CN114241158A - Component identification method, device, equipment and medium based on fan point cloud data - Google Patents
Component identification method, device, equipment and medium based on fan point cloud data Download PDFInfo
- Publication number
- CN114241158A CN114241158A CN202111559981.XA CN202111559981A CN114241158A CN 114241158 A CN114241158 A CN 114241158A CN 202111559981 A CN202111559981 A CN 202111559981A CN 114241158 A CN114241158 A CN 114241158A
- Authority
- CN
- China
- Prior art keywords
- fan
- point
- cloud data
- skeleton
- component
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000006870 function Effects 0.000 claims abstract description 45
- 230000011218 segmentation Effects 0.000 claims abstract description 20
- 238000001914 filtration Methods 0.000 claims description 13
- 238000003860 storage Methods 0.000 claims description 9
- 238000005260 corrosion Methods 0.000 claims description 7
- 230000007797 corrosion Effects 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 7
- 238000002372 labelling Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 238000007689 inspection Methods 0.000 abstract description 17
- 238000010586 diagram Methods 0.000 description 12
- 238000010248 power generation Methods 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000010339 dilation Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method, a device, equipment and a medium for identifying components based on point cloud data of a fan. The method comprises the following steps: acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image, wherein the fan skeleton image is a binary image; performing skeleton segmentation on the fan skeleton image by adopting a preset path-finding mode, and combining the characteristics of fan parts to obtain a part fitting function of each fan part; and dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components. The invention solves the problem that the fan inspection work needs a large amount of manpower because the fan components cannot be accurately identified by the computer at present, realizes the quick and accurate division of the fan point cloud data to obtain the component point cloud data of each fan component, is convenient for carrying out accurate inspection on each component, quickly completes the inspection task, saves time and labor, saves money and ensures the safe operation of the wind turbine generator.
Description
Technical Field
The embodiment of the invention relates to the technical field of image processing, in particular to a method, a device, equipment and a medium for identifying components based on point cloud data of a fan.
Background
With the development of science and technology, the operation and maintenance modes of new energy power generation enterprises such as wind power and photovoltaic are gradually transiting from traditional manpower inspection to automatic and intelligent inspection, and in the process, the increasingly mature unmanned aerial vehicle technology gradually becomes the main means of relying on of power production enterprises.
In recent years, in the domestic power inspection work, inspection personnel mainly use an unmanned aerial vehicle flight platform as a carrier, carry a high-resolution visible light camera to observe and record the surface state of a blade, and return an image to operation and maintenance personnel. Unmanned aerial vehicle patrols and examines and can not leak each detail of fan, does all-round multi-angle "physical examination" for the blade. Unmanned aerial vehicle uses can greatly promote the efficiency that visualizes and detect in the middle of the wind-powered electricity generation blade detects, reduces and shuts down the loss, prevents the blade failure enlargement.
Because the fan image is that unmanned aerial vehicle gathers from each angle of fan and obtains, when carrying out the problem investigation to the image, the fan part in the image is very difficult to distinguish to the computer, consequently mainly relies on the manual work to carry out the investigation to passback image at present, looks for the trouble that the fan probably exists, still needs a large amount of manpowers to patrolling and examining of fan.
Disclosure of Invention
The invention provides a component identification method, a component identification device, equipment and a medium based on fan point cloud data, and aims to accurately identify fan component data in power generation fan point cloud data.
In a first aspect, an embodiment of the present invention provides a component identification method based on wind turbine point cloud data, including:
acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image, wherein the fan skeleton image is a binary image;
performing skeleton segmentation on the fan skeleton image by adopting a preset path-finding mode, and combining the characteristics of fan parts to obtain a part fitting function of each fan part;
and dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components.
Optionally, the fan point cloud data is projected to a blade plane and skeleton extraction is performed to obtain a fan skeleton image, including:
performing coordinate transformation on the fan point cloud data to obtain earth point cloud data under an earth coordinate system;
determining a blade plane according to the earth point cloud data, and projecting the earth point cloud data to the blade plane to obtain an initial image of the fan;
and performing expansion operation and corrosion operation on the initial image of the fan to obtain a fan skeleton image.
Optionally, the method of performing skeleton segmentation on the fan skeleton image by using a preset path finding mode, and obtaining a component fitting function of each fan component by combining with characteristics of the fan components includes:
performing skeleton division on an initial skeleton in the fan skeleton image by adopting a preset path-finding mode to obtain at least one initial path point set, and performing burr filtering on the fan skeleton image according to each initial path point set to obtain a target skeleton image;
performing skeleton division on a target skeleton in the target skeleton image by adopting a preset path finding mode to obtain at least one component path point set;
and respectively carrying out linear fitting on each component path point set to obtain a corresponding component fitting function, and carrying out component name labeling on each component fitting function according to the characteristics of the fan components.
Optionally, the step of presetting the way searching mode includes:
determining all end points of a skeleton in an image, wherein the image is the fan skeleton image or the target skeleton image, and the skeleton is the initial skeleton or the target skeleton;
establishing a path point set aiming at each end point, adding the end point to the path point set, determining the connecting point number of the path point connected with the path finding starting point by taking the end point as the path finding starting point, if the connecting point number is equal to 1, adding the path point to the path point set, and re-determining the path point as a new path finding starting point to search the path point, wherein the path point set is the initial path point set or the component path point set.
Optionally, the method for determining the endpoint includes:
aiming at each non-edge pixel point in the image, dividing a nine-pixel point set by taking the non-edge pixel point as a center, determining a total pixel value of the nine-pixel point set, and determining a non-edge pixel point corresponding to the nine-pixel point set with the total pixel value of 2 as an end point;
aiming at each edge pixel point in the image, dividing a six-pixel point set by taking the edge pixel point as a center, determining a total pixel value of the six-pixel point set, and determining an edge pixel point corresponding to the six-pixel point set with the total pixel value of 2 as an end point;
aiming at each corner pixel point in the image, dividing a four-pixel point set by taking the corner pixel point as a center, determining a total pixel value of the four-pixel point set, and determining the corner pixel point corresponding to the four-pixel point set with the total pixel value of 2 as an end point.
Optionally, the filtering burrs of the fan skeleton image according to each initial path point set to obtain a target skeleton image includes:
determining the initial path point number of the initial path point contained in each initial path point set;
and filtering the initial path point set of which the initial path point number is less than the burr point number threshold value to obtain a target skeleton image.
Optionally, after obtaining the component point cloud data of each fan component, the method further includes:
and optimizing the point cloud data of each component, and determining the component attitude information of each fan component.
In a second aspect, an embodiment of the present invention further provides a component identification apparatus based on wind turbine point cloud data, where the apparatus includes:
the skeleton image determining module is used for acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image;
the component skeleton segmentation module is used for performing skeleton segmentation on the fan skeleton image in a preset path finding mode and obtaining a component fitting function of each fan component by combining the characteristics of the fan components;
and the point cloud data identification module is used for dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components.
In a third aspect, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the method for identifying a component based on wind turbine point cloud data according to any embodiment of the present invention when executing the program.
In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a method for identifying a component based on wind turbine point cloud data according to any embodiment of the present invention.
The method comprises the steps of obtaining fan point cloud data, projecting the fan point cloud data to a blade plane, extracting a framework to obtain a fan framework image, performing framework segmentation on the fan framework image by adopting a preset path finding mode, obtaining component fitting functions of each fan component by combining fan component characteristics, dividing the fan point cloud data according to the component fitting functions to obtain component point cloud data of each fan component, solving the problem that a large amount of manpower is needed for fan inspection work due to the fact that a computer cannot accurately identify the fan components at present, achieving the purpose of rapidly and accurately dividing the fan point cloud data to obtain the component point cloud data of each fan component, facilitating accurate inspection on each fan component, rapidly completing inspection tasks, saving time, labor and money, and simultaneously ensuring safe operation of a wind turbine unit.
Drawings
Fig. 1 is a flowchart of a component identification method based on wind turbine point cloud data according to an embodiment of the present invention;
fig. 2a is a flowchart of a component identification method based on wind turbine point cloud data according to a second embodiment of the present invention;
fig. 2b is a schematic diagram of a first principle of a component identification method based on wind turbine point cloud data according to a second embodiment of the present invention;
fig. 2c is a schematic diagram of a second principle of the component identification method based on the wind turbine point cloud data according to the second embodiment of the present invention;
fig. 2d is a schematic diagram of a third principle of the component identification method based on the point cloud data of the wind turbine according to the second embodiment of the present invention;
fig. 2e is a fourth schematic diagram of a component identification method based on wind turbine point cloud data according to the second embodiment of the present invention;
fig. 3 is a block diagram of a component identification apparatus based on wind turbine point cloud data according to a third embodiment of the present invention;
fig. 4 is a block diagram of a computer device according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be noted that, for convenience of description, only a part of the structures related to the present invention, not all of the structures, are shown in the drawings, and furthermore, embodiments of the present invention and features of the embodiments may be combined with each other without conflict.
Example one
Fig. 1 is a flowchart of a method for identifying a component based on fan point cloud data according to an embodiment of the present invention, where the embodiment is applicable to a case of identifying a component of fan point cloud data of a power generation fan, and the method may be executed by a component identification device based on fan point cloud data, and the device may be implemented by software and/or hardware.
As shown in fig. 1, the method specifically includes the following steps:
and 110, acquiring fan point cloud data, projecting the fan point cloud data to a blade plane, and extracting a skeleton to obtain a fan skeleton image.
The fan point cloud data can be three-dimensional data acquired by a laser radar. The fan skeleton image may be a binary image.
Specifically, at the in-process of patrolling and examining the electricity generation fan, unmanned aerial vehicle lidar carries out data acquisition, and data analysis carries out after uploading the airborne computer on the data. The coordinate system change can be carried out after the point cloud data of the fan is obtained, the point cloud data under the radar coordinate system of the unmanned aerial vehicle is changed to be under the geodetic coordinate system, the point cloud data of the fan is analyzed to determine the blade plane of the fan, the three-dimensional point cloud data is projected to the blade plane, and a two-dimensional fan image is obtained. And continuously performing skeleton extraction operations such as expansion corrosion and the like on the fan image to obtain a fan skeleton image.
And 120, performing skeleton segmentation on the fan skeleton image by adopting a preset path-finding mode, and combining the characteristics of fan parts to obtain a part fitting function of each fan part.
In this embodiment, for a fan skeleton graph in a fan skeleton image, a preset path finding mode can be adopted to traverse a skeleton path from a skeleton endpoint, the skeleton path is segmented when bifurcation exists, and the segmented skeleton is subjected to straight line fitting to obtain a part fitting function of each part of the fan. In practical applications, the power generation fan generally has three blades, the difference between the blades is 120 degrees, and the tower of the fan is generally vertical to the ground. According to the characteristics of the fan component, the fitting function of the segmented component skeleton can be identified and labeled.
Optionally, the implementation step of the preset way searching manner may include:
a. all endpoints of the skeleton in the image are determined.
b. And establishing a path point set aiming at each end point, adding the end points to the path point set, determining the connecting points of the path points connected with the path finding starting points by taking the end points as the path finding starting points, adding the path points to the path point set if the connecting points are equal to 1, and re-determining the path points as new path finding starting points to search the path points.
And step 130, dividing the fan point cloud data according to the fitting functions of the components to obtain the component point cloud data of the components of the fan.
Specifically, the original fan point cloud data can be correspondingly classified according to the component fitting function of each fan component, so as to obtain the partitioned and classified component point cloud data.
According to the technical scheme, fan point cloud data are obtained, the fan point cloud data are projected to a blade plane and extracted through a framework, a fan framework image is obtained, framework segmentation is carried out on the fan framework image through a preset path finding mode, fan part characteristics are combined, part fitting functions of all fan parts are obtained, the fan point cloud data are divided according to the part fitting functions, part point cloud data of all fan parts are obtained, the problem that a large amount of manpower is needed for fan inspection work due to the fact that a computer cannot accurately identify the fan parts at present is solved, quick and accurate division of the fan point cloud data is achieved, the part point cloud data of all fan parts are obtained, accurate inspection can be conveniently carried out on all fan parts, inspection tasks are rapidly completed, time, labor and money are saved, and meanwhile safe operation of a wind turbine unit can be guaranteed.
Example two
Fig. 2a is a flowchart of a component identification method based on wind turbine point cloud data according to a second embodiment of the present invention. On the basis of the above embodiments, the method for identifying a component based on wind turbine point cloud data is further optimized.
As shown in fig. 2a, the method specifically includes:
and step 210, acquiring fan point cloud data, and performing coordinate transformation on the fan point cloud data to obtain geodetic point cloud data under a geodetic coordinate system.
Specifically, in the inspection process, the laser radar performs data acquisition, the component recognition device obtains fan point cloud data acquired by the laser radar and can perform initial pose transformation on the fan point cloud data, the RT matrix obtained by adding rotation transformation and displacement between the laser radar and the unmanned aerial vehicle and unmanned aerial vehicle posture transformation is added, coordinate system conversion is performed on the fan point cloud data, and geodetic point cloud data under a geodetic coordinate system are obtained.
And step 220, determining a blade plane according to the earth point cloud data, and projecting the earth point cloud data to the blade plane to obtain an initial image of the fan.
Specifically, the point cloud data are three-dimensional data, so that a blade plane where three blades of the fan are located can be determined according to the earth point cloud data, and the three-dimensional point cloud data are projected to the blade plane to obtain a two-dimensional fan initial image. The problem that point cloud data are not uniformly distributed may be caused due to the included angle between laser beams of data of the multi-line laser radar, in order to guarantee that the data are uniform, the data after projection can be subjected to down-sampling, and the two-dimensional data after down-sampling form a fan initial image.
Fig. 2b is a schematic diagram of a first principle of a component identification method based on wind turbine point cloud data according to a second embodiment of the present invention. Fig. 2b shows a wind turbine initial image obtained by projecting earth point cloud data of a certain power generation wind turbine to a blade plane.
And step 230, performing expansion operation and corrosion operation on the initial image of the fan to obtain a fan skeleton image.
In this embodiment, the fan skeleton image may be obtained by performing an expansion operation on the fan initial image and then performing a corrosion operation.
Fig. 2c is a schematic diagram of a second principle of the component identification method based on the wind turbine point cloud data according to the second embodiment of the present invention. Fig. 2c is an image obtained after the dilation operation is performed on fig. 2 b.
Fig. 2d is a schematic diagram of a third principle of the component identification method based on the wind turbine point cloud data according to the second embodiment of the present invention. Fig. 2d is an image obtained after the etching operation is performed on fig. 2 c.
And 240, performing skeleton division on an initial skeleton in the fan skeleton image by adopting a preset path finding mode to obtain at least one initial path point set, and performing burr filtering on the fan skeleton image according to each initial path point set to obtain a target skeleton image.
Optionally, step 240 may be implemented by the following steps:
s2401, determining all end points of an initial skeleton in the fan skeleton image.
Further, the method for determining the endpoint may include:
aiming at each non-edge pixel point in the fan skeleton image, dividing a nine-pixel point set by taking the non-edge pixel point as a center, determining a total pixel value of the nine-pixel point set, and determining a non-edge pixel point corresponding to the nine-pixel point set with the total pixel value of 2 as an end point;
aiming at each edge pixel point in the fan skeleton image, dividing a six-pixel point set by taking the edge pixel point as a center, determining a total pixel value of the six-pixel point set, and determining an edge pixel point corresponding to the six-pixel point set with the total pixel value of 2 as an end point;
aiming at each corner pixel point in the fan skeleton image, dividing a four-pixel point set by taking the corner pixel point as a center, determining a total pixel value of the four-pixel point set, and determining the corner pixel point corresponding to the four-pixel point set with the total pixel value of 2 as an end point.
In this embodiment, the fan skeleton image is a binary image, as shown in fig. 2d, a pixel value of a pixel point where the initial skeleton is located may be represented by 1, and a pixel value of the black background may be represented by 0. The pixel points can be divided into non-edge pixel points, edge pixel points and corner pixel points according to the positions of the pixel points in the image. The corner pixel points can be understood as four pixel points at four corners of the image, the edge pixel points can be understood as pixel points at the edge of the image but not the corner pixel points, and other pixel points except the corner pixel points and the edge pixel points in the image can be called non-edge pixel points.
When searching for the end point of the initial skeleton, each pixel point and the surrounding pixel points can be analyzed to determine whether the pixel point is an end point of the initial skeleton. For each non-edge pixel point, a 3 x 3 pixel point grid can be divided by taking the non-edge pixel point as a center, a set formed by nine pixel points in the grid is called a nine-pixel-point set of the non-edge pixel point, values of all the pixel points in the nine-pixel-point set are added to obtain a total pixel value, and when the total pixel value is 2, the non-edge pixel point can be considered as an end point. For each edge pixel point, the edge pixel point can be taken as the center, a set formed by five pixel points around the edge pixel point is called a six-pixel-point set of the edge pixel point, values of all the pixel points in the six-pixel-point set are added to obtain a total pixel value, and when the total pixel value is 2, the edge pixel point can be considered as an end point. For each corner pixel point, the corner pixel point can be taken as a center, a set formed by the corner pixel point and three surrounding pixel points is called a four-pixel-point set of the corner pixel point, values of all pixel points in the four-pixel-point set are added to obtain a total pixel value, and when the total pixel value is 2, the corner pixel point can be considered as an end point.
S2402, aiming at each end point, establishing an initial path point set, adding the end point to the initial path point set, determining the connecting point number of the path point connected with the path searching starting point by taking the end point as the path searching starting point, if the connecting point number is equal to 1, adding the path point to the initial path point set, and re-determining the path point as a new path searching starting point to search the path point.
In this embodiment, the route searching may be performed with the end point as a starting point, the skeleton path point connected to the end point is added to the initial path point set corresponding to the end point, and the route searching is stopped when the branch point is searched. The basis for judging whether the branch point is reached may be to judge whether the number of connected points of the path point connected to the seek start point is equal to 1, if equal to 1, the path point is not a branch point, and if greater than 1, the path point may be regarded as a branch point. In practical application, a 3 × 3 pixel point grid can be divided by taking the path-finding starting point as a center, the pixel values of all the pixel points in the grid are summed, and the pixel value summation is greater than 3, so that the pixel value summation can be regarded as a branch point. In the path searching process, the pixel value of the path point added to the initial path point set can be temporarily set to be 0, when the number of the connecting points of the path point connected with the path searching starting point is determined, a 3 x 3 pixel point grid can be divided by taking the path searching starting point as the center, the pixel values of all the pixel points in the grid are summed, and the pixel value summation is greater than 2, so that the pixel value summation can be regarded as a branch point. After the path searching is carried out on each endpoint, an initial path point set corresponding to each endpoint can be obtained.
S2403, determining initial path point numbers of the initial path points contained in each initial path point set.
Specifically, the number of initial path points included in each initial path point set may be different, and the number of initial path points included in each initial path point set is counted to obtain the number of initial path points corresponding to each initial path point set.
S2404, filtering the initial path point set of which the initial path point number is less than the burr point number threshold value to obtain a target skeleton image.
In this embodiment, a threshold of a number of burrs may be preset, and when the number of initial path points of a certain initial path point set is less than the threshold of the number of burrs, a skeleton path corresponding to the initial path point set may be considered as a burr, and needs to be filtered out. And forming a target skeleton image by using the residual initial path point set after the burrs are removed.
Fig. 2e is a fourth schematic diagram of a component identification method based on wind turbine point cloud data according to the second embodiment of the present invention. The skeleton image shown in fig. 2e is the target skeleton image composed of the remaining initial path point set after the burr filtering is performed on fig. 2 d.
And 250, performing skeleton division on the target skeleton in the target skeleton image by adopting a preset path finding mode to obtain at least one component path point set.
Optionally, step 250 may be implemented by the following steps:
s2501, determining all end points of the target skeleton in the target skeleton image.
The method of endpoint determination for the target skeleton may be similar to the method of endpoint determination for the initial skeleton:
aiming at each non-edge pixel point in the target skeleton image, dividing a nine-pixel point set by taking the non-edge pixel point as a center, determining a total pixel value of the nine-pixel point set, and determining a non-edge pixel point corresponding to the nine-pixel point set with the total pixel value of 2 as an end point;
aiming at each edge pixel point in the target skeleton image, dividing a six-pixel point set by taking the edge pixel point as a center, determining a total pixel value of the six-pixel point set, and determining an edge pixel point corresponding to the six-pixel point set with the total pixel value of 2 as an end point;
aiming at each corner pixel point in the target skeleton image, dividing a four-pixel point set by taking the corner pixel point as a center, determining a total pixel value of the four-pixel point set, and determining the corner pixel point corresponding to the four-pixel point set with the total pixel value of 2 as an end point.
S2502, aiming at each end point, a component path point set is established, the end points are added to the component path point set, the end points are used as path searching starting points, the connecting points of the path points connected with the path searching starting points are determined, if the connecting points are equal to 1, the path points are added to the component path point set, and the path points are determined to be new path searching starting points again to search the path points.
In this embodiment, the route searching may be performed with the end point as a starting point, the skeleton path point connected to the end point is added to the component path point set corresponding to the end point, and the route searching is stopped when the branch point is searched. The basis for judging whether the branch point is reached may be to judge whether the number of connected points of the path point connected to the seek start point is equal to 1, if equal to 1, the path point is not a branch point, and if greater than 1, the path point may be regarded as a branch point. In practical application, a 3 × 3 pixel point grid can be divided by taking the path-finding starting point as a center, and the pixel values of the pixel points in the grid are summed, wherein the pixel value summation is greater than 3, and the pixel value summation can be regarded as a branch point. After the path searching is carried out on each end point, a component path point set corresponding to each end point can be obtained.
And 260, respectively performing linear fitting on the path point sets of each component to obtain corresponding component fitting functions, and labeling the component names of the component fitting functions according to the characteristics of the fan components.
Specifically, straight line fitting can be performed on each component path point set to obtain a feature vector, namely position information of different fan components in the geodetic coordinate system. In addition, in practical application, the power generation fan generally has three blades, the difference between the blades is 120 degrees, the tower drum of the fan is generally perpendicular to the ground, and the fitted component fitting function can be identified and labeled according to the characteristics of the fan components.
And 270, dividing the fan point cloud data according to the fitting functions of the components to obtain the component point cloud data of the components of the fan.
Specifically, because different part fitting functions can represent the position information of the corresponding fan parts, the fan point cloud data can be correspondingly classified based on the part fitting functions, and the part point cloud data of the fan parts are obtained after segmentation and classification.
And 280, optimizing the point cloud data of each part, and determining the part attitude information of each fan part.
Specifically, the optimization algorithms such as least square and the like can be utilized to analyze and calculate the point cloud data of the components to obtain relatively accurate attitude information of the fan components.
The technical scheme of the embodiment includes that the method comprises the steps of obtaining wind turbine point cloud data, performing coordinate transformation on the wind turbine point cloud data to obtain earth point cloud data under an earth coordinate system, determining a blade plane according to the earth point cloud data, projecting the earth point cloud data to the blade plane to obtain a wind turbine initial image, performing expansion operation and corrosion operation on the wind turbine initial image to obtain a wind turbine skeleton image, performing skeleton division on an initial skeleton in the wind turbine skeleton image in a preset path-finding mode to obtain at least one initial path point set, performing burr filtering on the wind turbine skeleton image according to the initial path point sets to obtain a target skeleton image, performing skeleton division on a target skeleton in the target skeleton image in the preset path-finding mode to obtain at least one component path point set, performing linear fitting on the component path point sets respectively to obtain corresponding component fitting functions, and according to the characteristics of the fan components, labeling component names of the fitting functions of the components, dividing the fan point cloud data according to the fitting functions of the components to obtain component point cloud data of the fan components, optimizing the point cloud data of the components, and determining component attitude information of the fan components. The problem that the current computer can not accurately identify the fan components to cause the fan to patrol and examine the requirement on a large amount of manpower is solved, the point cloud data of the fan are quickly and accurately divided, the point cloud data of the components of the fan components are obtained, the fan components are conveniently and accurately patrolled and examined, a patrol and examine task is quickly completed, and the safe operation of a wind turbine generator set can be guaranteed while time, labor and money are saved.
EXAMPLE III
The fan point cloud data-based component identification device provided by the embodiment of the invention can execute the fan point cloud data-based component identification method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. Fig. 3 is a block diagram of a component identification apparatus based on wind turbine point cloud data according to a third embodiment of the present invention, and as shown in fig. 3, the apparatus includes: a skeleton image determination module 310, a component skeleton segmentation module 320, and a point cloud data identification module 330.
And the skeleton image determining module 310 is configured to acquire fan point cloud data, project the fan point cloud data to a blade plane, and perform skeleton extraction to obtain a fan skeleton image.
And the component skeleton segmentation module 320 is configured to perform skeleton segmentation on the fan skeleton image in a preset path finding manner, and obtain a component fitting function of each fan component by combining with characteristics of the fan components.
And the point cloud data identification module 330 is configured to divide the fan point cloud data according to each component fitting function to obtain component point cloud data of each fan component.
According to the technical scheme, fan point cloud data are obtained, the fan point cloud data are projected to a blade plane and extracted through a framework, a fan framework image is obtained, framework segmentation is carried out on the fan framework image through a preset path finding mode, fan part characteristics are combined, part fitting functions of all fan parts are obtained, the fan point cloud data are divided according to the part fitting functions, part point cloud data of all fan parts are obtained, the problem that a large amount of manpower is needed for fan inspection work due to the fact that a computer cannot accurately identify the fan parts at present is solved, quick and accurate division of the fan point cloud data is achieved, the part point cloud data of all fan parts are obtained, accurate inspection can be conveniently carried out on all fan parts, inspection tasks are rapidly completed, time, labor and money are saved, and meanwhile safe operation of a wind turbine unit can be guaranteed.
Optionally, the skeleton image determining module 310 includes:
the point cloud data acquisition unit is used for acquiring point cloud data of the fan;
the data coordinate conversion unit is used for carrying out coordinate conversion on the fan point cloud data to obtain earth point cloud data under an earth coordinate system;
the initial image determining unit is used for determining a blade plane according to the earth point cloud data and projecting the earth point cloud data to the blade plane to obtain an initial image of the fan;
and the skeleton image determining unit is used for performing expansion operation and corrosion operation on the initial image of the fan to obtain a fan skeleton image.
Optionally, the component skeleton segmentation module 320 includes:
the framework burr filtering unit is used for performing framework division on an initial framework in the fan framework image by adopting a preset path finding mode to obtain at least one initial path point set, and performing burr filtering on the fan framework image according to each initial path point set to obtain a target framework image;
the skeleton path dividing unit is used for performing skeleton division on a target skeleton in the target skeleton image by adopting a preset path finding mode to obtain at least one component path point set;
and the component characteristic fitting unit is used for respectively performing linear fitting on each component path point set to obtain a corresponding component fitting function, and performing component name labeling on each component fitting function according to the fan component characteristics.
Optionally, the step of presetting the way searching mode includes:
determining all end points of a skeleton in an image, wherein the image is the fan skeleton image or the target skeleton image, and the skeleton is the initial skeleton or the target skeleton;
establishing a path point set aiming at each end point, adding the end point to the path point set, determining the connecting point number of the path point connected with the path finding starting point by taking the end point as the path finding starting point, if the connecting point number is equal to 1, adding the path point to the path point set, and re-determining the path point as a new path finding starting point to search the path point, wherein the path point set is the initial path point set or the component path point set.
Optionally, the method for determining the endpoint includes:
aiming at each non-edge pixel point in the image, dividing a nine-pixel point set by taking the non-edge pixel point as a center, determining a total pixel value of the nine-pixel point set, and determining a non-edge pixel point corresponding to the nine-pixel point set with the total pixel value of 2 as an end point;
aiming at each edge pixel point in the image, dividing a six-pixel point set by taking the edge pixel point as a center, determining a total pixel value of the six-pixel point set, and determining an edge pixel point corresponding to the six-pixel point set with the total pixel value of 2 as an end point;
aiming at each corner pixel point in the image, dividing a four-pixel point set by taking the corner pixel point as a center, determining a total pixel value of the four-pixel point set, and determining the corner pixel point corresponding to the four-pixel point set with the total pixel value of 2 as an end point.
Optionally, the filtering burrs of the fan skeleton image according to each initial path point set to obtain a target skeleton image includes:
determining the initial path point number of the initial path point contained in each initial path point set;
and filtering the initial path point set of which the initial path point number is less than the burr point number threshold value to obtain a target skeleton image.
Optionally, the apparatus further comprises a component attitude fitting module, wherein the component attitude fitting module is configured to:
and after the component point cloud data of each fan component is obtained, optimizing the component point cloud data, and determining the component attitude information of each fan component.
Example four
Fig. 4 is a block diagram of a computer apparatus according to a fourth embodiment of the present invention, as shown in fig. 4, the computer apparatus includes a processor 410, a memory 420, an input device 430, and an output device 440; the number of the processors 410 in the computer device may be one or more, and one processor 410 is taken as an example in fig. 4; the processor 410, the memory 420, the input device 430 and the output device 440 in the computer apparatus may be connected by a bus or other means, and the connection by the bus is exemplified in fig. 4.
The memory 420 may be used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the wind turbine point cloud data-based component identification method in the embodiment of the present invention (for example, the skeleton image determination module 310, the component skeleton segmentation module 320, and the point cloud data identification module 330 in the wind turbine point cloud data-based component identification device). The processor 410 executes various functional applications and data processing of the computer device by running software programs, instructions and modules stored in the memory 420, that is, the above-mentioned component identification method based on the wind turbine point cloud data is realized.
The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 420 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 420 may further include memory located remotely from processor 410, which may be connected to a computer device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 430 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the computer apparatus. The output device 440 may include a display device such as a display screen.
EXAMPLE five
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are executed by a computer processor to perform a component identification method based on wind turbine point cloud data, and the method includes:
acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image, wherein the fan skeleton image is a binary image;
performing skeleton segmentation on the fan skeleton image by adopting a preset path-finding mode, and combining the characteristics of fan parts to obtain a part fitting function of each fan part;
and dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components.
Of course, the storage medium provided by the embodiment of the present invention includes computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the component identification method based on wind turbine point cloud data provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the component identification device based on wind turbine point cloud data, each unit and each module included in the embodiment are only divided according to functional logic, but are not limited to the above division, as long as corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A component identification method based on wind turbine point cloud data is characterized by comprising the following steps:
acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image, wherein the fan skeleton image is a binary image;
performing skeleton segmentation on the fan skeleton image by adopting a preset path-finding mode, and combining the characteristics of fan parts to obtain a part fitting function of each fan part;
and dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components.
2. The fan point cloud data-based component identification method according to claim 1, wherein the projecting the fan point cloud data to a blade plane and performing skeleton extraction to obtain a fan skeleton image comprises:
performing coordinate transformation on the fan point cloud data to obtain earth point cloud data under an earth coordinate system;
determining a blade plane according to the earth point cloud data, and projecting the earth point cloud data to the blade plane to obtain an initial image of the fan;
and performing expansion operation and corrosion operation on the initial image of the fan to obtain a fan skeleton image.
3. The fan point cloud data-based component identification method according to claim 1, wherein the fan skeleton image is subjected to skeleton segmentation in a preset path finding manner, and a component fitting function of each fan component is obtained by combining fan component characteristics, and the method comprises the following steps:
performing skeleton division on an initial skeleton in the fan skeleton image by adopting a preset path-finding mode to obtain at least one initial path point set, and performing burr filtering on the fan skeleton image according to each initial path point set to obtain a target skeleton image;
performing skeleton division on a target skeleton in the target skeleton image by adopting a preset path finding mode to obtain at least one component path point set;
and respectively carrying out linear fitting on each component path point set to obtain a corresponding component fitting function, and carrying out component name labeling on each component fitting function according to the characteristics of the fan components.
4. The wind turbine point cloud data-based component identification method according to claim 3, wherein the step of presetting a way-finding mode comprises:
determining all end points of a skeleton in an image, wherein the image is the fan skeleton image or the target skeleton image, and the skeleton is the initial skeleton or the target skeleton;
establishing a path point set aiming at each end point, adding the end point to the path point set, determining the connecting point number of the path point connected with the path finding starting point by taking the end point as the path finding starting point, if the connecting point number is equal to 1, adding the path point to the path point set, and re-determining the path point as a new path finding starting point to search the path point, wherein the path point set is the initial path point set or the component path point set.
5. The wind turbine point cloud data-based part identification method according to claim 4, wherein the end point determination method comprises:
aiming at each non-edge pixel point in the image, dividing a nine-pixel point set by taking the non-edge pixel point as a center, determining a total pixel value of the nine-pixel point set, and determining a non-edge pixel point corresponding to the nine-pixel point set with the total pixel value of 2 as an end point;
aiming at each edge pixel point in the image, dividing a six-pixel point set by taking the edge pixel point as a center, determining a total pixel value of the six-pixel point set, and determining an edge pixel point corresponding to the six-pixel point set with the total pixel value of 2 as an end point;
aiming at each corner pixel point in the image, dividing a four-pixel point set by taking the corner pixel point as a center, determining a total pixel value of the four-pixel point set, and determining the corner pixel point corresponding to the four-pixel point set with the total pixel value of 2 as an end point.
6. The fan point cloud data-based component identification method of claim 3, wherein the deburring the fan skeleton image according to each of the initial path point sets to obtain a target skeleton image comprises:
determining the initial path point number of the initial path point contained in each initial path point set;
and filtering the initial path point set of which the initial path point number is less than the burr point number threshold value to obtain a target skeleton image.
7. The method of claim 1, wherein after obtaining the component point cloud data for each of the wind turbine components, the method further comprises:
and optimizing the point cloud data of each component, and determining the component attitude information of each fan component.
8. The utility model provides a part recognition device based on fan point cloud data which characterized in that includes:
the skeleton image determining module is used for acquiring fan point cloud data, projecting the fan point cloud data to a blade plane and extracting a skeleton to obtain a fan skeleton image;
the component skeleton segmentation module is used for performing skeleton segmentation on the fan skeleton image in a preset path finding mode and obtaining a component fitting function of each fan component by combining the characteristics of the fan components;
and the point cloud data identification module is used for dividing the fan point cloud data according to the component fitting functions to obtain the component point cloud data of the fan components.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method for fan point cloud data based component identification as claimed in any one of claims 1-7.
10. A storage medium containing computer executable instructions for performing the method of any of claims 1-7 when executed by a computer processor for identifying a component based on wind turbine point cloud data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111559981.XA CN114241158A (en) | 2021-12-20 | 2021-12-20 | Component identification method, device, equipment and medium based on fan point cloud data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111559981.XA CN114241158A (en) | 2021-12-20 | 2021-12-20 | Component identification method, device, equipment and medium based on fan point cloud data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114241158A true CN114241158A (en) | 2022-03-25 |
Family
ID=80758999
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111559981.XA Pending CN114241158A (en) | 2021-12-20 | 2021-12-20 | Component identification method, device, equipment and medium based on fan point cloud data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114241158A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115097867A (en) * | 2022-08-23 | 2022-09-23 | 无锡海纳智能科技有限公司 | Method for determining unmanned aerial vehicle shooting attitude under fan inspection route |
CN115272248A (en) * | 2022-08-01 | 2022-11-01 | 无锡海纳智能科技有限公司 | Intelligent detection method for fan attitude and electronic equipment |
-
2021
- 2021-12-20 CN CN202111559981.XA patent/CN114241158A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115272248A (en) * | 2022-08-01 | 2022-11-01 | 无锡海纳智能科技有限公司 | Intelligent detection method for fan attitude and electronic equipment |
CN115272248B (en) * | 2022-08-01 | 2024-02-13 | 无锡海纳智能科技有限公司 | Intelligent detection method for fan gesture and electronic equipment |
CN115097867A (en) * | 2022-08-23 | 2022-09-23 | 无锡海纳智能科技有限公司 | Method for determining unmanned aerial vehicle shooting attitude under fan inspection route |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107808133B (en) | Unmanned aerial vehicle line patrol-based oil and gas pipeline safety monitoring method and system and software memory | |
CN114241158A (en) | Component identification method, device, equipment and medium based on fan point cloud data | |
CN109060826B (en) | Wind-powered electricity generation blade detection device that does not shut down | |
CN111311650A (en) | Point cloud data registration method and device and storage medium | |
CN111460967A (en) | Illegal building identification method, device, equipment and storage medium | |
CN112305559A (en) | Power transmission line distance measuring method, device and system based on ground fixed-point laser radar scanning and electronic equipment | |
CN112364843A (en) | Plug-in aerial image target positioning detection method, system and equipment | |
CN115631356A (en) | Road facility missing identification method and device, storage medium and electronic device | |
CN110597937A (en) | Unmanned intelligent inspection method, device, equipment and storage medium | |
CN114138012A (en) | Fan blade inspection method, device, equipment and storage medium | |
CN116052097A (en) | Map element detection method and device, electronic equipment and storage medium | |
CN115272262A (en) | Outdoor insulator surface defect identification method and device and electronic equipment | |
Fang et al. | A framework of power pylon detection for UAV-based power line inspection | |
CN116091709B (en) | Three-dimensional reconstruction method and device for building, electronic equipment and storage medium | |
CN117197571A (en) | Photovoltaic module fault detection method and device, electronic equipment and storage medium | |
CN115937492A (en) | Transformer equipment infrared image identification method based on feature identification | |
CN116309828A (en) | Three-dimensional laser point cloud data-based power pole information extraction method | |
Namouchi et al. | Piecewise horizontal 3d roof reconstruction from aerial lidar | |
CN115760575A (en) | Laser point cloud data processing method and device, electronic equipment and storage medium | |
Shen et al. | Substation one-line diagram automatic generation based on image recongnition | |
CN111144420A (en) | Small sample-based electric tower defect monitoring method and system | |
Li et al. | Design of porcelain insulator defect recognition system based on UAV line inspection image | |
Yan et al. | Research on fault diagnosis of transmission line based on SIFT feature | |
Pang et al. | Detection of Electrical Diagram based on QueryDet | |
CN114581890B (en) | Method and device for determining lane line, electronic equipment and storage medium |
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: Room 1404, Guanghua Building, No. 777, Guangzhou Road, Binjiang Street, Haimen District, Nantong City, Jiangsu Province, 226100 Applicant after: Fuya Intelligent Technology Development (Jiangsu) Co.,Ltd. Address before: 200433 3rd floor, No.1 Lane 127, Guotai Road, Yangpu District, Shanghai Applicant before: SHANGHAI FUYA INTELLIGENT TECHNOLOGY Co.,Ltd. |
|
CB02 | Change of applicant information |