Power transmission line point cloud data classification method based on multi-data fusion

Abstract

The application discloses a power transmission line point cloud data classification method based on multi-data fusion, which is applied to a point cloud data classification device and comprises the following steps: acquiring point cloud data in a power transmission line area and power transmission line remote sensing images; performing point cloud data classification processing on the point cloud data to obtain a first classification result; performing image area classification processing on the power transmission line remote sensing images to obtain a second classification result; and performing fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result. According to the scheme of the embodiment of the application, in the process of power transmission line construction, the point cloud data classification device can improve the efficiency and precision of point cloud data classification through image fusion, realizes automatic classification of point cloud data, and is favorable for subsequent construction of a power transmission line engineering three-dimensional visualization intelligent system.

Description

Technical Field

[0001] This application relates to the field of point cloud data processing technology, and in particular to a method for classifying point cloud data of transmission lines based on multi-data fusion. Background Technology

[0002] With the continuous development of living standards, the rising demand for electricity has led to increasingly complex power transmission lines. This complexity makes it difficult to effectively manage and supervise transmission line construction using traditional manual methods. Therefore, the need for a multi-functional intelligent 3D visualization system for power transmission line projects is becoming more urgent for engineers. The construction of such a system requires the creation of a database; therefore, efficient analysis and classification of the large amounts of scanning data (point cloud data) generated by airborne LiDAR during power transmission line inspections is crucial.

[0003] In related technologies, the main method for classifying the large amount of point cloud data collected is manual classification. This involves manually selecting point cloud data and manually dividing it into categories. This method is labor-intensive and inefficient, making it difficult to support the construction of a multi-functional intelligent system for 3D visualization of power transmission line projects. Summary of the Invention

[0004] This application aims to at least partially address one of the technical problems in related technologies. To this end, this application proposes a method, apparatus, equipment, and storage medium for classifying point cloud data of transmission lines based on multi-data fusion. This method can improve the efficiency and accuracy of point cloud data classification through image fusion, achieving automated classification of point cloud data, which is beneficial for the subsequent construction of a multi-functional intelligent system for 3D visualization of transmission line projects.

[0005] In a first aspect, embodiments of this application provide a method for classifying point cloud data of transmission lines based on multi-data fusion, characterized in that it is applied to a point cloud data classification device, comprising: acquiring point cloud data and remote sensing images of the transmission line within a transmission line area; performing point cloud data classification processing on the point cloud data to obtain a first classification result; performing image region classification processing on the remote sensing images of the transmission line to obtain a second classification result; and performing fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result.

[0006] The point cloud data classification method for transmission lines based on multi-data fusion in this application embodiment has at least the following beneficial effects: A point cloud data classification device acquires point cloud data and remote sensing images of the transmission line within a transmission line area; performs point cloud data classification processing on the point cloud data to obtain a first classification result; performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result; and performs fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result. According to the scheme of this application embodiment, the point cloud data classification device acquires point cloud data and remote sensing images of the transmission line within a transmission line area, then performs point cloud data classification processing on the point cloud data to obtain a first classification result, and performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result. Finally, the first classification result and the second classification result are fused to obtain a point cloud data classification result. That is to say, in the scheme of this application embodiment, during the construction of transmission lines, the point cloud data classification device can improve the efficiency and accuracy of point cloud data classification by fusing images, realizing automated classification of point cloud data, which is beneficial for the subsequent construction of a three-dimensional visualization intelligent system for transmission line projects.

[0007] Optionally, in one embodiment of this application, the step of performing point cloud data classification processing on the point cloud data to obtain a first classification result includes: filtering the point cloud data to obtain ground point cloud data and non-ground point cloud data; inputting the non-ground point cloud data into a preset point cloud data classification model to classify the non-ground point cloud data into pole point cloud data, power line point cloud data, vegetation point cloud data, and building point cloud data; and obtaining the first classification result based on the pole point cloud data, the power line point cloud data, the vegetation point cloud data, and the building point cloud data.

[0008] Optionally, in one embodiment of this application, before inputting the non-ground point cloud data into the preset point cloud data classification model, the method further includes: acquiring pre-saved historical point cloud data and the data categories corresponding to the historical point cloud data, wherein the data categories include poles, power lines, vegetation, and buildings; and training the historical point cloud data and the data categories according to a machine learning algorithm to establish the preset point cloud data classification model.

[0009] Optionally, in one embodiment of this application, the step of performing image region classification processing on the remote sensing image of the transmission line to obtain a second classification result includes: performing supervised classification processing on the remote sensing image of the transmission line to divide the remote sensing image of the transmission line into tower regions, power line regions, vegetation regions and building regions; and obtaining a second classification result based on the tower regions, the power line regions, the vegetation regions and the building regions.

[0010] Optionally, in one embodiment of this application, the step of fusing the first classification result and the second classification result to obtain a point cloud data classification result includes: registering the point cloud data and the remote sensing image of the transmission line; when the point cloud data and the remote sensing image of the transmission line are registered, performing overlay region extraction processing on the first classification result and the second classification result to obtain overlapping regions of towers, power lines, vegetation, and buildings; and obtaining the point cloud data classification result based on the overlapping regions of towers, power lines, vegetation, and buildings.

[0011] Optionally, in one embodiment of this application, before performing point cloud data classification processing on the point cloud data, the method further includes: performing point cloud data preprocessing on the point cloud data, wherein the point cloud data preprocessing is at least one of the following: noise reduction processing and outlier removal processing.

[0012] Optionally, in one embodiment of this application, before performing image region classification processing on the remote sensing image of the transmission line, the method further includes: performing image preprocessing on the remote sensing image of the transmission line to obtain the remote sensing image of the transmission line with enhanced image quality, wherein the image preprocessing is at least one of the following: radiometric correction processing, geometric correction processing, and dehazing processing.

[0013] Secondly, embodiments of this application provide a point cloud data classification device, comprising: a data acquisition module for acquiring point cloud data and remote sensing images of a transmission line area; and a data classification processing module for performing point cloud data classification processing on the point cloud data to obtain a first classification result, and further performing image region classification processing on the remote sensing images of the transmission line to obtain a second classification result, and performing fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result.

[0014] The point cloud data classification device of this application embodiment has at least the following beneficial effects: the point cloud data classification device acquires point cloud data and remote sensing images of the transmission line area; performs point cloud data classification processing on the point cloud data to obtain a first classification result; performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result; and performs fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result. According to the scheme of this application embodiment, the point cloud data classification device acquires point cloud data and remote sensing images of the transmission line area, then performs point cloud data classification processing on the point cloud data to obtain a first classification result, and performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result. Finally, the first classification result and the second classification result are fused to obtain a point cloud data classification result. That is to say, in the scheme of this application embodiment, during the construction of the transmission line, the point cloud data classification device can improve the efficiency and accuracy of point cloud data classification by fusing images, realizing automated classification of point cloud data, which is beneficial to the subsequent construction of a three-dimensional visualization intelligent system for transmission line projects.

[0015] Thirdly, embodiments of this application provide a point cloud data classification device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the point cloud data classification method as described in the first aspect.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the point cloud data classification method as described in the first aspect. Attached Figure Description

[0017] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0018] Figure 1 This is a schematic diagram of a point cloud data classification device provided in an embodiment of this application;

[0019] Figure 2 This is a flowchart illustrating a method for classifying point cloud data of transmission lines based on image quality assessment, provided in an embodiment of this application.

[0020] Figure 3 This application Figure 2 A flowchart illustrating the specific method of step S220;

[0021] Figure 4 This application Figure 2A flowchart illustrating the specific method of step S230;

[0022] Figure 5 This application Figure 2 A flowchart illustrating the specific method of step S240;

[0023] Figure 6 This is a schematic diagram of the structure of a point cloud data classification device based on multi-data fusion provided in one embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] It should be noted that although functional modules are divided in the system diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the system or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0026] With the continuous development of living standards, the rising demand for electricity has led to increasingly complex power transmission lines. This complexity makes it difficult to effectively manage and supervise transmission line construction using traditional manual methods. Therefore, the need for a multi-functional intelligent 3D visualization system for power transmission line projects is becoming more urgent. The construction of such a system requires the creation of a database; therefore, efficient analysis and classification of the large amounts of scanning data (point cloud data) generated by airborne LiDAR during power transmission line inspections is crucial.

[0027] In related technologies, the main method for classifying the large amount of point cloud data collected is manual classification. This involves manually selecting point cloud data and manually dividing it into categories. This method is labor-intensive and inefficient, making it difficult to support the construction of a multi-functional intelligent system for 3D visualization of power transmission line projects.

[0028] Based on this, this application proposes a point cloud data classification method, a point cloud data classification device, a point cloud data classification equipment, and a computer-readable storage medium based on multi-data fusion for transmission line point cloud data classification. During the engineering database establishment phase, the point cloud data classification device can acquire point cloud data and remote sensing images of the transmission line within the transmission line area. Then, it performs point cloud data classification processing on the point cloud data to obtain a first classification result, and performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result. Finally, it fuses the first and second classification results to obtain the final point cloud data classification result. In other words, in the scheme of this application embodiment, during the construction of transmission lines, the point cloud data classification device can improve the efficiency and accuracy of point cloud data classification by fusing images, achieving automated point cloud data classification, which is beneficial for the subsequent construction of a three-dimensional visualization intelligent system for transmission line engineering.

[0029] The embodiments of this application will be further described below with reference to the accompanying drawings.

[0030] Reference Figure 1 , Figure 1 This is a schematic diagram of a point cloud data classification device according to an embodiment of this application. The point cloud data classification device includes a data acquisition module 110 and a data classification processing module 120, and the data acquisition module 110 and the data classification processing module 120 are communicatively connected.

[0031] The data acquisition module 110 is used to acquire point cloud data and remote sensing images of the transmission line area. Specifically, the data acquisition module can receive point cloud data and remote sensing images of the transmission line area returned by the airborne lidar, and send the point cloud data and remote sensing images of the transmission line to the data classification and processing module 120.

[0032] The data classification and processing module 120 is used to classify point cloud data to obtain a first classification result, and to classify image regions of remote sensing images of transmission lines to obtain a second classification result. The first and second classification results are then fused to obtain the final point cloud data classification result. Specifically, the data classification and processing module 120 receives point cloud data and remote sensing images of transmission lines from the data acquisition module, classifies the point cloud data and remote sensing images of transmission lines respectively to obtain a first classification result and a second classification result, and finally fuses the first and second classification results to obtain the final point cloud data classification result. This fusion of images improves the efficiency and accuracy of point cloud data classification, enabling automated point cloud data classification, which is beneficial for the subsequent construction of a 3D visualization intelligent system for transmission line projects.

[0033] It is understandable that the point cloud data classification device may also include a preprocessing module. This preprocessing is used to preprocess the point cloud data, including at least one of the following: noise reduction and outlier removal. It is also used to preprocess the remote sensing image of the transmission line to obtain an enhanced image quality, wherein the image preprocessing includes at least one of the following: radiometric correction, geometric correction, and dehazing. Preprocessing the point cloud data and the remote sensing image of the transmission line, and then sending the preprocessed point cloud data and the remote sensing image of the transmission line to the data classification processing module 120, helps to improve the efficiency and accuracy of point cloud data classification.

[0034] According to an embodiment of this application, a point cloud data classification device can be used to process large amounts of point cloud data during the engineering database establishment stage, thereby obtaining point cloud data with high classification accuracy. The point cloud data classification device can classify the acquired point cloud data and remote sensing images of the transmission line area separately to obtain a first classification result and a second classification result. The first classification result and the second classification result are then fused to obtain the point cloud data classification result. By fusing images, the efficiency and accuracy of point cloud data classification can be improved, and automated classification of point cloud data can be achieved, which is beneficial for the subsequent construction of a three-dimensional visualization intelligent system for transmission line engineering.

[0035] The schematic diagrams of device functional modules and application scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. Those skilled in the art will understand that, with the evolution of device functional modules and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems. It will be understood by those skilled in the art that... Figure 1 The system functional modules shown do not constitute a limitation on the embodiments of this application. They may include more or fewer modules than shown, or combine certain modules, or have different combinations of modules.

[0036] Reference Figure 2 , Figure 2 This is a flowchart illustrating a method for classifying transmission line point cloud data based on multi-data fusion, according to an embodiment of this application; this method is applied to... Figure 1 The point cloud data classification device shown includes, but is not limited to, steps S210, S220, S230 and S240.

[0037] Step S210: Acquire point cloud data and remote sensing images of the transmission line area;

[0038] Step S220: Perform point cloud data classification processing on the point cloud data to obtain the first classification result;

[0039] Step S230: Perform image region classification processing on the remote sensing image of the transmission line to obtain the second classification result;

[0040] Step S240: The first classification result and the second classification result are fused to obtain the point cloud data classification result.

[0041] According to the scheme of this application embodiment, during the engineering database establishment stage, engineers can process large amounts of point cloud data using a point cloud data classification device to obtain point cloud data with high classification accuracy. Specifically, the point cloud data classification device acquires point cloud data and remote sensing images of the transmission line area, then performs point cloud data classification processing to obtain a first classification result, and performs image region classification processing on the remote sensing images of the transmission line to obtain a second classification result. Finally, the first and second classification results are fused to obtain the point cloud data classification result. During the construction of the transmission line, the point cloud data classification device can improve the efficiency and accuracy of point cloud data classification by fusing images, realizing automated classification of point cloud data, which is beneficial to the subsequent construction of a three-dimensional visualization intelligent system for transmission line engineering. It is understood that the acquired point cloud data and remote sensing images of the transmission line correspond to the same transmission line area and can be stored accordingly.

[0042] Reference Figure 3 , Figure 3 This application Figure 2 A flowchart illustrating the specific method of step S220. Step S220: Perform point cloud data classification processing on the point cloud data to obtain the first classification result, including but not limited to steps S310, S320, and S330.

[0043] Step S310: Filter the point cloud data to obtain ground point cloud data and non-ground point cloud data.

[0044] In this step, the point cloud data classification device filters the point cloud data to obtain ground point cloud data and non-ground point cloud data. This is equivalent to a preliminary classification of the point cloud data, that is, the filtering process initially divides the point cloud data into ground point cloud data and non-ground point cloud data. This is beneficial for subsequent further classification processing in the non-ground point cloud data, that is, distinguishing different ground features (including buildings, vegetation, etc.) in the non-ground point cloud data. The filtering process is beneficial for subsequent generation of digital terrain models or digital elevation models, as well as for subsequent processing such as ground feature identification, ground feature extraction, and 3D reconstruction based on point cloud data.

[0045] Specifically, an interpolation-based filtering method is used to filter the point cloud data. It is understood that the point cloud data filtering method used can also be mathematical morphology filtering, iterative linear least squares interpolation filtering, moving window filtering, terrain slope-based filtering algorithms, etc. This application does not impose specific restrictions on the point cloud data filtering method used, nor does it provide a detailed description of the implementation process of each filtering algorithm. Engineers can choose the filtering algorithm according to the actual situation.

[0046] In one embodiment of this application, before performing point cloud data classification processing, the method further includes: preprocessing the point cloud data, wherein the preprocessing includes at least one of the following: noise reduction processing and outlier removal processing. It is understood that the environment in which power transmission lines are located is relatively complex and affected by complex external factors. The raw laser point cloud data collected by airborne lidar may contain some messy and abnormal laser noise points, exhibiting significant gross errors. Removing outliers and performing noise reduction processing can improve the reliability of the point cloud data.

[0047] Step S320: Input the non-ground point cloud data into the preset point cloud data classification model and classify the non-ground point cloud data into pole point cloud data, power line point cloud data, vegetation point cloud data and building point cloud data.

[0048] In this step, the point cloud data classification device inputs non-ground point cloud data into a preset point cloud data classification model, classifying the non-ground point cloud data into pole / tower point cloud data, power line point cloud data, vegetation point cloud data, and building point cloud data. Using the preset point cloud data classification model to classify the point cloud data improves the efficiency of point cloud data classification compared to manual classification methods. It is understandable that the preset point cloud data classification model needs to be constructed before inputting the non-ground point cloud data.

[0049] In one embodiment of this application, pre-saved historical point cloud data and corresponding data categories are acquired. These data categories include power poles, power lines, vegetation, and buildings. A preset point cloud data classification model is established by training the historical point cloud data and data categories using a machine learning algorithm. Here, the point cloud data consists of three-dimensional coordinate data of multiple points on objects within the power transmission line area, collected by an airborne lidar. These objects include various types of power poles, power lines, buildings, vegetation, and the ground, etc. The three-dimensional coordinate data is based on coordinates in a land coordinate system.

[0050] Specifically, a large amount of historical non-ground point cloud data is acquired in advance, along with the corresponding categories for each point cloud data point in this historical data. The selected categories are: power poles, power lines, vegetation, and buildings. That is, point cloud data points for power poles, power lines, vegetation, and buildings are selected from the historical non-ground point cloud data for sample training. Considering that power poles have different tower types, power lines have different line types, and vegetation and buildings have different types in reality, targeted classification training can be performed on power poles, power lines, vegetation, and buildings according to different types. This allows the established pre-defined point cloud data classification model to more accurately identify different categories of point cloud data, enabling more efficient and accurate point cloud data classification processing.

[0051] After obtaining training samples, features are extracted from them. These extracted features are then associated with their corresponding categories to form a virtual classifier. Historical non-terrestrial point cloud data and the categories corresponding to each point cloud data point are input into the virtual classifier for training, resulting in a pre-defined point cloud data classification model. After each batch of point cloud data undergoes classification processing to obtain a first classification result, the point cloud data and the corresponding first classification result can be used to train a machine learning algorithm, further improving the classification ability of the pre-defined point cloud data classification model.

[0052] It is understandable that the classification of point cloud data can also include other types, such as roads, rivers, and mountains. Those skilled in the art can train a preset point cloud data classification model according to actual needs and continuously updated point cloud data, so that it can distinguish more point cloud data types.

[0053] Step S330: Obtain the first classification result based on the point cloud data of poles, power lines, vegetation, and buildings.

[0054] Through step S330, the point cloud data classification device can obtain the first classification result including pole point cloud data, power line point cloud data, vegetation point cloud data and building point cloud data.

[0055] Reference Figure 4 , Figure 4 This application Figure 2 A flowchart illustrating the specific method of step S230. Step S230: Perform image region classification processing on the remote sensing image of the transmission line to obtain a second classification result, including but not limited to steps S410 and S420.

[0056] Step S410: Perform supervised classification processing on the remote sensing images of transmission lines, dividing the remote sensing images of transmission lines into tower areas, power line areas, vegetation areas, and building areas.

[0057] In this step, the point cloud data classification device performs supervised classification processing on the remote sensing image of the power transmission line, dividing it into tower areas, power line areas, vegetation areas, and building areas. Specifically, the maximum likelihood method is used to classify the remote sensing image of the power transmission line, obtaining the classification results. It is understood that although this application only specifically involves four types—towers, power lines, vegetation, and buildings—in practical applications, deep learning can be used to further differentiate between these types, which will not be elaborated upon here. Furthermore, other supervised classification algorithms can also be used to classify the remote sensing image of the power transmission line, such as the minimum distance method, parallel algorithms, etc. This application does not impose specific limitations on these algorithms, as long as they can achieve the function of classifying the remote sensing image of the power transmission line.

[0058] Step S420: Obtain the second classification result based on the pole area, power line area, vegetation area, and building area.

[0059] Through step S420, the point cloud data classification device can obtain a second classification result including pole area, power line area, vegetation area and building area.

[0060] In one embodiment of this application, before performing image region classification processing on the remote sensing image of the transmission line, the method further includes: image preprocessing of the remote sensing image of the transmission line to obtain an enhanced image quality remote sensing image of the transmission line, wherein the image preprocessing includes at least one of the following: radiometric correction processing, geometric correction processing, and dehazing processing. Specifically, a high-resolution digital camera mounted on an airborne lidar remote sensing platform is used to acquire true-color images of the transmission line area. Combining the color image with point cloud data helps to compensate for the colorless state of the point cloud data and improve the accuracy of the point cloud data classification results. The quality of acquired images is affected by multiple environmental factors. On the one hand, errors generated by the high-resolution digital camera itself and atmospheric radiation can cause radiation distortion, resulting in uneven grayscale in the acquired remote sensing images of power transmission lines, forming stripes and noise. On the other hand, factors such as the flight status of UAVs and the Earth's rotation can cause distortion in the obtained remote sensing images of power transmission lines compared to the actual terrain. Therefore, performing image preprocessing on remote sensing images of power transmission lines, including at least one of radiometric correction, geometric correction, and dehazing, is beneficial to improving image quality and the accuracy of subsequent registration and classification processes.

[0061] Reference Figure 5 , Figure 5 This application Figure 2A flowchart illustrating the specific method of step S240. Step S240: The first classification result and the second classification result are fused to obtain the point cloud data classification result, including but not limited to steps S510, S520 and S530.

[0062] Step S510: Perform registration processing on the point cloud data and the remote sensing image of the transmission line;

[0063] Step S520: With the point cloud data and the remote sensing image of the transmission line registered, the first classification result and the second classification result are processed to extract the superimposed region to obtain the tower overlapping region, the power line overlapping region, the vegetation overlapping region and the building overlapping region.

[0064] Step S530: Obtain point cloud data classification results based on the overlapping areas of poles, power lines, vegetation, and buildings.

[0065] In one embodiment of this application, the point cloud data classification device performs registration processing on point cloud data and remote sensing images of transmission lines. Specifically, the registration processing is coordinate registration. Then, when the point cloud data and remote sensing images of transmission lines are registered, i.e., when it is confirmed that the point cloud data and remote sensing images of transmission lines correspond to the same transmission line area, the first classification result and the second classification result are subjected to superimposed region extraction processing to obtain pole overlapping areas, power line overlapping areas, vegetation overlapping areas, and building overlapping areas. Specifically, based on the overlap between the pole point cloud data in the first classification result and the pole area in the second classification result, the area overlapping between the pole point cloud data and the pole area is determined as the pole overlapping area. Similarly, the area overlapping between the power line point cloud data and the power line area is determined as the power line overlapping area, the area overlapping between the vegetation point cloud data and the vegetation area is determined as the vegetation overlapping area, and the area overlapping between the building point cloud data and the building area is determined as the building overlapping area. Finally, point cloud data classification results were obtained, including overlapping areas of towers, power lines, vegetation, and buildings, completing the point cloud data classification processing through multi-data fusion. While point cloud data boasts high accuracy and reliability, it lacks spectral information, making it difficult to directly obtain semantic information (such as texture, color, etc.) of ground targets. The type of ground objects cannot be directly determined from point cloud data, inevitably leading to errors in classification results. However, fusing classification results from remote sensing images of transmission lines rich in spectral information can improve the accuracy and efficiency of point cloud data classification to a certain extent, enhancing the automation of point cloud data classification and providing support for the subsequent construction of a 3D visualization intelligent system for transmission line projects.

[0066] Reference Figure 6 , Figure 6This is a schematic diagram of the structure of a point cloud data classification device based on multi-data fusion according to an embodiment of this application. The point cloud data classification device 600 based on multi-data fusion according to this embodiment includes one or more processors 610 and a memory 620. Figure 6 The example uses a processor 610 and a memory 620. The processor 610 and the memory 620 can be connected via a bus or other means. Figure 6 Taking the example of a connection between China and Israel via a bus.

[0067] Memory 620, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory 620 may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 620 may optionally include remotely located memories 620 relative to processor 610, which can be connected to the multi-data fusion-based point cloud data classification device 600 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0068] Those skilled in the art will understand that Figure 6 The device structure shown does not constitute a limitation on the point cloud data classification device 600 based on image quality assessment, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0069] The non-transient software program and instructions required to implement the transmission line point cloud data classification method applied to the point cloud data classification device 600 based on image quality assessment in the above embodiments are stored in the memory 620. When executed by the processor 610, the transmission line point cloud data classification method applied to the point cloud data classification device 600 based on image quality assessment in the above embodiments is executed. For example, the above-described method is executed. Figure 2 Method steps S210 to S240, Figure 3 Method steps S310 to S330, Figure 4 Method steps S410 to S420 and Figure 5 Method steps S510 to S530.

[0070] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0071] Furthermore, one embodiment of this application also provides a computer-readable storage medium storing computer-executable instructions that are executed by one or more processors, for example, by one of the processors 610, causing the one or more processors 610 to perform the control method described in the above method embodiment, for example, to perform the above-described control method. Figure 2 Method steps S210 to S240, Figure 3 Method steps S310 to S330, Figure 4 Method steps S410 to S420 and Figure 5 Method steps S510 to S530.

[0072] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

Claims

1. A method for classifying point cloud data of transmission lines based on multi-data fusion, characterized in that, Applications to point cloud data classification devices include: Acquire point cloud data and remote sensing images of the transmission line area; The point cloud data is subjected to point cloud data classification processing to obtain a first classification result; The remote sensing image of the transmission line is subjected to image region classification processing to obtain a second classification result; The first classification result and the second classification result are fused to obtain the point cloud data classification result; The step of performing point cloud data classification processing on the point cloud data to obtain a first classification result includes: The point cloud data is filtered to obtain ground point cloud data and non-ground point cloud data; The non-ground point cloud data is input into a preset point cloud data classification model to classify the non-ground point cloud data into pole point cloud data, power line point cloud data, vegetation point cloud data and building point cloud data. The first classification result is obtained based on the point cloud data of the poles, the point cloud data of the power lines, the point cloud data of the vegetation, and the point cloud data of the buildings; The step of performing image region classification processing on the remote sensing image of the transmission line to obtain a second classification result includes: performing supervised classification processing on the remote sensing image of the transmission line to divide the remote sensing image of the transmission line into tower regions, power line regions, vegetation regions, and building regions; and obtaining a second classification result based on the tower regions, power line regions, vegetation regions, and building regions. The step of fusing the first classification result and the second classification result to obtain the point cloud data classification result includes: registering the point cloud data and the remote sensing image of the transmission line; with the point cloud data and the remote sensing image of the transmission line registered, extracting the overlapping regions of the first classification result and the second classification result to obtain overlapping regions of towers, power lines, vegetation, and buildings; and obtaining the point cloud data classification result based on the overlapping regions of towers, power lines, vegetation, and buildings.

2. The point cloud data classification method according to claim 1, characterized in that, Before inputting the non-ground point cloud data into the preset point cloud data classification model, the method further includes: Acquire pre-saved historical point cloud data and the corresponding data categories of the historical point cloud data, including poles, power lines, vegetation, and buildings; The preset point cloud data classification model is established by training the historical point cloud data and the data categories using machine learning algorithms.

3. The point cloud data classification method according to claim 1, characterized in that, Before performing point cloud data classification processing on the point cloud data, the method further includes: The point cloud data is preprocessed, and the point cloud data preprocessing includes at least one of the following: noise reduction processing and outlier removal processing.

4. The point cloud data classification method according to claim 1, characterized in that, Before performing image region classification processing on the remote sensing image of the transmission line, the method further includes: The remote sensing image of the transmission line is preprocessed to obtain an enhanced remote sensing image of the transmission line, wherein the image preprocessing is at least one of the following: radiometric correction, geometric correction, and dehazing.

5. A point cloud data classification device, characterized in that, include: The data acquisition module is used to acquire point cloud data and remote sensing images of the transmission line area; The data classification and processing module is used to perform point cloud data classification processing on the point cloud data to obtain a first classification result, and is also used to perform image region classification processing on the remote sensing image of the transmission line to obtain a second classification result, and to perform fusion processing on the first classification result and the second classification result to obtain a point cloud data classification result. The step of performing point cloud data classification processing on the point cloud data to obtain a first classification result includes: The point cloud data is filtered to obtain ground point cloud data and non-ground point cloud data; The non-ground point cloud data is input into a preset point cloud data classification model to classify the non-ground point cloud data into pole point cloud data, power line point cloud data, vegetation point cloud data and building point cloud data. The first classification result is obtained based on the point cloud data of the poles, the point cloud data of the power lines, the point cloud data of the vegetation, and the point cloud data of the buildings; The step of performing image region classification processing on the remote sensing image of the transmission line to obtain a second classification result includes: performing supervised classification processing on the remote sensing image of the transmission line to divide the remote sensing image of the transmission line into tower regions, power line regions, vegetation regions, and building regions; and obtaining a second classification result based on the tower regions, power line regions, vegetation regions, and building regions. The step of fusing the first classification result and the second classification result to obtain the point cloud data classification result includes: registering the point cloud data and the remote sensing image of the transmission line; with the point cloud data and the remote sensing image of the transmission line registered, extracting the overlapping regions of the first classification result and the second classification result to obtain overlapping regions of towers, power lines, vegetation, and buildings; and obtaining the point cloud data classification result based on the overlapping regions of towers, power lines, vegetation, and buildings.

6. A point cloud data classification device, characterized in that, The method includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the point cloud data classification method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the point cloud data classification method as described in any one of claims 1 to 4.

Images