer of the tower intersectkK is the number of the layers obtained by layering;
preferably, step 3 comprises:
step 3.1: the projection results in a ZX projection. And finding out the column of the center of the tower and the column of the left and right boundaries of the tower according to the condition that the number of the points of the column of the electric tower is greater than that of the points of the column of the electric wire, and setting the pixel value to be 0 to remove the electric tower.
Step 3.2: and taking the lowest point of the ZX image as a seed point, growing to obtain the interference of residual tree barriers on the ground and the like, and setting the pixel value of the part of the point as 0, so that the rest part of the image basically only comprises the electric wire.
Step 3.3: and (3) performing edge extraction on the ZX projection image, then performing Hough line detection on the edge image, and screening according to the vertical distance of the slope of the line, thereby obtaining the position of each layer of electric wire in the row of the tower center column.
And 4, step 4: cutting to obtain point clouds of each layer based on each layer of position Z obtained by layering, wherein the uppermost layer is used as a top layer, and the types of insulator strings in the non-top layer are judged to be vertical string layers and horizontal string layers;
preferably, step 4 comprises:
step 4.1: the top layer is regarded as the top layer and does not participate in type judgment. And for the non-top layer, respectively taking small-range point clouds above and below the calculated Z position, and projecting downwards to obtain an XY projection diagram.
Step 4.2: and counting the continuous conditions of the pixel points in a certain row range from top to bottom according to the row of the center of the tower, wherein if the continuous conditions indicate that the layer has the tower, the layer belongs to a horizontal string layer, and otherwise, the layer belongs to a vertical string layer.
And 5: processing the point clouds on each layer of the electric tower by adopting different methods based on the types of the point clouds, and finally obtaining suspension points on the top layer, obtaining corresponding vertical insulator strings on the vertical string layer and obtaining corresponding horizontal insulator strings on the horizontal layer;
preferably, step 5 comprises:
step 5.1: the uppermost layer of the electric tower layering result is regarded as the top layer, and the suspension point is extracted by processing
Step 5.2: for the part of the electric tower with the non-top layer and the type judged as the vertical string layer, the horizontal insulator string is extracted by processing
Step 5.3: for the part of the electric tower with the non-top layer and the type judged as the transverse string layer, the vertical insulator string is extracted by processing
Preferably, step 5.1 comprises:
step 5.1.1: and (4) projecting the top-layer point cloud to obtain a ZX projection diagram, counting the number of each row of points, finding out an area boundary row with the maximum density, and calculating the X coordinate of the hanging point according to the mean value of the boundary row.
Step 5.1.2: starting from the boundary columns found in the step 5.1.1 to two sides respectively, until the columns with small enough width are found, and calculating the Z coordinate of the hanging point according to the initial row of the columns.
Step 5.1.3: and (4) projecting the point cloud to obtain an XY projection image. And moving the wire to the row where the hanging point is located from the upper end and the lower end, thereby calculating the Y coordinate of the hanging point.
Step 5.1.4: and (3) rotating the calculated three-dimensional coordinates of the hanging points by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the hanging points in the original coordinate system.
Preferably, step 5.2 comprises:
step 5.2.1: and (4) performing downward projection on the point cloud to obtain an XY projection diagram, counting the number of the electric wires according to the interval existing between different electric wires on the same layer, and obtaining the range of each electric wire Y.
Step 5.2.2: and (3) obtaining a ZX projection image by the point cloud projection of each wire, filling, refining and deburring the ZX projection, searching a three-fork point for the refined image, wherein the point is the intersection point of the insulator string and the wire, and calculating the ZX coordinate of the intersection point.
Step 5.2.3: and taking a certain range of point cloud by using the ray corresponding to the intersection point ZX, and taking the average value of the cloud point Y of the point cloud as the Y coordinate of the intersection point, thereby obtaining the three-dimensional coordinate of the intersection point.
Step 5.2.4: and (3) upwards taking a certain range of point cloud according to the three-dimensional coordinates of the intersection point to obtain ZY projection, finding a row with a certain width, namely the position of a suspension point Z from the row where the intersection point is located, and keeping the XY coordinates of the suspension point consistent with the XY coordinates of the intersection point, thereby obtaining the three-dimensional coordinates of the suspension point.
Step 5.2.5: and (3) rotating by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical string in the original coordinate system according to the three-dimensional coordinates of the vertical string intersection point and the suspension point calculated by each wire.
Preferably, step 5.3 comprises:
step 5.3.1: and (4) downwards projecting the point cloud to obtain an XY projection image, and cutting and removing the part outside a certain range from top to bottom of the central line of the tower in the XY projection.
Step 5.3.2: analyzing the connected domain of the rest part of the cutting scissors, solving the angular points at the two sides of the two connected domains, namely 4 angular points at the inner side of the H-shaped pattern, and slightly expanding the coordinates of the angular points to obtain XY coordinates of the hanging points.
Step 5.3.3: and (3) cutting according to the XY coordinates of the suspension points and the position of the center X of the tower to obtain a part of point cloud which only comprises the tower and does not comprise the electric wire and the insulator string, projecting to obtain a ZY projection diagram, and finding out lines with jumping of 255 points as the Z coordinates of the suspension points of the transverse string.
Step 5.3.4: the layer of point cloud projection obtains a ZX projection diagram, and the highest row of the electric wire, the lowest row of the electric wire and the highest row of the suspended object are traversed and recorded row by row according to the principle that the left suspension point is found to the left and the right suspension point is found to the right; and (4) taking the difference between the wire highest behavior reference condition, the wire lowest row and the suspension object highest row as a judgment condition, and carrying out classification discussion and fitting to obtain a ZX value.
Step 5.3.5: and cutting the layer of point cloud according to the ZX coordinates of the cross-string intersection points, taking a certain range of point cloud by using rays corresponding to the intersection points ZX, dividing into two clusters according to the Y coordinates of the cluster cloud, and taking the average value of each cluster of point cloud Y as the Y coordinate of each cross-string intersection point.
Step 5.3.6: and (3) rotating the three-dimensional coordinates of each transverse string intersection point and each suspension point calculated according to the layer point cloud by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical strings in the original coordinate system.
Step 6: in consideration of the problem of data defect, in order to enhance robustness, a series of post-processing is carried out on the horizontal strings, and the possibly existing horizontal insulator strings which are not fully extracted are completed.
Preferably, step 6 comprises:
step 6.1: and for step 5.3.4, the intersection point possibly cannot be found in the ZX projection, and the method is completed by adopting a post-processing method. If the layer has the horizontal string extracted, the length of the horizontal string in the three-dimensional space is calculated. And calculating the length of the insulator string in the ZX projection according to the direction vector of the insulator string. From this length the intersection ZX coordinate is directly calculated.
Step 6.2: the method is almost the same as the step 6.1, only when the horizontal string is not extracted from the same layer, if the electric tower has the extracted horizontal string, the length of the insulator string extracted from the electric tower is taken as the length of the horizontal string in the three-dimensional space, and the intersecting point ZX coordinate is completely supplemented.
Step 6.3: the method is almost the same as the step 6.1, except that when the electric tower does not extract the transverse string, a predefined length is taken as the length of the transverse string in the three-dimensional space, and the intersection point ZX coordinate is supplemented.
It will be understood by those skilled in the art that the foregoing is merely a preferred embodiment of this invention, and is not intended to limit the invention to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. An automatic extraction method of an electric tower insulator string based on three-dimensional point cloud is characterized by comprising the following steps:
step 1: the method comprises the steps of obtaining three-dimensional point cloud data of an electric tower and the center of the electric tower, further cutting the three-dimensional point cloud data of the electric tower in the horizontal XY direction and the vertical Z direction to obtain a cut point cloud;
step 2: for the clipped point cloud data2(x, y, z) acquiring certain range of point cloud data at the topmost layer3(x, y, z), calculating the spin-yaw angle kappa, and then point cloud data2Rotating (x, Y, z) to make the direction of the tower of the electric tower consistent with the direction of the coordinate axis Y, and obtaining point cloud data4(x,y,z)；
And step 3: rotating point cloud data4(x, y, Z), performing projection to obtain a ZX projection picture, layering according to the wire height to obtain Z of intersection of wires at two sides of each layer of the electric towerAt the approximate position z _ layerkK is the number of the layers obtained by layering;
and 4, step 4: cutting to obtain point clouds of each layer based on each layer of position Z obtained by layering, wherein the uppermost layer is used as a top layer, and the types of insulator strings in the non-top layer are judged to be vertical string layers and horizontal string layers;
and 5: processing the point clouds on each layer of the electric tower by adopting different methods based on the types of the point clouds, and finally obtaining suspension points on the top layer, obtaining corresponding vertical insulator strings on the vertical string layer and obtaining corresponding horizontal insulator strings on the horizontal layer;
step 6: in consideration of the problem of data defect, in order to enhance robustness, a series of post-processing is carried out on the horizontal strings, and the possibly existing horizontal insulator strings which are not fully extracted are completed.
2. The automatic extraction method for the electric tower insulator string based on the three-dimensional point cloud as claimed in claim 1, wherein the three-dimensional point cloud data of the electric tower in the step 1 is as follows:
data0(x,y,z)＝{data0 k(xk,yk,zk),k∈[1,K]}
wherein, the data0(x, y, z) is the electric tower three-dimensional point cloud data, data0 k(xk,yk,zk) Is the pixel, x, of the kth point in the three-dimensional point cloud data of the electric towerkIs the X-direction coordinate, y of the kth point in the three-dimensional point cloud data of the electric towerkIs the Y-direction coordinate, z of the kth point in the three-dimensional point cloud data of the electric towerkThe Z-direction coordinate of the kth point in the electric tower three-dimensional point cloud data is shown, and K is the number of three-dimensional points in the electric tower three-dimensional point cloud data;
the horizontal XY direction cutting in the step 1 comprises the following steps:
according to the electric tower center, i.e. center0(x0,y0,z0)；
Three-dimensional point cloud data of electric tower0Search for abs (x) in (x, y, z)k-x0)＜threshold0And abs (y)k-y0)＜threshold0Point of (2), construct horizontal direction cutThe subsequent point cloud is data1(x,y,z)，threshold0Is a threshold parameter;
step 1, the vertical Z direction cutting is as follows:
combining the point cloud (data) cut in the horizontal direction1(x, y, z), search for abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The minimum value of Z-direction coordinates in the point cloud in the range is defined as Zmin(ii) a Search for satisfied abs (x)k-x0)＜threshold1And abs (y)k-y0)＜threshold1The maximum value of Z-direction coordinates in the point cloud in the range is defined as Zmax，threshold1Is a threshold parameter;
defining the interlayer spacing as z _ interval 1.5;
the number of layers is calculated as:
z_num＝ceil[(zmax-zmin)/z_interval]
wherein ceil represents an rounding-up function, and Z _ num represents the number of layers;
defining the point cloud density of each layer as:
z_density＝{z_densityl,l∈[1,z_num]}
and initialize z _ diversityl0, and layer 1 is the highest layer;
traversal data1(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval]Calculating the z _ now of the layer to which the kth point belongs, and counting the number of the points of each layer;
from layer 1, find that z _ dense is satisfieda+s＞10，s＝[1,6]The layer a of (a) calculates a retention value Z _ up in the Z direction from Z _ up ═ Z _ max- (a-1) × Z _ interval;
point cloud data1(x, y, z) satisfies zk≥z_min+threshold2And z iskKeeping points less than or equal to z _ up, updating the highest point z _ max to z _ up of the point cloud, and updating the lowest point z _ min to z _ min + threshold2，threshold2Is a threshold parameter;
finally, the point cloud data after cutting is obtained2(x,y,z)。
3. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 1, wherein the step 2 comprises the following steps:
step 2.1: from the point cloud data2Starting from the highest point (x, y, z) downwards, setting a layering interval z _ interval1The number of layers z _ num is calculated as 0.11＝ceil[(zmax-zmin)/z_interval1]Defining the density z _ density of each layer point cloud1＝{z_density1 k,k∈[1,z_num1]And initialize z _ dense1 k0, and layer 1 is the highest layer; traversal point cloud data2(x, y, z) according to z _ now ═ ceil [ (z)max-zk)/z_interval1]Calculating the layer to which the kth point belongs, and counting the number of points in each layer; from layer 1, find that z _ dense is satisfied1 kLayer k more than 100, calculating the maximum value Z of the top point cloud in the Z directiontopMax, point-to-point cloud data2(x, y, z) Retention satisfies zk≥ztopMax-3 and zk≤ztopTaking the point of max as the top-level point cloud data3(x,y,z)；
Step 2.2: data of top point cloud3Projecting downwards (x, y, z) to obtain XY projection image (all the projections mentioned in the invention are binary projections, and the specific method for generating XY projection at the position is as follows, and the other conditions are similar; according to point cloud data3Setting the two-dimensional grid interval to grid _ interval 0.1 according to the maximum value and the minimum value of XY coordinates of (x, y, z), and according to row ceil [ (x _ max-x _ min)/grid _ interval]Calculating the number of XY projection drawing lines, col ═ ceil [ (y _ max-y _ min)/grid _ interval]Calculating the number of XY projection diagram rows and traversing the point cloud data3(x, y, z) according to row _ now ═ ceil [ (x _ max-x) xk)/grid_interval]Calculating the line number of the projection drawing where the current point falls according to col _ now ═ ceil [ (y _ max-y)k)/grid_interval]Calculating the column number of the current point falling on the projection graph, if a pixel has a point cloud falling in the range, the pixel value is 255, otherwise, the pixel value is 0), performing morphological filling and corrosion operation on the XY projection, and extracting the maximum outline; calculating a rotation deviation angle kappa according to the first main direction of the maximum contour by utilizing a PCA method;
step 2.3: according to the calculated rotating deflection angle kappa, the point cloud data after cutting is processed2(x, y, z) at the center of the electric tower0(x0,y0,z0) The rotation center is the Z axis of the coordinate system, the rotation angle is calculated according to the deflection angle kappa, the rotation is carried out, the direction of the tower is consistent with the direction of the Y axis of the coordinate system, and the point cloud data after rotation is obtained4(x,y,z)。
4. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 1, wherein the step 3 comprises the following steps:
step 3.1: obtaining a ZX projection image by projection; according to the fact that the number of points of the row of the electric tower is larger than that of the points of the row of the electric wire, the row of the center of the tower and the row of the left and right boundaries of the tower are found, the pixel value is set to be 0, and the electric tower is removed;
step 3.2: taking the lowest point of the ZX image as a seed point, growing to obtain the interference of residual tree barriers on the ground and the like, and setting the pixel value of the part of the point as 0, so that the rest part of the image basically only comprises electric wires;
step 3.3: and (3) performing edge extraction on the ZX projection image, then performing Hough line detection on the edge image, and screening according to the vertical distance of the slope of the line, thereby obtaining the position of each layer of electric wire in the row of the tower center column.
5. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 1, wherein the step 4 comprises the following steps:
step 4.1: the uppermost layer is regarded as the top layer and does not participate in type judgment; for the non-top layer, respectively taking small-range point clouds up and down according to the calculated Z position, and projecting downwards to obtain an XY projection diagram;
step 4.2: and counting the continuous conditions of the pixel points in a certain row range from top to bottom according to the row of the center of the tower, wherein if the continuous conditions indicate that the layer has the tower, the layer belongs to a horizontal string layer, and otherwise, the layer belongs to a vertical string layer.
6. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 1, wherein the step 5 comprises the following steps:
step 5.1: the uppermost layer of the electric tower layering result is regarded as the top layer, and the suspension point is extracted by processing;
step 5.2: for the non-top layer of the electric tower layering result, the type is judged to be the part of the vertical string layer, and the horizontal insulator string is processed and extracted;
step 5.3: and (4) for the non-top layer of the electric tower layering result, the type is judged to be the part of the transverse string layer, and the vertical insulator string is extracted by processing.
7. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 6,
step 5.1 comprises:
step 5.1.1: the top point cloud projection obtains a ZX projection diagram, the number of each row of points is counted, an area boundary row with the maximum density is found, and the X coordinate of the hanging point is calculated according to the mean value of the boundary row;
step 5.1.2: starting from the boundary columns found in the step 5.1.1 to two sides respectively until the columns with small enough width are found, and calculating the Z coordinate of the hanging point according to the initial row of the columns;
step 5.1.3: performing point cloud projection to obtain an XY projection image; moving the wire from the upper end and the lower end to a line where the hanging point is located, and calculating the Y coordinate of the hanging point;
step 5.1.4: and (3) rotating the calculated three-dimensional coordinates of the hanging points by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the hanging points in the original coordinate system.
8. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 6,
step 5.2 comprises the following steps:
step 5.2.1: the point cloud is projected downwards to obtain an XY projection diagram, the number of the electric wires is counted according to the interval existing between different electric wires on the same layer, and the range of each electric wire Y is obtained;
step 5.2.2: obtaining a ZX projection image by the point cloud projection of each wire, filling, refining and deburring the ZX projection image, searching a three-way point for the refined image, wherein the point is the intersection point of the insulator string and the wire, and calculating the ZX coordinate of the intersection point;
step 5.2.3: taking a certain range of point cloud by using the ray corresponding to the intersection point ZX, and taking the mean value of the cloud Y of the point cloud as the Y coordinate of the intersection point, thus obtaining the three-dimensional coordinate of the intersection point;
step 5.2.4: according to the three-dimensional coordinates of the intersection point, a certain range of point cloud is upwards taken to obtain ZY projection, a row with a certain width, namely the position of a suspension point Z, is upwards found from the row where the intersection point is located, and the XY coordinates of the suspension point are consistent with those of the intersection point, so that the three-dimensional coordinates of the suspension point are obtained;
step 5.2.5: and (3) rotating by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical string in the original coordinate system according to the three-dimensional coordinates of the vertical string intersection point and the suspension point calculated by each wire.
9. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 6,
step 5.3 comprises the following steps:
step 5.3.1: the point cloud is projected downwards to obtain an XY projection image, and parts outside a certain range above and below the center line of the tower are cut and removed in the XY projection;
step 5.3.2: analyzing connected domains of the rest parts of the cutting scissors, solving angular points at two sides of the two connected domains, namely 4 angular points at the inner side of the H-shaped pattern, and slightly expanding the coordinates of the angular points to obtain XY coordinates of the hanging points;
step 5.3.3: cutting according to the XY coordinates of the suspension points and the position of the center X of the tower to obtain a part of point cloud which only comprises the tower and does not comprise the electric wire and the insulator string, projecting to obtain a ZY projection diagram, and finding a row with jumping of 255 points as a row of the Z coordinate of the transverse string suspension point;
step 5.3.4: the layer of point cloud projection obtains a ZX projection diagram, and the highest row of the electric wire, the lowest row of the electric wire and the highest row of the suspended object are traversed and recorded row by row according to the principle that the left suspension point is found to the left and the right suspension point is found to the right; taking the difference between the wire highest behavior reference condition and the wire lowest row and the suspension object highest row as a judgment condition, and carrying out classification discussion and fitting to obtain a ZX value;
step 5.3.5: cutting the layer of point cloud according to the ZX coordinates of the cross-string intersection points, taking a certain range of point cloud by using rays corresponding to the intersection points ZX, dividing the point cloud into two clusters according to the Y coordinates of the cluster cloud, and taking the average value of each cluster of point cloud Y as the Y coordinate of each cross-string intersection point;
step 5.3.6: and (3) rotating the three-dimensional coordinates of each transverse string intersection point and each suspension point calculated according to the layer point cloud by a-kappa angle according to the rotating direction in the step (2) to obtain the three-dimensional coordinates of the vertical strings in the original coordinate system.
10. The automatic extraction method of the three-dimensional point cloud-based electric tower insulator string according to claim 1, wherein the step 6 comprises the following steps:
step 6.1: for the step 5.3.4, the intersection point possibly cannot be found in the ZX projection, a post-processing method is adopted for perfection; if the layer has the horizontal strings to be extracted, calculating the length of the horizontal strings in the three-dimensional space; calculating the length of the insulator string in the ZX projection according to the direction vector of the insulator string; directly calculating the ZX coordinate of the intersection point according to the length;
step 6.2: the processing method is approximately the same as the processing method in the step 6.1, and only when the horizontal string is not extracted from the same layer, if the electric tower has the extracted horizontal string, the length of the insulator string extracted from the electric tower is taken as the length of the horizontal string in the three-dimensional space to complement the ZX coordinate of the intersection point;
step 6.3: the method is almost the same as the step 6.1, except that when the electric tower does not extract the transverse string, a predefined length is taken as the length of the transverse string in the three-dimensional space, and the intersection point ZX coordinate is supplemented.
CN202110300702.1A 2021-03-22 2021-03-22 Automatic extraction method of electric tower insulator string based on three-dimensional point cloud Active CN112991303B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110300702.1A CN112991303B (en) 2021-03-22 2021-03-22 Automatic extraction method of electric tower insulator string based on three-dimensional point cloud

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110300702.1A CN112991303B (en) 2021-03-22 2021-03-22 Automatic extraction method of electric tower insulator string based on three-dimensional point cloud

Publications (2)

Publication Number Publication Date
CN112991303A true CN112991303A (en) 2021-06-18
CN112991303B CN112991303B (en) 2023-08-18

Family

ID=76332783

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110300702.1A Active CN112991303B (en) 2021-03-22 2021-03-22 Automatic extraction method of electric tower insulator string based on three-dimensional point cloud

Country Status (1)

Country Link
CN (1) CN112991303B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103744661A (en) * 2013-12-23 2014-04-23 广东电网公司电力科学研究院 Multi-sensor data integration processing method and system of minimum altitude unmanned aerial vehicle
CN107680102A (en) * 2017-08-28 2018-02-09 国网甘肃省电力公司电力科学研究院 A kind of airborne cloud data electric force pole tower extraction method based on space constraint
CN110780681A (en) * 2019-11-26 2020-02-11 贵州电网有限责任公司 Unmanned aerial vehicle autonomous inspection insulator path planning method based on laser point cloud
CN111830528A (en) * 2020-06-29 2020-10-27 西安交通大学 Tower characteristic point automatic identification and inclination parameter automatic measurement method based on laser point cloud

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103744661A (en) * 2013-12-23 2014-04-23 广东电网公司电力科学研究院 Multi-sensor data integration processing method and system of minimum altitude unmanned aerial vehicle
CN107680102A (en) * 2017-08-28 2018-02-09 国网甘肃省电力公司电力科学研究院 A kind of airborne cloud data electric force pole tower extraction method based on space constraint
CN110780681A (en) * 2019-11-26 2020-02-11 贵州电网有限责任公司 Unmanned aerial vehicle autonomous inspection insulator path planning method based on laser point cloud
CN111830528A (en) * 2020-06-29 2020-10-27 西安交通大学 Tower characteristic point automatic identification and inclination parameter automatic measurement method based on laser point cloud

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
LIU CHE 等: "Automatic Recognition Method of Substation Pillar Insulator Strings Based On D-S Evidence Rheory", IEEE *
马志学 等: "基于交比不变性的绝缘子串中轴线提取方法", 电子测量技术, vol. 40, no. 6 *

Also Published As

Publication number Publication date
CN112991303B (en) 2023-08-18

Similar Documents

Publication Publication Date Title
CN112070769B (en) Layered point cloud segmentation method based on DBSCAN
CN106022381B (en) Automatic extraction method of street lamp pole based on vehicle-mounted laser scanning point cloud
CN106650640B (en) Negative obstacle detection method based on laser radar point cloud local structure characteristics
CN106815847B (en) Trees dividing method and single tree extracting method based on laser radar point cloud
CN109165549B (en) Road identification obtaining method based on three-dimensional point cloud data, terminal equipment and device
CN106529469B (en) Unmanned aerial vehicle-mounted LiDAR point cloud filtering method based on self-adaptive gradient
CN114488073A (en) Method for processing point cloud data acquired by laser radar
CN109215112B (en) Method for marking single-side point cloud model
CN111783721B (en) Lane line extraction method of laser point cloud and electronic equipment
CN113920134B (en) Slope ground point cloud segmentation method and system based on multi-line laser radar
CN111462134A (en) Single-tree segmentation method and system for fusing high-resolution remote sensing image and laser radar point cloud
CN112883878B (en) Automatic point cloud classification method under transformer substation scene based on three-dimensional grid
JP6341264B2 (en) Deposition object recognition method, apparatus, and sorting system using equipment
CN108919295A (en) Airborne LiDAR point cloud road information extracting method and device
CN115690081A (en) Tree counting method, system, storage medium, computer equipment and terminal
CN115205690B (en) Method and device for extracting street tree in monomer mode based on MLS point cloud data
Zhang et al. Filtering photogrammetric point clouds using standard lidar filters towards dtm generation
CN108765478A (en) It is a kind of to build the density clustering algorithm that separating monomer is built in point cloud
CN111950589B (en) Point cloud region growing optimization segmentation method combined with K-means clustering
CN115063555A (en) Method for extracting vehicle-mounted LiDAR point cloud street tree growing in Gaussian distribution area
AU2010200144A1 (en) Extraction processes
CN113379919A (en) Vegetation canopy height rapid extraction method based on unmanned aerial vehicle RGB camera
CN115825984A (en) Power transmission line inspection system and inspection method based on laser point cloud
CN114862715A (en) TIN (triangulated irregular network) progressive encryption denoising method fusing terrain feature semantic information
CN114332134A (en) Building facade extraction method and device based on dense point cloud

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
GR01 Patent grant
GR01 Patent grant