CN117975275A - Distribution line pole tower identification method and device, electronic equipment and storage medium - Google Patents
Distribution line pole tower identification method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN117975275A CN117975275A CN202410195969.2A CN202410195969A CN117975275A CN 117975275 A CN117975275 A CN 117975275A CN 202410195969 A CN202410195969 A CN 202410195969A CN 117975275 A CN117975275 A CN 117975275A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- tower
- target
- determining
- processed
- 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 56
- 238000013145 classification model Methods 0.000 claims abstract description 56
- 238000012549 training Methods 0.000 claims description 52
- 238000004590 computer program Methods 0.000 claims description 16
- 230000009467 reduction Effects 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 description 14
- 230000008569 process Effects 0.000 description 12
- 238000001914 filtration Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 5
- 238000012423 maintenance Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for identifying a distribution line pole tower, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring a point cloud image to be processed corresponding to a target distribution line, and determining at least one tower to be identified corresponding to the target distribution line; determining point cloud images to be identified corresponding to the towers to be identified based on the point cloud images to be processed; and determining the tower type corresponding to each tower to be identified based on the point cloud image to be identified and the target tower classification model. Based on the technical scheme, the point cloud image of the tower to be identified is identified through the target tower classification model, the tower type corresponding to the tower to be identified is determined, the tower identification efficiency is improved, and the accuracy of the identification result is ensured.
Description
Technical Field
The present invention relates to the field of power operation and maintenance technologies, and in particular, to a method and apparatus for identifying a distribution line tower, an electronic device, and a storage medium.
Background
With the continuous development of power technology, in order to improve the operation and maintenance efficiency of the distribution line, a digital power grid model corresponding to the distribution line is constructed, and because the power towers are important constituent structures of the distribution line, the reliability and the safety of the distribution line are directly affected, so that the correct identification of the types of towers in the distribution line is important in constructing the digital power grid model.
The traditional pole and tower type identification method is to carry out visual identification on site by operation and maintenance personnel, but with the rapid development of a power network, the operation and maintenance personnel need to identify a large number of pole and towers, and the manual identification scheme is low in processing efficiency and cannot guarantee the accuracy of an identification result.
Disclosure of Invention
The invention provides a method, a device, electronic equipment and a storage medium for identifying a power distribution line tower, which are used for identifying point cloud images of the tower to be identified through a target tower classification model, determining the tower type corresponding to the tower to be identified, improving the tower identification efficiency and ensuring the accuracy of identification results.
According to an aspect of the present invention, there is provided a method for identifying a distribution line tower, the method comprising:
Acquiring a point cloud image to be processed corresponding to a target distribution line, and determining at least one tower to be identified corresponding to the target distribution line;
determining point cloud images to be identified corresponding to the towers to be identified based on the point cloud images to be processed;
And determining the tower type corresponding to each tower to be identified based on the point cloud image to be identified and a target tower classification model, wherein the target tower classification model is a neural network model obtained through pre-training.
According to another aspect of the present invention, there is provided an identification device for a distribution line tower, the device comprising:
The tower to be identified determining module is used for acquiring a point cloud image to be processed corresponding to a target distribution line and determining at least one tower to be identified corresponding to the target distribution line;
The to-be-identified image determining module is used for determining to-be-identified point cloud images corresponding to the towers to be identified based on the to-be-processed point cloud images;
The tower identification module is used for determining tower types corresponding to the towers to be identified based on the point cloud images to be identified and a target tower classification model, wherein the target tower classification model is a neural network model obtained through training in advance.
According to another aspect of the present invention, there is provided an electronic apparatus including:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of identifying a distribution line pole as described in any one of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute the method for identifying a distribution line tower according to any embodiment of the present invention.
According to the technical scheme, the point cloud images to be processed corresponding to the target distribution lines are obtained, at least one tower to be identified corresponding to the target distribution lines is determined, the point cloud images to be identified corresponding to the towers to be identified are determined based on the point cloud images to be processed, the tower types corresponding to the towers to be identified are determined based on the point cloud images to be identified and the target tower classification model, and the target tower classification model is a neural network model trained in advance. Based on the technical scheme, the point cloud image of the tower to be identified is identified through the target tower classification model, the tower type corresponding to the tower to be identified is determined, the tower identification efficiency is improved, and the accuracy of the identification result is ensured.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for identifying a distribution line tower according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for identifying a distribution line tower according to an embodiment of the present invention;
FIG. 3 is a block diagram of a device for identifying a distribution line tower according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
Fig. 1 is a schematic flow chart of a method for identifying a distribution line tower, which is provided by the embodiment of the present invention, where the method may be implemented by an identifying device for a distribution line tower, where the identifying device for a distribution line tower may be implemented in a hardware and/or software form, and the identifying device for a distribution line tower may be configured in an electronic device, where the electronic device may be a terminal device or a server device, etc., where the identifying device for a distribution line tower identifies a tower point cloud image to be identified according to a target tower classification model obtained by training in advance.
As shown in fig. 1, the method includes:
s110, acquiring a point cloud image to be processed corresponding to a target distribution line, and determining at least one tower to be identified corresponding to the target distribution line.
The target distribution line can be a distribution line which needs to be subjected to digital model construction, and can be determined according to requirements. The point cloud image to be processed can be understood as a point cloud image corresponding to the distribution line, which is obtained through laser radar scanning. The tower to be identified may be a tower for which tower type identification is required.
Specifically, the to-be-processed point cloud image corresponding to the target distribution line is obtained, and at least one tower to be identified corresponding to the target distribution line is determined, for example, a line number corresponding to the target distribution line may be determined, the corresponding to-be-processed point cloud image is queried from a preset image database based on the line number, and a tower of a type not standard to the target distribution line is determined and used as the tower to be identified. It should be noted that, point cloud images associated with each distribution line can be acquired in advance through a laser radar, the acquired point cloud images are stored in a preset image database, and when the point cloud images need to be called, the corresponding point cloud images can be queried from the database based on line numbers corresponding to the distribution lines.
On the basis of the technical scheme, before the to-be-processed point cloud image corresponding to the target distribution line is acquired, the method further comprises the following steps: acquiring historical tower point cloud data, and determining a target training data set based on the historical tower point cloud data; and training the tower classification model to be trained based on the target training data set to determine the target tower classification model.
The historical tower point cloud data can be tower point cloud data obtained through historical collection, and the historical tower point cloud data can comprise point cloud images and tower types corresponding to the images. The target training dataset may be understood as a dataset for model training. The tower classification model to be trained can be an initial neural network model which is not trained, and the tower classification model to be trained can be a point cloud network model constructed based on PointNet.
Specifically, before the towers to be identified in the target distribution line are classified, historical tower point cloud data is required to be obtained from a database, a target training data set is determined based on the historical tower point cloud data, and then the target tower classification model is trained and determined according to the target training data set. For example, the collected historical tower point cloud data may be preprocessed to obtain preprocessed historical tower point cloud data, the preprocessing step may include removing noise, filtering, downsampling, coordinate transformation, standardization and the like, the preprocessed point cloud data is labeled, a tower type tag corresponding to the point cloud data may be determined through a manual labeling mode, and then the labeled data set is divided into a training set, a verification set and a test set, and further a target training data set is obtained. It should be noted that, in the training process of the target tower classification model, a deep learning network architecture (such as PointNet, pointNet ++, DGCNN, etc.) specially designed for point cloud data may be used to extract characteristics of the point cloud, a classification model is built according to a selected network structure, a loss function may be set, a cross entropy loss function may be set, a back propagation algorithm is used to update model parameters, further in the training process, model parameters are continuously iterated and optimized through a training set, and performance on a verification set is monitored to prevent overfitting, further in the training process, model superparameters such as learning rate, weight attenuation, batch size, etc. may be adjusted based on performance of the verification set, further indexes such as accuracy, recall rate, F1 value, etc. are calculated after training is completed, and a final target tower classification model is determined according to the index values of the model.
On the basis of the technical scheme, the determining the target training data set based on the historical tower point cloud data comprises the following steps: determining a tower point cloud image to be processed based on the historical tower point cloud data, and determining a noise point cloud area corresponding to the tower point cloud image to be processed; and determining the target training data set based on the noise point cloud area and the to-be-processed tower point cloud image.
The tower point cloud image to be processed can be understood to be the tower point cloud image which needs noise reduction processing. The noise point cloud region may be a noise region contained in the tower point cloud image.
Specifically, the to-be-processed tower point cloud image is determined based on the historical tower point cloud data, the noise point cloud area corresponding to the to-be-processed tower point cloud image is determined, and the target training data set is determined based on the noise point cloud area and the to-be-processed tower point cloud image, for example, one or more tower point cloud images can be selected from the historical tower point cloud data to serve as the to-be-processed tower point cloud image, whether the to-be-processed tower point cloud image comprises the noise point cloud area or not is determined, if the noise point cloud area exists, the noise point cloud area can be deleted from the to-be-processed tower point cloud image, and further the training process of the model is prevented from being influenced by the noise data.
On the basis of the technical scheme, the determining the noise point cloud area corresponding to the to-be-processed tower point cloud image comprises the following steps: dividing the point cloud image of the tower to be processed to obtain at least one point cloud area to be processed corresponding to the point cloud image of the tower to be processed; determining noise similarity of each point cloud area to be processed and typical noise point cloud data, and determining a target point cloud area based on the noise similarity and a preset noise threshold; and determining the noise point cloud area based on the target point cloud area and the point cloud area to be processed.
The point cloud region to be processed can be understood as a point cloud region obtained by dividing the point cloud image to be processed. The representative noise cloud may be pre-stored point cloud data of representative noise, for example, may be representative noise point cloud containing a black cloud occlusion and representative noise point cloud containing a fog occlusion, representative noise point cloud containing a building occlusion, and the like. The noise similarity may be a data similarity between the point cloud region to be processed and a typical noise cloud. The preset noise threshold may be understood as a preset similarity threshold. The target point cloud region may be a noise point that is finally determined.
Specifically, dividing a to-be-processed tower point cloud image to obtain at least one to-be-processed point cloud region corresponding to the to-be-processed tower point cloud image, determining noise similarity of each to-be-processed point cloud region and typical noise point cloud data, determining a target point cloud region based on the noise similarity and a preset noise threshold, and finally determining the noise point cloud region by the target point cloud region and the to-be-processed point cloud region. For example, feature extraction may be performed on the point cloud region to be processed and the typical noise point cloud data, noise similarity between the point cloud region to be processed and the typical noise point cloud data is calculated based on the extracted feature information, the target point cloud region is determined based on the noise similarity and a preset noise threshold, for example, euclidean distance between each point cloud region to be processed and each typical noise point cloud data may be calculated, and the obtained euclidean distance value is used as a final noise similarity value. It should be noted that, for the method for dividing the point cloud image to be processed, the point cloud image to be processed may be divided into a plurality of point cloud areas to be processed with the same size as the typical noise according to the typical noise size of the typical noise cloud, so as to facilitate calculation of the similarity between each point cloud area to be processed and the typical noise cloud.
On the basis of the above technical solution, the determining the target point cloud area based on the noise similarity and the preset noise threshold includes: and taking the current point cloud area to be processed as the target point cloud area under the condition that the noise similarity is larger than the preset noise threshold.
Specifically, it may be determined whether the point cloud area to be processed has sufficient similarity with the typical noise point cloud data by setting one or more thresholds, so as to mark the point cloud area to be processed as a noise area, and it may be understood that if the noise similarity of the current area to be processed is greater than a preset noise threshold, the current area to be processed may be used as a target point cloud area, and a final noise area may be determined based on the target point cloud area. It should be noted that the preset noise threshold may be a value set empirically by a technician.
On the basis of the above technical solution, the determining the noise point cloud area based on the target point cloud area and the point cloud area to be processed includes: determining adjacent point cloud areas corresponding to the target point cloud areas from the point cloud areas to be processed, and determining point cloud distance values between the target point cloud areas and the adjacent point cloud areas; and determining an average point cloud distance value based on the point cloud distance value, and determining a noise point cloud area corresponding to the to-be-processed tower point cloud image based on the average point cloud distance value.
The adjacent point cloud area can be understood as the entire point cloud area adjacent to the target point cloud area. The point cloud distance value may be a distance value between the target point cloud region and each neighboring point cloud region. The average distance value may be a distance average value determined from the point cloud distance values.
Specifically, an adjacent point cloud area corresponding to the target point cloud area is determined from the point cloud area to be processed, point cloud distance values between the target point cloud area and each adjacent point cloud area are determined, further, an average point cloud distance value is determined based on the point cloud distance values, and finally, a noise point cloud area corresponding to the tower point cloud image to be processed is determined based on the average point cloud distance value. The similarity between the point cloud area to be processed and the typical noise point cloud is calculated, and whether the similarity exceeds a preset similarity threshold T is judged; if the point cloud area to be processed is more than the point cloud area to be processed, determining the current point cloud area to be the point P, namely the target point cloud area; for each target point cloud area point P, all adjacent point cloud areas adjacent to the target point cloud area point P, namely K nearest neighbors, are determined, then point cloud distance values between the target point cloud area point P and the K nearest neighbors are calculated, and then an average distance value D_avg is determined according to the point cloud distance values. It should be noted that, after the target point cloud area is determined, the target point cloud center coordinates corresponding to the target point cloud area may be determined, then the neighboring point cloud areas neighboring the target point cloud area are determined, and then the neighboring point cloud center coordinates corresponding to each neighboring point cloud area are determined, and then the point cloud distance value between the target point cloud area and each neighboring point cloud area is determined based on the target point cloud center coordinates and the neighboring point cloud center coordinates, or for each point cloud area, the coordinate values of all points in the area in the three-dimensional space may be calculated, and the three axes x, y, z may be summed, and then the average value (i.e., the sum divided by the point number in the area) on each axis is calculated, and the center coordinate of the point cloud area in the three-dimensional space is obtainedWherein/> Where (x i,yi,zi) is the coordinates of the i-th point in the region and N is the number of points in the region.
On the basis of the above technical solution, the determining the target training data set based on the noise point cloud area and the to-be-processed tower point cloud image includes: carrying out noise reduction treatment on the to-be-treated tower point cloud image based on the noise point cloud area, and taking the treated image as a target sample image; a target tower type corresponding to the target sample image is acquired, and the target training dataset is determined based on the target sample image and the target tower type.
The target sample image may be an image obtained after performing noise reduction processing. The target tower type may be understood as tower label data corresponding to the target sample image, for example, the tower type may include any one of a tangent tower, a corner tower, and a terminal tower.
Specifically, noise reduction processing is performed on a tower point cloud image to be processed based on a noise point cloud area, the processed image is used as a target sample image, a target tower type corresponding to the target sample image is further obtained, and a target training data set is determined based on the target sample image and the target tower type. For example, if the average distance d_avg between the point cloud target area P and K adjacent point cloud areas is greater than the preset distance threshold, recording all the adjacent point cloud areas around the point of the point cloud target area P as noise point cloud areas, recording the current point cloud area P and the adjacent point cloud areas around the point of the point cloud target area P as a set of tower laser point clouds containing noise points, and deleting the noise point cloud areas from the point cloud image to be processed, thereby obtaining the point cloud image of the target point.
S120, determining point cloud images to be identified corresponding to the towers to be identified based on the point cloud images to be processed.
The point cloud image to be identified can be a point cloud image corresponding to each tower to be identified.
Specifically, the to-be-identified point cloud image corresponding to each to-be-identified tower is determined based on the to-be-identified point cloud image, for example, the geographic position information of the tower corresponding to each to-be-identified tower may be obtained, the image acquisition position information corresponding to each to-be-identified point cloud image is determined, further, the to-be-identified point cloud image corresponding to each to-be-identified tower is determined based on the geographic position information of the tower and the image acquisition position information, it is to be noted that in the process of acquiring the point cloud image, the acquisition position of the image may be marked in the acquired point cloud image in the form of a watermark, further, the corresponding acquisition position information may be determined directly from the image in the process of subsequent processing, and after the to-be-identified point cloud image is determined, the to-be-identified point cloud image may be noise-reduced based on a preset noise reduction algorithm to ensure the accuracy of the identification result, and the preset median noise reduction algorithm may be a statistical filtering (such as mean filtering, neighbor filtering), a consistency checking, etc.
S130, determining the tower type corresponding to each tower to be identified based on the point cloud image to be identified and the target tower classification model.
The target pole tower classification model is a neural network model which is obtained through training in advance. The tower type may include any one of a tangent tower, an angle tower, and a terminal tower.
Specifically, the point cloud images to be identified are used as input of a target tower classification model, the target tower model can classify each tower point cloud image to be identified, give out a corresponding tower type prediction result, and further determine the tower type corresponding to each tower to be identified. The target tower classification model is utilized to process the point cloud images to be identified, the specific type of each tower is finally determined, and accurate data support is provided for follow-up operation and maintenance management, fault detection, update and reconstruction and other works of the electric power facilities.
According to the technical scheme, the point cloud images to be processed corresponding to the target distribution lines are obtained, at least one tower to be identified corresponding to the target distribution lines is determined, the point cloud images to be identified corresponding to the towers to be identified are determined based on the point cloud images to be processed, the tower types corresponding to the towers to be identified are determined based on the point cloud images to be identified and the target tower classification model, and the target tower classification model is a neural network model trained in advance. Based on the technical scheme, the point cloud image of the tower to be identified is identified through the target tower classification model, the tower type corresponding to the tower to be identified is determined, the tower identification efficiency is improved, and the accuracy of the identification result is ensured.
Example two
Fig. 2 is a flowchart of a method for identifying a distribution line tower according to an embodiment of the present invention, where the method for identifying a distribution line tower is further optimized based on the above embodiment. The specific implementation manner can be seen in the technical scheme of the embodiment. Wherein, the technical terms identical to or corresponding to the above embodiments are not repeated herein.
As shown in fig. 2, the method of the embodiment of the present invention includes:
Determining a target pole tower classification model: specifically, historical tower point cloud data is obtained from a database, a target training data set is determined based on the historical tower point cloud data, and then a target tower classification model is determined by training the tower classification model to be trained according to the target training data set. The method includes the steps of carrying out noise interference recognition and filtration on historical tower point cloud data to obtain a final target training data set, further training by utilizing PointNet based on the target training data set to obtain a target tower classification model, determining whether point cloud areas containing noise points in each point cloud image to be processed are contained or not through feature extraction and recognition on the tower laser point cloud, if yes, carrying out filtration and deletion on the point cloud areas containing the noise points, and obtaining the final target training data set after noise filtration is completed. Further, dividing the current to-be-processed tower point cloud image into a plurality of to-be-processed point cloud areas, calculating the similarity between the to-be-processed point cloud areas and the typical noise point cloud, and judging whether the similarity exceeds a preset similarity threshold T; if the point cloud area to be processed is more than the point cloud area to be processed, determining the current point cloud area to be the point P, namely the target point cloud area; for each target point cloud area point P, all adjacent point cloud areas adjacent to the target point cloud area point P, namely K nearest neighbors, are determined, then point cloud distance values between the target point cloud area point P and the K nearest neighbors are calculated, and then an average distance value D_avg is determined according to the point cloud distance values. It should be noted that, after the target point cloud area is determined, the target point cloud center coordinates corresponding to the target point cloud area may be determined, then the neighboring point cloud areas neighboring the target point cloud area are determined, and then the neighboring point cloud center coordinates corresponding to each neighboring point cloud area are determined, and then the point cloud distance value between the target point cloud area and each neighboring point cloud area is determined based on the target point cloud center coordinates and the neighboring point cloud center coordinates, or for each point cloud area, the coordinate values of all points in the area in the three-dimensional space may be calculated, and the three axes x, y, z may be summed, and then the average value (i.e., the sum divided by the point number in the area) on each axis is calculated, and the center coordinate of the point cloud area in the three-dimensional space is obtained Wherein/> Where (x i,yi,zi) is the coordinates of the i-th point in the region and N is the number of points in the region. Further, if the average distance d_avg between the point-target point cloud area P and K adjacent point cloud areas is greater than the preset distance threshold, recording all the plurality of adjacent point cloud areas around the point of the target point cloud area P as noise point cloud areas, recording the current target point cloud area P and the plurality of adjacent point cloud areas around the point of the target point cloud area P as a set of tower laser point clouds containing noise points, and deleting the noise point cloud areas from the point cloud image to be processed, thereby obtaining the target point cloud image. And then the target training data set can be determined by the target point cloud image, and training is performed based on the target training data set and PointNet, so that the high-efficiency and accurate pole tower point cloud classification model can be obtained.
According to the technical scheme provided by the embodiment of the invention, the accurate point cloud data can be obtained by carrying out laser scanning on the towers on the distribution network line, so that the identification and analysis of the towers are realized. By dividing the laser point cloud data and identifying noise interference, the point cloud data of the target tower can be accurately extracted, noise points are filtered, and therefore clear tower point cloud data are obtained. Constructing a standard pole tower point cloud data set and a training pole tower point cloud classification model: the point cloud data of the target tower is marked manually, a standard tower point cloud data set is constructed, and the training is carried out by utilizing PointNet in a deep learning algorithm, so that a high-efficiency and accurate tower point cloud classification model can be obtained.
Determining a point cloud image to be identified, and determining the type of a pole tower: specifically, the geographic position information of the towers corresponding to each to-be-identified is obtained, the image acquisition position information corresponding to each to-be-identified point cloud image is determined, the to-be-identified point cloud image corresponding to each to-be-identified tower is determined based on the geographic position information of the towers and the image acquisition position information, it is to be noted that in the process of acquiring the point cloud image, the acquisition position of the image can be marked in the acquired point cloud image in the form of a watermark, the corresponding acquisition position information can be determined directly from the image in the subsequent processing process, and after the to-be-identified point cloud image is determined, the noise reduction processing can be performed on the to-be-identified point cloud image based on a preset noise reduction algorithm to ensure the accuracy of an identification result, and the preset noise reduction algorithm can be a noise reduction algorithm such as a statistical filtering (e.g. a mean value filtering, a median filtering) and a consistency checking. And finally, taking the point cloud images to be identified as the input of a target tower classification model, wherein the target tower model can classify each tower point cloud image to be identified, give out a corresponding tower type prediction result, and further determine the tower type corresponding to each tower to be identified.
According to the technical scheme, the point cloud images to be processed corresponding to the target distribution lines are obtained, at least one tower to be identified corresponding to the target distribution lines is determined, the point cloud images to be identified corresponding to the towers to be identified are determined based on the point cloud images to be processed, the tower types corresponding to the towers to be identified are determined based on the point cloud images to be identified and the target tower classification model, and the target tower classification model is a neural network model trained in advance. Based on the technical scheme, the point cloud image of the tower to be identified is identified through the target tower classification model, the tower type corresponding to the tower to be identified is determined, the tower identification efficiency is improved, and the accuracy of the identification result is ensured.
Example III
Fig. 3 is a block diagram of a device for identifying a tower of a distribution line according to an embodiment of the present invention. As shown in fig. 3, the apparatus includes: a tower to be identified determination module 310, an image to be identified determination module 320, and a tower identification module 330.
The tower to be identified determining module 310 is configured to acquire a point cloud image to be processed corresponding to a target distribution line, and determine at least one tower to be identified corresponding to the target distribution line;
the to-be-identified image determining module 320 is configured to determine to-be-identified point cloud images corresponding to the towers to be identified based on the to-be-processed point cloud images;
The tower identification module 330 is configured to determine a tower type corresponding to each of the towers to be identified based on the point cloud image to be identified and a target tower classification model, where the target tower classification model is a neural network model obtained by training in advance.
On the basis of the technical scheme, the device further comprises: the classification model determining module is used for acquiring historical tower point cloud data before the point cloud image to be processed corresponding to the target power distribution cable is acquired, and determining a target training data set based on the historical tower point cloud data; and training the tower classification model to be trained based on the target training data set to determine the target tower classification model.
On the basis of the technical scheme, the classification model determining module is used for determining a tower point cloud image to be processed based on the historical tower point cloud data and determining a noise point cloud area corresponding to the tower point cloud image to be processed; and determining the target training data set based on the noise point cloud area and the to-be-processed tower point cloud image.
On the basis of the technical scheme, the classification model determining module is used for dividing the point cloud image of the pole tower to be processed to obtain at least one point cloud area to be processed corresponding to the point cloud image of the pole tower to be processed; determining noise similarity of each point cloud area to be processed and typical noise point cloud data, and determining a target point cloud area based on the noise similarity and a preset noise threshold; and determining the noise point cloud area based on the target point cloud area and the point cloud area to be processed.
On the basis of the technical scheme, the classification model determining module is used for taking the current point cloud area to be processed as the target point cloud area under the condition that the noise similarity is larger than the preset noise threshold value.
On the basis of the technical scheme, the classification model determining module is used for determining adjacent point cloud areas corresponding to the target point cloud areas from the point cloud areas to be processed and determining point cloud distance values between the target point cloud areas and the adjacent point cloud areas; and determining an average point cloud distance value based on the point cloud distance value, and determining a noise point cloud area corresponding to the to-be-processed tower point cloud image based on the average point cloud distance value.
On the basis of the technical scheme, the classification model determining module is used for carrying out noise reduction processing on the to-be-processed tower point cloud image based on the noise point cloud area, and taking the processed image as a target sample image; a target tower type corresponding to the target sample image is acquired, and the target training dataset is determined based on the target sample image and the target tower type.
According to the technical scheme, the point cloud images to be processed corresponding to the target distribution lines are obtained, at least one tower to be identified corresponding to the target distribution lines is determined, the point cloud images to be identified corresponding to the towers to be identified are determined based on the point cloud images to be processed, the tower types corresponding to the towers to be identified are determined based on the point cloud images to be identified and the target tower classification model, and the target tower classification model is a neural network model trained in advance. Based on the technical scheme, the point cloud image of the tower to be identified is identified through the target tower classification model, the tower type corresponding to the tower to be identified is determined, the tower identification efficiency is improved, and the accuracy of the identification result is ensured.
The identification device for the distribution line pole tower provided by the embodiment of the invention can execute the identification method for the distribution line pole tower provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Example IV
Fig. 4 shows a schematic diagram of the structure of an electronic device 10 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 4, the electronic device 10 includes at least one processor 11, and a memory, such as a Read Only Memory (ROM) 12, a Random Access Memory (RAM) 13, etc., communicatively connected to the at least one processor 11, in which the memory stores a computer program executable by the at least one processor, and the processor 11 may perform various appropriate actions and processes according to the computer program stored in the Read Only Memory (ROM) 12 or the computer program loaded from the storage unit 18 into the Random Access Memory (RAM) 13. In the RAM13, various programs and data required for the operation of the electronic device 10 may also be stored. The processor 11, the ROM12 and the RAM13 are connected to each other via a bus 14. An input/output (I/O) interface 15 is also connected to bus 14.
Various components in the electronic device 10 are connected to the I/O interface 15, including: an input unit 16 such as a keyboard, a mouse, etc.; an output unit 17 such as various types of displays, speakers, and the like; a storage unit 18 such as a magnetic disk, an optical disk, or the like; and a communication unit 19 such as a network card, modem, wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 11 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the identification of the distribution line towers.
In some embodiments, the method of identifying a distribution line tower may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 10 via the ROM12 and/or the communication unit 19. When the computer program is loaded into RAM13 and executed by processor 11, one or more steps of the above-described method of identifying a distribution line tower may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the method of identifying the distribution line tower in any other suitable manner (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method of identifying a distribution line tower, comprising:
Acquiring a point cloud image to be processed corresponding to a target distribution line, and determining at least one tower to be identified corresponding to the target distribution line;
determining point cloud images to be identified corresponding to the towers to be identified based on the point cloud images to be processed;
And determining the tower type corresponding to each tower to be identified based on the point cloud image to be identified and a target tower classification model, wherein the target tower classification model is a neural network model obtained through pre-training.
2. The method of claim 1, further comprising, prior to the acquiring the cloud of points to be processed corresponding to the target distribution line:
acquiring historical tower point cloud data, and determining a target training data set based on the historical tower point cloud data;
and training the tower classification model to be trained based on the target training data set to determine the target tower classification model.
3. The method of claim 2, wherein the determining a target training data set based on the historical tower point cloud data comprises:
Determining a tower point cloud image to be processed based on the historical tower point cloud data, and determining a noise point cloud area corresponding to the tower point cloud image to be processed;
and determining the target training data set based on the noise point cloud area and the to-be-processed tower point cloud image.
4. A method according to claim 3, wherein said determining a noise point cloud region corresponding to the tower point cloud image to be processed comprises:
Dividing the point cloud image of the tower to be processed to obtain at least one point cloud area to be processed corresponding to the point cloud image of the tower to be processed;
Determining noise similarity of each point cloud area to be processed and typical noise point cloud data, and determining a target point cloud area based on the noise similarity and a preset noise threshold;
and determining the noise point cloud area based on the target point cloud area and the point cloud area to be processed.
5. The method of claim 4, wherein the determining the target point cloud region based on the noise similarity and a preset noise threshold comprises:
And taking the current point cloud area to be processed as the target point cloud area under the condition that the noise similarity is larger than the preset noise threshold.
6. The method of claim 4, wherein the determining the noise point cloud region based on the target point cloud region and the point cloud region to be processed comprises:
Determining adjacent point cloud areas corresponding to the target point cloud areas from the point cloud areas to be processed, and determining point cloud distance values between the target point cloud areas and the adjacent point cloud areas;
And determining an average point cloud distance value based on the point cloud distance value, and determining a noise point cloud area corresponding to the to-be-processed tower point cloud image based on the average point cloud distance value.
7. A method according to claim 3, wherein said determining said target training data set based on said noise point cloud region and said tower point cloud image to be processed comprises:
Carrying out noise reduction treatment on the to-be-treated tower point cloud image based on the noise point cloud area, and taking the treated image as a target sample image;
a target tower type corresponding to the target sample image is acquired, and the target training dataset is determined based on the target sample image and the target tower type.
8. An identification device for a distribution line pole tower, comprising:
The tower to be identified determining module is used for acquiring a point cloud image to be processed corresponding to a target distribution line and determining at least one tower to be identified corresponding to the target distribution line;
The to-be-identified image determining module is used for determining to-be-identified point cloud images corresponding to the towers to be identified based on the to-be-processed point cloud images;
The tower identification module is used for determining tower types corresponding to the towers to be identified based on the point cloud images to be identified and a target tower classification model, wherein the target tower classification model is a neural network model obtained through training in advance.
9. An electronic device, the electronic device comprising:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of identifying a distribution line pole of any one of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the method of identifying a distribution line tower of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410195969.2A CN117975275A (en) | 2024-02-22 | 2024-02-22 | Distribution line pole tower identification method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410195969.2A CN117975275A (en) | 2024-02-22 | 2024-02-22 | Distribution line pole tower identification method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117975275A true CN117975275A (en) | 2024-05-03 |
Family
ID=90864256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410195969.2A Pending CN117975275A (en) | 2024-02-22 | 2024-02-22 | Distribution line pole tower identification method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117975275A (en) |
-
2024
- 2024-02-22 CN CN202410195969.2A patent/CN117975275A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113705425A (en) | Training method of living body detection model, and method, device and equipment for living body detection | |
| CN113947188A (en) | Training method of target detection network and vehicle detection method | |
| CN114386537A (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN115346171A (en) | Power transmission line monitoring method, device, equipment and storage medium | |
| CN113205041A (en) | Structured information extraction method, device, equipment and storage medium | |
| CN115272262A (en) | Outdoor insulator surface defect identification method and device and electronic equipment | |
| CN114445663A (en) | Method, apparatus and computer program product for detecting challenge samples | |
| CN118038153A (en) | Method, device, equipment and medium for identifying external damage prevention of distribution overhead line | |
| CN116842366A (en) | Abnormality warning method, device, equipment and medium for hydroelectric generating set | |
| CN116957539A (en) | Cable state evaluation method, device, electronic equipment and storage medium | |
| CN115690143B (en) | Image segmentation method, device, electronic equipment and storage medium | |
| CN115249281B (en) | Image occlusion and model training method, device, equipment and storage medium | |
| CN116304798A (en) | Partial discharge type identification method, device, equipment and medium | |
| CN117975275A (en) | Distribution line pole tower identification method and device, electronic equipment and storage medium | |
| CN113344064A (en) | Event processing method and device | |
| CN118052877A (en) | Positioning method, device, equipment and medium for strand breakage defect of power cable | |
| CN113360688B (en) | Method, device and system for constructing information base | |
| CN114092739B (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN117853789A (en) | Image detection method and device | |
| US20230142243A1 (en) | Device environment identification method and apparatus, electronic device, and autonomous vehicle | |
| CN117131990A (en) | Power grid infrastructure information management method and device, electronic equipment and storage medium | |
| CN116630863A (en) | Power distribution network safety identification recognition method, device, equipment and storage medium | |
| CN117115714A (en) | Power grid infrastructure safety management method and device, electronic equipment and storage medium | |
| CN115773821A (en) | Method, device, equipment and medium for detecting thermal fault of dry-type air-core reactor | |
| CN118230094A (en) | Line defect detection model determining method, device, equipment and 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 |