CN109214457A - A kind of classification method and device of power circuit - Google Patents
A kind of classification method and device of power circuit Download PDFInfo
- Publication number
- CN109214457A CN109214457A CN201811043754.XA CN201811043754A CN109214457A CN 109214457 A CN109214457 A CN 109214457A CN 201811043754 A CN201811043754 A CN 201811043754A CN 109214457 A CN109214457 A CN 109214457A
- Authority
- CN
- China
- Prior art keywords
- data
- power line
- category attribute
- power circuit
- image
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
Abstract
The present invention provides the classification method and device of a kind of power circuit, wherein the classification method of the power circuit includes: to obtain the power line data including power line data and shaft tower data;Processing power line data obtains power circuit figure, and generates the 3 d grid image including multiple sub- grating images corresponding with power line data according to power circuit figure;Obtain the category attribute table of comparisons including the corresponding category attribute label of sub- grating image;Using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, 3 d grid image is analyzed, obtains category attribute data set corresponding with 3 d grid image;Combination of power track data and category attribute data set obtain classification data corresponding with power line data.The classification method and device of power circuit described in the invention can be improved the nicety of grading of power circuit, improve the classification effectiveness of power circuit, and realize the data processing of magnanimity.
Description
Technical field
The present invention relates to electric inspection process fields, in particular to the classification method and device of a kind of power circuit.
Background technique
It, usually will be based on power line and shaft tower included by power circuit, to carry out during electric inspection process
The sunykatuib analysis of dangerous point detection and various operating conditions, therefore power line and the relevant acquisition of information of shaft tower seem especially heavy with classification
It wants.In practice, it has been found that current power line and tower classification method generally use be manual sort method, however, people
Although the method precision of work point class is higher, the efficiency of this method is lower, cannot achieve mass data processing.
Summary of the invention
In view of the above problems, the present invention provides a kind of classification method of power circuit and device, it can be improved power line
The nicety of grading on road, improves the classification effectiveness of power circuit, and realizes the data processing of magnanimity.
To achieve the goals above, the present invention adopts the following technical scheme that:
In a first aspect, the present invention provides a kind of classification methods of power circuit, comprising:
Obtain the power line data including power line data and shaft tower data;
It handles the power line data and obtains power circuit figure, and generated and the electric power according to the power circuit figure
The corresponding 3 d grid image including multiple sub- grating images of track data;
Obtain the category attribute table of comparisons including the corresponding category attribute label of sub- grating image;
Using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, to the 3 d grid image
It is analyzed, obtains category attribute data set corresponding with the 3 d grid image;
The power line data and the category attribute data set are combined, is obtained corresponding with the power line data
Classification data.
As an alternative embodiment, the combination power line data and the category attribute data set,
After the step of obtaining classification data corresponding with the power line data, the method also includes:
Processing is optimized to the classification data with preset searching algorithm, obtains Classified optimization data.
As an alternative embodiment, described obtain the power line data including power line data and shaft tower data
The step of include:
Laser radar is controlled to the power circuit emission detection signal including power line and shaft tower;
Receive the echo-signal of the power circuit reflection;
Signal processing is carried out to the detectable signal and the echo-signal, obtains including power line data and shaft tower data
Laser radar point cloud data;
The laser radar point cloud data is determined as power line data.
As an alternative embodiment, the processing power line data, obtains including multiple sub- grid maps
The step of 3 d grid image of picture includes:
It handles the power line data and obtains power circuit figure;
Using preset reference face as foundation, slicing treatment is carried out to the power circuit figure, obtains sectioning image;
The sectioning image is divided according to default division mode, obtains multiple sub- grating images, and acquisition and institute
State multiple sub- grating images multiple gray values correspondingly;
The sectioning image and the multiple gray value are combined, 3 d grid image is obtained.
As an alternative embodiment, the acquisition includes the corresponding category attribute label of every sub- grating image
The step of category attribute table of comparisons includes:
The multiple sub- grating image that the 3 d grid image includes is sampled, sample data is obtained;
Using preset label insertion standard as foundation, the insertion of category attribute label is carried out to the sample data, is obtained
The category attribute table of comparisons including the corresponding category attribute label of sub- grating image.
Second aspect, the present invention provides a kind of sorters of power circuit, including obtain module, processing module, divide
Analyse module and composite module, wherein
The acquisition module is for obtaining the power line data including power line data and shaft tower data;
The processing module obtains power circuit figure for handling the power line data, and according to the power circuit
Figure generates the 3 d grid image including multiple sub- grating images corresponding with the power line data;
The module that obtains is also used to obtain the category attribute control including the corresponding category attribute label of sub- grating image
Table;
The analysis module is used for using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, right
The 3 d grid image is analyzed, and category attribute data set corresponding with the 3 d grid image is obtained;
The composite module obtains and the electricity for combining the power line data and the category attribute data set
The corresponding classification data of line of force circuit-switched data.
As an alternative embodiment, the processing module is also used to preset searching algorithm to the classification number
According to processing is optimized, Classified optimization data are obtained.
As an alternative embodiment, the acquisition module includes sampling unit and embedded unit, wherein
The sampling unit is obtained for sampling to the multiple sub- grating image that the 3 d grid image includes
To sample data;
The embedded unit carries out category attribute label to the sample data using preset label insertion standard as foundation
Insertion, obtain include the corresponding category attribute label of sub- grating image the category attribute table of comparisons.
The third aspect, the present invention provides a kind of computer equipment, the computer equipment includes memory and processing
Device, the memory run the computer program so that the computer is set for storing computer program, the processor
It is standby to execute a kind of classification method of power circuit described in first aspect present invention.
Fourth aspect, the present invention provides a kind of computer readable storage mediums, are stored with third aspect present invention institute
Computer program used in the computer equipment stated.
The classification method and device of the power circuit provided according to the present invention, can preferentially obtain including power line data and
The power line data of shaft tower data, and the power line data is handled, obtain 3 d grid image;Meanwhile it obtaining
The category attribute table of comparisons, in order to which the sorter is according to both above-mentioned category attribute table of comparisons and the neural network of artificial intelligence
Above-mentioned 3 d grid image is analyzed, corresponding category attribute data set is obtained;Last power combining circuit data and institute
Category attribute data set is stated, the corresponding total classification data of power line data is obtained.As it can be seen that implementing this embodiment, energy
It is enough that power line data is handled to obtain corresponding 3 d grid image and obtains the above-mentioned category attribute table of comparisons, so that
This method can carry out analysis classification to above-mentioned 3 d grid image by the neural network of artificial intelligence, to obtain power line
The classification data of circuit-switched data.Therefore, this method can be analyzed by the neural network of artificial intelligence to improve power circuit
Nicety of grading and classification effectiveness, and by data automation obtain with processing realize magnanimity data processing.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of the scope of the invention.
Fig. 1 is a kind of flow diagram of the classification method for power circuit that first embodiment of the invention provides;
Fig. 2 is a kind of flow diagram of the classification method for power circuit that second embodiment of the invention provides;
Fig. 3 is a kind of structural schematic diagram of the sorter for power circuit that third embodiment of the invention provides.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
For the problems of the prior art, the present invention provides a kind of classification method of power circuit, the power circuit
Classification method can preferentially obtain the power line data including power line data and shaft tower data, and to the power line data
It is handled, obtains 3 d grid image;Meanwhile the category attribute table of comparisons is obtained, in order to which the sorter is according to above-mentioned class
Both neural networks of the other attribute table of comparisons and artificial intelligence analyze above-mentioned 3 d grid image, obtain corresponding classification
Attribute data collection;It is corresponding total to obtain power line data for last power combining circuit data and the category attribute data set
Classification data.As it can be seen that implementing this embodiment, power line data can be handled to obtain corresponding 3 d grid
Image simultaneously obtains the above-mentioned category attribute table of comparisons, to enable the method to the neural network by artificial intelligence to above-mentioned three-dimensional
Grating image carries out analysis classification, to obtain the classification data of power line data.Therefore, this method can pass through artificial intelligence
The neural network of energy is analyzed to improve the nicety of grading of power circuit and classification effectiveness, and is obtained by the automation of data
Take the data processing that magnanimity is realized with processing.It is described below by embodiment.
Wherein, above-mentioned technical method can also be realized using relevant software or hardware, in this present embodiment
No longer add to repeat.For the classification method and device of the power circuit, it is described below by embodiment.
Embodiment 1
Referring to Fig. 1, being a kind of flow diagram of the classification method of power circuit provided in this embodiment, the power line
The classification method on road the following steps are included:
S101, the power line data including power line data and shaft tower data is obtained.
In the present embodiment, power line data may include the distributed data and power transmission of power transmission line in power circuit
All data informations relevant to power transmission such as line specification data.Among these, power transmission line refers in power plant, power transformation
It stands for transmitting the route of electric energy between power consumer, which is the important component in power supply system, is undertaken
Conveying and distribution electric energy task.
In the present embodiment, power transmission line be will become, distribution substation links up with each electric power users or electrical equipment, by electricity
Source (becoming, distribution substation) conveys and distributes the conductor circuit of electric energy to load side (electric power users or electrical equipment).Also, electric power
Transmission line can be distinguish by the height of voltage, i.e. high-tension line and low-voltage circuit, wherein high-tension line refers to 1kV or more
The power transmission line of voltage, low-voltage circuit refer to 1kV power transmission line below.Meanwhile also some by 1kV to 10kV (or 35kV)
Power transmission line be known as medium-voltage line, 35kV is known as high-tension line with the power transmission line up to 110kV (or 220kV), and
Power transmission line more than 220kV (or 330kV) is known as supertension line.
In the present embodiment, power transmission line should have reliable, easy to operate and flexible operation of power supply safety etc. to imitate substantially
Fruit.
In the present embodiment, power transmission line can be divided into overhead transmission line, cable run and indoor route etc. by structure type,
For not being limited in any way in structure type the present embodiment of power transmission line.
In the present embodiment, shaft tower is the supporter for being used to supporting electric power transmission lines in overhead transmission line, and wherein shaft tower is more
It is made of steel or armored concrete, is the main support structure of overhead transmission line.
In the present embodiment, shaft tower data include purposes of the structure of shaft tower, the classification of shaft tower and shaft tower etc. and shaft tower phase
All data informations closed.
As an alternative embodiment, power line data is including that power line data and shaft tower data can also wrap
Include the data of other structures or component in relation to power transmission.
Implement this embodiment, can further improve power line data, so that this method can obtain more
More data (or feature), and then realize the raising to power circuit nicety of grading.
S102, the processing power line data obtain power circuit figure, and according to power circuit figure generation and institute
State the corresponding 3 d grid image including multiple sub- grating images of power line data.
In the present embodiment, power line data can be the point cloud data got by laser radar technique.
In the present embodiment, power line data can be the bitmap data with multiple pixels, and each pixel has
There is its unique gray value, wherein the bitmap originally may be considered grating image, and only the grating image is two-dimensional, ash
Angle value exists as three-dimensional supplement.
Implement this embodiment, the data conversion that can be will acquire continues image subsequent at corresponding image
Processing, to be that data are explicitly more intuitive, while being also beneficial to subsequent processing.
As an alternative embodiment, handling the power line data obtains power circuit figure, and according to described
Power circuit figure generates the step of the 3 d grid image including multiple sub- grating images corresponding with the power line data
Suddenly may include:
Processing power line data obtains power circuit figure, obtains the grey scale pixel value in power line data, uniformly draws
Divide the power circuit figure, and combine the power circuit figure and above-mentioned grey scale pixel value after dividing, so that obtaining includes multiple sub- grid
The 3 d grid image of table images;Plurality of sub- grating image is to mark off above-mentioned the step of being evenly dividing the power circuit figure
Come, in addition the division not cuts power circuit figure open to come, only to the region division of power circuit figure.
Implement this embodiment, the acquisition modes of 3 d grid image can be embodied, so that 3 d grid image exists
Except machine directly generates, it can also be obtained by implementing above-mentioned steps.
In the present embodiment, multiple sub- grating images are to mark off to come from 3 d grid image, it can be understood as multiple
Sub- grating image is a part of 3 d grid image.
In the present embodiment, 3 d grid image is the 3 d grid figure with plane two dimensions and gray value third dimension
Picture, in data handling, the 3 d grid image can carry out respective handling by third-order matrix.
S103, the category attribute table of comparisons including the corresponding category attribute label of sub- grating image is obtained.
In the present embodiment, the category attribute table of comparisons includes more seed grating images and a pair of with more seed grating images one
The category attribute label answered.Wherein, category attribute tags, which can be, artificially marks, and is also possible to machine storage, to this
It is not limited in any way in the present embodiment.
In the present embodiment, the category attribute table of comparisons obtains the corresponding category attribute of sub- grating image for matching.
In the present embodiment, the category attribute table of comparisons (is manually labeled because it is individually processed or machine is individually marked
Note), and there are the advantages such as feature (label) quantity is more, feature (label) content is detailed, convenient for improving the classification essence of subsequent step
Degree.
For example, sub- grating image is a shaft tower, and category attribute is then to illustrate that the sub- grating image is one assorted
The shaft tower of sample has what effect etc..
S104, using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, to 3 d grid image into
Row analysis, obtains category attribute data set corresponding with 3 d grid image.
In the present embodiment, Three dimensional convolution neural network may include input layer, rigid line layer, the first convolutional layer, down-sampled layer,
Second convolutional layer, down-sampled layer and third convolutional layer.
In the present embodiment, the sorter of power circuit obtains all features in the category attribute table of comparisons, and (such as classification belongs to
Property etc.), and all features are input in the input layer of Three dimensional convolution neural network, so that Three dimensional convolution neural network is to all
Feature carries out black box for processing, to obtain largely for comparing 3 d grid image neutron grating image with which kind of feature
Data acquisition system.
In the present embodiment, analysis is carried out to 3 d grid image and refers to and is obtained by artificial intelligence (Three dimensional convolution neural network)
The result got analyzes 3 d grid image.Wherein, it is got by artificial intelligence (Three dimensional convolution neural network)
The result is that a large amount of sub- grating image table of comparisons corresponding with category attribute.
S105, combination of power track data and category attribute data set obtain classification number corresponding with power line data
According to.
In the present embodiment, combination of power track data and category attribute data set can make each in power line data
Part is all superimposed with corresponding category attribute, so that power line data classification is completed, obtains corresponding classification data.
In the classification method of the power circuit described in Fig. 1, the classification method of power circuit, the classification of the power circuit
Method can preferentially obtain the power line data including power line data and shaft tower data, and carry out to the power line data
Processing, obtains 3 d grid image;Meanwhile the category attribute table of comparisons is obtained, in order to which the sorter is according to above-mentioned classification category
Property both the table of comparisons and the neural network of artificial intelligence analyze above-mentioned 3 d grid image, obtain corresponding category attribute
Data set;Last power combining circuit data and the category attribute data set obtain corresponding total point of power line data
Class data.As it can be seen that implementing the classification method of power circuit described in Fig. 1, power line data can be handled to obtain
Corresponding 3 d grid image simultaneously obtains the above-mentioned category attribute table of comparisons, to enable the method to the nerve by artificial intelligence
Network carries out analysis classification to above-mentioned 3 d grid image, to obtain the classification data of power line data.Therefore, this method
It can be analyzed to improve the nicety of grading of power circuit and classification effectiveness, and be passed through by the neural network of artificial intelligence
The automation of data obtains the data processing that magnanimity is realized with processing.
Embodiment 2
Referring to Fig. 2, Fig. 2 is a kind of flow diagram of the classification method of power circuit provided in this embodiment.Such as Fig. 2
It is shown, the classification method of the power circuit the following steps are included:
S201, laser radar is controlled to the power circuit emission detection signal including power line and shaft tower.
In the present embodiment, laser radar be it is a kind of use laser as transmitting light source, using detecting technique means
Active remote sensing equipment.Specifically, laser radar is advanced detection mode of the laser technology in conjunction with modern detecting technique,
The detection mode is made of the part such as emission system, reception system, information processing.Wherein, emission system is various forms of sharp
Light device, such as the Solid State Laser of carbon dioxide laser, neodymium doped yttrium aluminium garnet laser, semiconductor laser and tunable wave length
The composition such as device and optical beam-expanding unit;Reception system uses telescope and various forms of photodetectors, such as photomultiplier transit
The combination such as pipe, semiconductor photo diode, avalanche photodide, infrared and visible light multiunit detector part.Meanwhile laser thunder
Up to two kinds of working methods of pulse or continuous wave are used, detection method can be divided into Mie scattering, Rayleigh according to the principle difference of detection
Scattering, Raman scattering, Brillouin scattering, fluorescence etc..
S202, the echo-signal for receiving power circuit reflection.
In the present embodiment, the echo-signal can be to be obtained after above-mentioned detectable signal is changed, wherein two signals
Originally it is same signal, has only been changed by extraneous interference.
S203, signal processing is carried out to detectable signal and echo-signal, obtains including power line data and shaft tower data
Laser radar point cloud data.
As an alternative embodiment, may include: to detectable signal and echo-signal progress signal processing
The distinction between detectable signal and echo-signal is obtained, and distinction is matched, is obtained corresponding
Data, to complete to handle.
Implement this embodiment, processing mode can be embodied, avoids black box work bring uncertain, to mention
High disposal precision and processing order of accuarcy.
S204, laser radar point cloud data is determined as power line data.
S205, processing power line data obtain power circuit figure.
S206, using preset reference face as foundation, to power circuit figure carry out slicing treatment, obtain sectioning image.
In the present embodiment, using preset reference face as foundation, carrying out slicing treatment to power line data be can be understood as point
Not along being horizontally and vertically sliced to sample data, wherein being horizontally and vertically pre-set
's.
In the present embodiment, datum level can be horizontal plane.
In the present embodiment, sectioning image can be bitmap (pixel map).
S207, sectioning image is divided according to default division mode, obtains multiple sub- grating images, and obtain with it is more
A sub- grating image multiple gray values correspondingly.
In the present embodiment, gray value is the particular community of each pixel in sub- grating image.
S208, combination sectioning image and multiple gray values, obtain 3 d grid image.
Implement this embodiment, third dimension can be superimposed in two dimensional image, obtains three-dimensional data (3 d grid figure
Picture).
S209, multiple sub- grating images that 3 d grid image includes are sampled, obtains sample data.
S210, standard is embedded in as foundation using preset label, the insertion of category attribute label is carried out to sample data, is obtained
The category attribute table of comparisons including the corresponding category attribute label of sub- grating image.
Implementation steps S209~S210 is it is known that the category attribute table of comparisons is the sampled images by 3 d grid image
It gets, it is seen then that category attribute list is originally that part is corresponding with 3 d grid image, and also exactly therefore, two have together
One property so as to improve the efficiency of data control and processing, while can also improve the precision of classification.
S211, using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, to 3 d grid image into
Row analysis, obtains category attribute data set corresponding with 3 d grid image.
S212, combination of power track data and category attribute data set obtain classification number corresponding with power line data
According to.
S213, processing is optimized to classification data with preset searching algorithm, obtains Classified optimization data.
In the present embodiment, preset searching algorithm be can be enumeration, depth-first search, breadth first search,
A* algorithm, backtracking algorithm, the search of Monte Carlo tree, hash function, local search algorithm, Neighborhood-region-search algorithm and change neighborhood are searched
One of rope algorithm scheduling algorithm is a variety of.
As an alternative embodiment, preset searching algorithm is Neighborhood-region-search algorithm.
In the present embodiment, search scale can be reduced, according to asking by optimizing processing to classification data using searching algorithm
The constraint condition of topic carries out beta pruning, using the intermediate solution in search process and avoids the advantage computed repeatedly.
For example, this method can select a part of sample from laser radar point cloud data to be processed, and mark
Power line and shaft tower classification in sample;Respectively along being both horizontally and vertically sliced to sample data, grid map is generated
Picture, obtains the gray value of grid, and records the category attribute of each grid;To pending data along both horizontally and vertically into
Row slice, generates grating image, obtains the gray value of grid;Pending data is analyzed by 3D-CNN algorithm, to every
A grid assigns a category attribute;Corresponding cloud is found according to the horizontal vertical coordinate of grid, by the corresponding classification of grid
Attribute assigns corresponding cloud;Neighborhood search, Optimum Classification result are carried out to the point cloud for being divided into power line and shaft tower classification.
Implement this embodiment, power line and tower classification can be carried out by deep learning algorithm, and with sample
It is abundant, improve its nicety of grading.
In the classification method of the power circuit described in Fig. 2, the classification method of power circuit, the classification of the power circuit
Method can preferentially obtain the power line data including power line data and shaft tower data, and carry out to the power line data
Processing, obtains 3 d grid image;Meanwhile the category attribute table of comparisons is obtained, in order to which the sorter is according to above-mentioned classification category
Property both the table of comparisons and the neural network of artificial intelligence analyze above-mentioned 3 d grid image, obtain corresponding category attribute
Data set;Last power combining circuit data and the category attribute data set obtain corresponding total point of power line data
Class data.As it can be seen that implementing the classification method of power circuit described in Fig. 2, power line data can be handled to obtain
Corresponding 3 d grid image simultaneously obtains the above-mentioned category attribute table of comparisons, to enable the method to the nerve by artificial intelligence
Network carries out analysis classification to above-mentioned 3 d grid image, to obtain the classification data of power line data.Therefore, this method
It can be analyzed to improve the nicety of grading of power circuit and classification effectiveness, and be passed through by the neural network of artificial intelligence
The automation of data obtains the data processing that magnanimity is realized with processing.
Embodiment 3
Referring to Fig. 3, being a kind of apparatus structure schematic diagram of the sorter of power circuit provided in this embodiment.
As shown in figure 3, the sorter of the power circuit includes obtaining module 310, processing module 320, analysis module 330
And composite module 340, wherein
Module 310 is obtained for obtaining the power line data including power line data and shaft tower data;
Processing module 320 obtains power circuit figure for handling power line data, and according to power circuit figure generate with
The corresponding 3 d grid image including multiple sub- grating images of power line data;
It obtains module 310 and is also used to obtain the category attribute control including the corresponding category attribute label of sub- grating image
Table;
Analysis module 330 is used for using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, to three-dimensional
Grating image is analyzed, and category attribute data set corresponding with 3 d grid image is obtained;
Composite module 340 is used for combination of power track data and category attribute data set, obtains and power line data pair
The classification data answered.
As an alternative embodiment, processing module 320 be also used to preset searching algorithm to classification data into
Row optimization processing obtains Classified optimization data.
As an alternative embodiment, obtaining module 310 includes sampling unit 311 and embedded unit 312,
In,
Sampling unit 311 obtains sample number for sampling to multiple sub- grating images that 3 d grid image includes
According to;
Embedded unit 312 carries out the embedding of category attribute label to sample data using preset label insertion standard as foundation
Enter, obtain include the corresponding category attribute label of sub- grating image the category attribute table of comparisons.
As it can be seen that can be handled to obtain to power line data using the sorter of power circuit described in Fig. 3
Corresponding 3 d grid image simultaneously obtains the above-mentioned category attribute table of comparisons, to enable the method to the nerve by artificial intelligence
Network carries out analysis classification to above-mentioned 3 d grid image, to obtain the classification data of power line data.Therefore, this method
It can be analyzed to improve the nicety of grading of power circuit and classification effectiveness, and be passed through by the neural network of artificial intelligence
The automation of data obtains the data processing that magnanimity is realized with processing
In addition, the computer equipment may include smart phone, put down the present invention also provides another computer equipment
Plate computer, vehicle-mounted computer, intelligent wearable device etc..The computer equipment includes memory and processor, and memory can be used for depositing
Store up computer program, processor by running above-mentioned computer program, thus make computer equipment execute the above method or on
State the function of each unit in device.
Memory may include storing program area and storage data area, wherein storing program area can storage program area, at least
Application program needed for one function (such as sound-playing function, image player function etc.) etc.;Storage data area can store root
Created data (such as audio data, phone directory etc.) etc. are used according to computer equipment.In addition, memory may include height
Fast random access memory, can also include nonvolatile memory, a for example, at least disk memory, flush memory device,
Or other volatile solid-state parts.
The present embodiment additionally provides a kind of computer storage medium, for storing calculating used in above-mentioned computer equipment
Machine program.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and structure in attached drawing
Figure shows the system frame in the cards of the device of multiple embodiments according to the present invention, method and computer program product
Structure, function and operation.In this regard, each box in flowchart or block diagram can represent a module, section or code
A part, a part of above-mentioned module, section or code includes one or more for implementing the specified logical function
Executable instruction.It should also be noted that function marked in the box can also be to be different from the implementation as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that in structure chart and/or flow chart
The combination of each box and the box in structure chart and/or flow chart, can function or movement as defined in executing it is dedicated
Hardware based system realize, or can realize using a combination of dedicated hardware and computer instructions.
In addition, each functional module or unit in each embodiment of the present invention can integrate one independence of formation together
Part, be also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
If described function is realized and when sold or used as an independent product in the form of software function module, can
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that the part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, it should
Computer software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
Smart phone, personal computer, server or network equipment etc.) execute the complete of method described by each embodiment of the present invention
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
Content described above, only a specific embodiment of the invention, but protection scope of the present invention is not limited to
In this, anyone skilled in the art in the technical scope disclosed by the present invention, can readily occur in variation or replace
It changes, should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with scope of protection of the claims
Subject to.
Claims (10)
1. a kind of classification method of power circuit characterized by comprising
Obtain the power line data including power line data and shaft tower data;
It handles the power line data and obtains power circuit figure, and generated and the power circuit according to the power circuit figure
The corresponding 3 d grid image including multiple sub- grating images of data;
Obtain the category attribute table of comparisons including the corresponding category attribute label of sub- grating image;
Using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, the 3 d grid image is carried out
Analysis, obtains category attribute data set corresponding with the 3 d grid image;
The power line data and the category attribute data set are combined, classification corresponding with the power line data is obtained
Data.
2. the classification method of power circuit according to claim 1, which is characterized in that the combination power line number
According to the category attribute data set, after the step of obtaining classification data corresponding with the power line data, the side
Method further include:
Processing is optimized to the classification data with preset searching algorithm, obtains Classified optimization data.
3. the classification method of power circuit according to claim 1, which is characterized in that described obtain includes power line data
Include: with the step of power line datas of shaft tower data
Laser radar is controlled to the power circuit emission detection signal including power line and shaft tower;
Receive the echo-signal of the power circuit reflection;
Signal processing is carried out to the detectable signal and the echo-signal, obtains including the sharp of power line data and shaft tower data
Optical radar point cloud data;
The laser radar point cloud data is determined as power line data.
4. the classification method of power circuit according to claim 1, which is characterized in that the processing power line number
According to obtaining including the steps that the 3 d grid image of multiple sub- grating images includes:
It handles the power line data and obtains power circuit figure;
Using preset reference face as foundation, slicing treatment is carried out to the power circuit figure, obtains sectioning image;
The sectioning image is divided according to default division mode, obtains multiple sub- grating images, and obtain with it is described more
A sub- grating image multiple gray values correspondingly;
The sectioning image and the multiple gray value are combined, 3 d grid image is obtained.
5. the classification method of power circuit according to claim 1, which is characterized in that described obtain includes every sub- grid
The step of category attribute table of comparisons of the corresponding category attribute label of image includes:
The multiple sub- grating image that the 3 d grid image includes is sampled, sample data is obtained;
Using preset label insertion standard as foundation, the insertion of category attribute label is carried out to the sample data, including
The category attribute table of comparisons of the corresponding category attribute label of sub- grating image.
6. a kind of sorter of power circuit, which is characterized in that the sorter of the power circuit includes obtaining module, place
Manage module, analysis module and composite module, wherein
The acquisition module is for obtaining the power line data including power line data and shaft tower data;
The processing module obtains power circuit figure for handling the power line data, and raw according to the power circuit figure
At the 3 d grid image including multiple sub- grating images corresponding with the power line data;
The module that obtains is also used to obtain the category attribute table of comparisons including the corresponding category attribute label of sub- grating image;
The analysis module is used for using preset Three dimensional convolution neural network and the category attribute table of comparisons as foundation, to described
3 d grid image is analyzed, and category attribute data set corresponding with the 3 d grid image is obtained;
The composite module obtains and the power line for combining the power line data and the category attribute data set
The corresponding classification data of circuit-switched data.
7. the sorter of power circuit according to claim 6, which is characterized in that
The processing module is also used to optimize processing to the classification data with preset searching algorithm, obtains Classified optimization
Data.
8. the sorter of power circuit according to claim 7, which is characterized in that the acquisition module includes that sampling is single
Member and embedded unit, wherein
The sampling unit obtains sample for sampling to the multiple sub- grating image that the 3 d grid image includes
Notebook data;
The embedded unit carries out the embedding of category attribute label to the sample data using preset label insertion standard as foundation
Enter, obtain include the corresponding category attribute label of sub- grating image the category attribute table of comparisons.
9. a kind of computer equipment, which is characterized in that including memory and processor, the memory is for storing computer
Program, the processor runs the computer program so that the computer equipment executes according to claim 1 to any in 5
A kind of classification method of power circuit described in.
10. a kind of computer readable storage medium, which is characterized in that it is stored in computer equipment as claimed in claim 9
Used computer program.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811043754.XA CN109214457B (en) | 2018-09-07 | 2018-09-07 | Power line classification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811043754.XA CN109214457B (en) | 2018-09-07 | 2018-09-07 | Power line classification method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109214457A true CN109214457A (en) | 2019-01-15 |
CN109214457B CN109214457B (en) | 2021-08-24 |
Family
ID=64986656
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811043754.XA Active CN109214457B (en) | 2018-09-07 | 2018-09-07 | Power line classification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109214457B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110120091A (en) * | 2019-04-28 | 2019-08-13 | 深圳供电局有限公司 | Electric inspection process image pattern production method, device and computer equipment |
CN112364878A (en) * | 2020-09-25 | 2021-02-12 | 江苏师范大学 | Power line classification method based on deep learning under complex background |
CN113238578A (en) * | 2021-05-11 | 2021-08-10 | 上海电力大学 | Routing planning method and system for power tower unmanned aerial vehicle inspection route |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104866840A (en) * | 2015-06-04 | 2015-08-26 | 广东中城规划设计有限公司 | Method for recognizing overhead power transmission line from airborne laser point cloud data |
CN106408011A (en) * | 2016-09-09 | 2017-02-15 | 厦门大学 | Laser scanning three-dimensional point cloud tree automatic classifying method based on deep learning |
GB2545602A (en) * | 2016-09-21 | 2017-06-21 | Univ Oxford Innovation Ltd | A neural network and method of using a neural network to detect objects in an environment |
-
2018
- 2018-09-07 CN CN201811043754.XA patent/CN109214457B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104866840A (en) * | 2015-06-04 | 2015-08-26 | 广东中城规划设计有限公司 | Method for recognizing overhead power transmission line from airborne laser point cloud data |
CN106408011A (en) * | 2016-09-09 | 2017-02-15 | 厦门大学 | Laser scanning three-dimensional point cloud tree automatic classifying method based on deep learning |
GB2545602A (en) * | 2016-09-21 | 2017-06-21 | Univ Oxford Innovation Ltd | A neural network and method of using a neural network to detect objects in an environment |
Non-Patent Citations (1)
Title |
---|
JING HUANG,SUYA YOU: "Point Cloud Labeling using 3D Convolutional Neural Network", 《2016 23RD INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110120091A (en) * | 2019-04-28 | 2019-08-13 | 深圳供电局有限公司 | Electric inspection process image pattern production method, device and computer equipment |
CN112364878A (en) * | 2020-09-25 | 2021-02-12 | 江苏师范大学 | Power line classification method based on deep learning under complex background |
CN113238578A (en) * | 2021-05-11 | 2021-08-10 | 上海电力大学 | Routing planning method and system for power tower unmanned aerial vehicle inspection route |
Also Published As
Publication number | Publication date |
---|---|
CN109214457B (en) | 2021-08-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
LaRue et al. | Structural diversity as a predictor of ecosystem function | |
Amos et al. | Determining the use of Sentinel-2A MSI for wildfire burning & severity detection | |
CN108764012B (en) | Urban road rod-shaped object recognition algorithm based on multi-frame combined vehicle-mounted laser radar data | |
CN109214457A (en) | A kind of classification method and device of power circuit | |
CN104020475B (en) | A kind of line of electric force based on on-board LiDAR data extracts and modeling method | |
Yang et al. | Using MODIS NDVI time series to identify geographic patterns of landslides in vegetated regions | |
Kim et al. | 3D classification of power-line scene from airborne laser scanning data using random forests | |
CN102999927B (en) | Fine partition method of soil pollutant content spatial distribution | |
CN108037514A (en) | One kind carries out screen of trees safety detection method using laser point cloud | |
CN108011367B (en) | Power load characteristic mining method based on depth decision tree algorithm | |
CN104951489A (en) | Meteorological data analyzing and processing method applied to power system | |
Vidal-Vázquez et al. | Multifractal analysis of soil properties along two perpendicular transects | |
Bucksch | Revealing the skeleton from imperfect point clouds | |
Xu et al. | Individual stem detection in residential environments with MLS data | |
Grigillo et al. | Determination of large wood accumulation in a steep forested torrent using laser scanning | |
Reese et al. | Combining point clouds from image matching with SPOT 5 multispectral data for mountain vegetation classification | |
Wang et al. | Improved progressive triangular irregular network densification filtering algorithm for airborne LiDAR data based on a multiscale cylindrical neighborhood | |
Heyns et al. | A multi-resolution approach towards point-based multi-objective geospatial facility location | |
CN116128429A (en) | Digital economic system and method based on blockchain technology | |
CN116009121A (en) | Lightning short-time proximity forecasting method based on deep learning | |
CN109948682B (en) | Laser radar point cloud power line classification method based on normal random sampling distribution | |
Roussel et al. | Classification of high-voltage power line structures in low density ALS data acquired over broad non-urban areas | |
CN107085662A (en) | A kind of method that earthquake intensity is extracted based on nighttime light data | |
Fuchs et al. | Intermittent Upwelling Events Trigger Delayed, Major, and Reproducible Pico‐Nanophytoplankton Responses in Coastal Oligotrophic Waters | |
KR101852129B1 (en) | Analysis system for environment research using environmental geographical information and textmining among big data |
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 | ||
CP03 | Change of name, title or address |
Address after: 100000 room 2301-2308, third floor, building 2, incubator of Dongbeiwang Zhongguancun Software Park, Haidian District, Beijing Patentee after: Beijing Digital Green Earth Technology Co.,Ltd. Address before: 100000 incubator 2a2308, Zhongguancun Software Park, Haidian District, Beijing Patentee before: BEIJING GREENVALLEY TECHNOLOGY Co.,Ltd. |
|
CP03 | Change of name, title or address |