Aircraft model identification method, airport berth guiding method and electronic equipment

Abstract

The invention provides an aircraft model identification method, an airport berth guiding method and electronic equipment, and the method comprises the steps: carrying out the feature extraction of to-be-processed airport three-dimensional point cloud data, and obtaining a plurality of feature points of a target aircraft; determining coordinate positions of the plurality of feature points according to the airport three-dimensional point cloud data; determining distances among the plurality of feature points according to the coordinate positions of the plurality of feature points; and identifying the identification model of the target aircraft from an aircraft model database according to the distance between the plurality of feature points, wherein the aircraft model database stores a corresponding relationship between the distance between the feature points and the identification model of the aircraft. According to the invention, the distances among the plurality of feature points are determined through the coordinate positions of the plurality of feature points determined by the airport three-dimensional point cloud data, and the identification model of the target aircraft can be identified from the aircraft model database according to the distances among the plurality of feature points even if part of the feature points or the areas among the feature points are shielded. And the accuracy of identifying the aircraft model is improved.

Description

Technical Field

[0001] This application relates to the technical fields of air traffic management, computer vision, and 3D perception. Specifically, it relates to an aircraft model identification method, an airport berth guidance method, and an electronic device. Background Art

[0002] Currently, in the process of an aircraft transferring from the flight area to the ground area, the technology for identifying the aircraft model usually relies on two-dimensional images or videos. For example, machine learning technology or deep learning network models are used to perform object detection on two-dimensional images or videos to obtain the identified aircraft model. However, in the specific practice process, it is found that relying on two-dimensional images or videos to identify the aircraft model is easily affected by complex lighting and weather (such as fog, rain, snow). For example, in heavy rain or foggy weather, part of the aircraft's contour will be blocked, resulting in the deep learning network model being unable to clearly extract the contour features of the aircraft. Therefore, the current accuracy of identifying the aircraft model is relatively low. Summary of the Invention

[0003] The purpose of the embodiments of this application is to provide an aircraft model identification method, an airport berth guidance method, and an electronic device, which are used to improve the problem of relatively low accuracy in identifying the aircraft model.

[0004] The embodiments of this application provide an aircraft model identification method, including: extracting features from the airport three-dimensional point cloud data to be processed to obtain multiple feature points of the target aircraft; determining the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data; determining the distances between the multiple feature points according to the coordinate positions of the multiple feature points; and identifying the identification model of the target aircraft from the aircraft model database according to the distances between the multiple feature points. The aircraft model database stores the corresponding relationship between the distances between the feature points and the identification models of the aircraft. In the implementation process of the above solution, the distances between multiple feature points are determined by the coordinate positions of the multiple feature points determined from the airport three-dimensional point cloud data. Even when some feature points or the areas between the feature points are blocked, the identification model of the target aircraft can still be identified from the aircraft model database according to the distances between the multiple feature points. The accuracy of identifying the aircraft model with relatively low overall contour features is improved by adopting the key feature point spacing matching method, effectively transforming the traditional overall contour matching problem into a geometric constraint problem of feature points, thereby improving the accuracy of identifying the aircraft model.

[0005] Optionally, in the embodiments of this application, before extracting features from the airport three-dimensional point cloud data to be processed, it further includes: obtaining a preset background model point cloud and a calibration reference model; aligning the calibration reference model with the calibration object model found in the airport three-dimensional point cloud data to obtain the aligned point cloud data; and deleting the preset background model point cloud from the aligned point cloud data.

[0006] In the implementation process of the above solution, by deleting the preset background model point cloud from the aligned point cloud data, the static environment and dynamic targets are fundamentally separated, avoiding misidentification caused by complex backgrounds (such as moving vehicles, temporary equipment) in traditional methods. Especially for the filtering of repetitive backgrounds such as apron markings and fixed facilities, the signal-to-noise ratio of feature extraction has been qualitatively improved, thereby improving the accuracy of identifying aircraft models.

[0007] Optionally, in the embodiments of the present application, the multiple feature points include: the nose tip point, the first engine center point, and the second engine center point; determining the distances between the multiple feature points according to the coordinate positions of the multiple feature points includes: calculating the engine spacing between the first engine center point and the second engine center point according to the coordinate positions of the first engine center point and the second engine center point; calculating the first straight-line distance between the nose tip point and the first engine center point according to the coordinate position of the nose tip point and the coordinate position of the first engine center point; calculating the second straight-line distance between the nose tip point and the second engine center point according to the coordinate position of the nose tip point and the coordinate position of the second engine center point.

[0008] In the implementation process of the above solution, through the triangular space structure formed by the nose and the twin engines, the system establishes a more stable geometric constraint system than single-point measurement. Even if there is an error in the detection of a certain feature point (such as the nose point deviating by 0.5 meters), the triangular inequality formed by the engine spacing and the straight-line distances on both sides will automatically correct the outliers, forming an error self-correction mechanism. This fault tolerance ability based on geometric topology enables the system to still maintain reliable identification when part of the point cloud is missing or there is noise interference, thereby improving the accuracy of identifying aircraft models.

[0009] Optionally, in the embodiments of the present application, the multiple feature points further include: a ground reference point; determining the distances between the multiple feature points according to the coordinate positions of the multiple feature points further includes: calculating the height of the nose tip point from the ground according to the coordinate positions of the ground reference point and the nose tip point; and / or calculating the height of the first engine center point from the ground according to the coordinate positions of the ground reference point and the first engine center point; and / or calculating the height of the second engine center point from the ground according to the coordinate positions of the ground reference point and the second engine center point.

[0010] In the implementation process of the above solution, by incorporating the nose and engine heights into the feature system, the system constructs a three-dimensional recognition network. The engine spacing in the horizontal direction and the height measurement in the vertical direction form a three-dimensional verification framework, upgrading the recognition process from planar geometry to spatial geometry analysis. This full-dimensional constraint enables the system to still complete reliable recognition through the spatial projection relationship between the nose height and the remaining engine even in extreme situations (such as complete occlusion of one side engine), thereby improving the accuracy of identifying aircraft models.

[0011] Optionally, in the embodiments of the present application, identifying the identification model of the target aircraft from the aircraft model database according to the distances between multiple feature points includes: matching multiple candidate aircraft models in the aircraft model database according to the distances between multiple feature points; calculating the error value between the preset distance of the candidate aircraft model and the distances between multiple feature points; and determining the candidate aircraft model with the smallest error value among the multiple candidate aircraft models as the identification model of the target aircraft.

[0012] In the implementation process of the above solution, through parallel matching of multiple candidate models and global error optimization, the electronic device is enabled to resist local feature distortion. When there are deviations in the measurement of some feature points (such as the distance calculation error caused by the temporary occlusion of one side of the engine), the constraint network composed of other feature distances will automatically balance the influence of outliers, ensuring that the final recognition result is not interfered by single-point failures. This distributed decision-making mechanism has inherently stronger fault tolerance than single-path matching, thereby improving the accuracy of identifying the aircraft model.

[0013] Optionally, in the embodiments of the present application, after identifying the identification model of the target aircraft from the aircraft model database according to the distances between multiple feature points, it further includes: querying the target berth information corresponding to the identification model of the target aircraft in the airport berth database; and generating target guidance information according to the target berth information and multiple feature points of the target aircraft, where the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0014] In the implementation process of the above solution, by deeply coupling aircraft model identification and berth guidance, after identifying the aircraft model according to the feature point distance, the accurate berth parameter information corresponding to the aircraft model (such as the position of the stop line, the hovering height of the engine, etc.) is immediately called to form a complete spatial cognition chain of "identification - positioning - guidance". This closed-loop design enables the guidance instruction to fully consider the structural characteristics of a specific aircraft model, avoiding the adaptation error caused by general guidance, thereby improving the accuracy of airport berth guidance.

[0015] Optionally, in the embodiments of the present application, before querying the target berth information corresponding to the identification model of the target aircraft in the airport berth database, it further includes: receiving the current flight model sent by the server; determining the preliminary berth information according to the airport three-dimensional point cloud data and the current flight model; and generating preliminary guidance information according to the current flight model and the preliminary berth information.

[0016] In the implementation process of the above solution, by obtaining the aircraft type information in the flight plan in advance, the system can start the pre-judgment process before the aircraft enters the scanning area. This forward shift in the time dimension enables subsequent actions such as berth resource allocation and ground handling preparation to obtain a key lead, transforming the traditional reactive scheduling into predictive operation. When the actual scanning data matches the planned information, the system is in the verification mode; when there are deviations (such as a temporary change in the aircraft type), an exception handling process is triggered, forming an optimal combination of decision-making flexibility and execution rigidity, thereby improving the robustness of airport berth guidance.

[0017] Optionally, in the embodiment of the present application, the multiple feature points include: the nose tip point of the aircraft; generating target guidance information according to the target berth information and the multiple feature points of the target aircraft, including: determining whether the nose tip point exceeds the stop line in the berth information; if not, generating distance reminder information in the target guidance information, and the distance reminder information includes the distance between the current position and the stop line.

[0018] In the implementation process of the above solution, by establishing a dynamic relationship between the nose tip point of the aircraft as the core spatial reference and the berth stop line, since the nose of the aircraft is the most prominent visual reference in the pilot's field of vision, the guidance information generated based on it naturally conforms to the pilot's cognitive habits. When the distance reminder information shows that "the nose of the aircraft is 5.3 meters away from the stop line", the pilot can intuitively judge the position relationship without complex spatial conversion. This ergonomic design significantly reduces the cognitive load of information interpretation, especially at night or in bad weather conditions, effectively avoiding manipulation delays caused by information conversion, thereby improving the accuracy of airport berth guidance.

[0019] Optionally, in the embodiment of the present application, generating target guidance information according to the target berth information and the multiple feature points of the target aircraft further includes: if the distance between the nose tip point of the aircraft and the reference center line in the target berth information is greater than a preset distance, generating deviation reminder information in the target guidance information, and the deviation reminder information includes the relative direction between the nose tip point of the aircraft and the reference center line.

[0020] In the implementation process of the above solution, by converting the lateral deviation between the nose position of the aircraft and the reference center line into a directional reminder (such as "0.5 meters to the left"), the system converts the abstract coordinate data into azimuth information that the pilot instinctively understands. This design that conforms to the basic law of spatial cognition enables the pilot to directly perceive the deviation direction without mental coordinate conversion, and can trigger a conditioned reflex correction operation in an emergency, significantly shortening the decision-making chain from perception to action.

[0021] Optionally, in the embodiments of the present application, generating target guidance information according to the target berth information and multiple feature points of the target aircraft includes: determining whether the current speed of the target aircraft is greater than the speed threshold; if so, generating overspeed reminder information in the target guidance information, otherwise, generating deceleration reminder information in the target guidance information, and the overspeed reminder information and the deceleration reminder information include the current speed of the target aircraft.

[0022] In the implementation process of the above solution, by real-time monitoring the aircraft speed and comparing it with the speed threshold, the system can timely determine whether the aircraft speed is appropriate and generate corresponding reminder information, which helps the pilot to more precisely control the aircraft speed and avoid inaccurate berthing caused by too fast or too slow speed. This overspeed reminder information enables the pilot to take deceleration measures in time to prevent the aircraft from rushing past the berth due to too fast speed and reduce the risk of collision with surrounding facilities or other aircraft. The deceleration reminder information helps to avoid too long berthing time caused by too slow aircraft speed and improve the efficiency of aircraft berthing at the airport.

[0023] The embodiments of the present application further provide an airport berth guidance method, including: extracting features from the to-be-processed airport three-dimensional point cloud data to obtain multiple feature points of the target aircraft; determining the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data; determining the distances between the multiple feature points according to the coordinate positions of the multiple feature points; identifying the identification model of the target aircraft from the aircraft type database according to the distances between the multiple feature points, and the corresponding relationship between the distances between the feature points and the identification model of the aircraft is stored in the aircraft type database; querying the target berth information corresponding to the identification model of the target aircraft in the airport berth database; generating target guidance information according to the target berth information and the multiple feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0024] In the implementation process of the above solution, by generating target guidance information according to the target berth information and the multiple feature points of the target aircraft, this guidance information is specific to a particular berth of a particular aircraft, can fully consider the size and shape of the aircraft as well as the specific location and space limitations of the berth, thus providing more personalized and accurate guidance to help the aircraft enter and park at the airport berth more smoothly. Further, since the guidance information is generated based on real-time three-dimensional point cloud data and can timely reflect the current states of the aircraft and the berth, the system can adjust the guidance information in real time according to the actual position and attitude of the aircraft to ensure the accuracy and timeliness of the guidance and improve the success rate of aircraft berthing.

[0025] The embodiment of the present application further provides an aircraft model recognition device, including: a point cloud feature extraction module, configured to extract features from the airport three-dimensional point cloud data to be processed to obtain multiple feature points of the target aircraft; a coordinate position determination module, configured to determine the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data; a feature distance determination module, configured to determine the distances between the multiple feature points according to the coordinate positions of the multiple feature points; an aircraft model recognition module, configured to identify the identification model of the target aircraft from the aircraft model database according to the distances between the multiple feature points, and the corresponding relationship between the distances between the feature points and the identification models of the aircraft is stored in the aircraft model database.

[0026] Optionally, in the embodiment of the present application, the aircraft model recognition device further includes: a point cloud model acquisition module, configured to acquire a preset background model point cloud and a calibration reference model; a point cloud model alignment module, configured to align the calibration reference model with the calibration object model found in the airport three-dimensional point cloud data to obtain the aligned point cloud data; a background model deletion module, configured to delete the preset background model point cloud from the aligned point cloud data.

[0027] Optionally, in the embodiment of the present application, the multiple feature points include: the nose tip point of the aircraft, the center point of the first engine, and the center point of the second engine; the feature distance determination module includes: an engine spacing calculation sub-module, configured to calculate the engine spacing between the center point of the first engine and the center point of the second engine according to the coordinate positions of the center point of the first engine and the center point of the second engine; a first distance calculation sub-module, configured to calculate the first straight-line distance between the nose tip point of the aircraft and the center point of the first engine according to the coordinate positions of the nose tip point of the aircraft and the center point of the first engine; a second distance calculation sub-module, configured to calculate the second straight-line distance between the nose tip point of the aircraft and the center point of the second engine according to the coordinate positions of the nose tip point of the aircraft and the center point of the second engine.

[0028] Optionally, in the embodiment of the present application, the multiple feature points further include: a ground reference point; the feature distance determination module further includes: a ground clearance calculation sub-module, configured to calculate the ground clearance of the nose tip point of the aircraft according to the coordinate positions of the ground reference point and the nose tip point of the aircraft; and / or, a first height calculation sub-module, configured to calculate the ground clearance of the center point of the first engine according to the coordinate positions of the ground reference point and the center point of the first engine; and / or, a second height calculation sub-module, configured to calculate the ground clearance of the center point of the second engine according to the coordinate positions of the ground reference point and the center point of the second engine.

[0029] Optionally, in the embodiments of the present application, the aircraft model recognition module includes: an aircraft model matching sub-module, configured to match multiple candidate aircraft models in the aircraft model database according to the distances between multiple feature points; a feature error calculation sub-module, configured to calculate the error value between the preset distance of the candidate aircraft model and the distances between the multiple feature points; and an aircraft model determination sub-module, configured to determine the candidate aircraft model with the smallest error value among the multiple candidate aircraft models as the identification model of the target aircraft.

[0030] Optionally, in the embodiments of the present application, the aircraft model recognition device further includes: a berth information determination module, configured to query the target berth information corresponding to the identification model of the target aircraft in the airport berth database; and a guidance information generation module, configured to generate target guidance information according to the target berth information and the multiple feature points of the target aircraft, where the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0031] Optionally, in the embodiments of the present application, the aircraft model recognition device further includes: a flight model receiving module, configured to receive the current flight model sent by the server; a preliminary berth determination module, configured to determine preliminary berth information according to the airport three-dimensional point cloud data and the current flight model; and a preliminary guidance determination module, configured to generate preliminary guidance information according to the current flight model and the preliminary berth information.

[0032] Optionally, in the embodiments of the present application, the multiple feature points include: the nose tip point of the aircraft; the guidance information generation module includes: a nose stop judgment sub-module, configured to judge whether the nose tip point exceeds the stop line in the berth information; and a distance reminder guidance sub-module, configured to generate the distance reminder information in the target guidance information if the nose tip point does not exceed the stop line in the berth information, where the distance reminder information includes the distance between the current position and the stop line.

[0033] Optionally, in the embodiments of the present application, the guidance information generation module further includes: a deviation reminder guidance sub-module, configured to generate the deviation reminder information in the target guidance information if the distance between the nose tip point and the reference center line in the target berth information is greater than the preset distance, where the deviation reminder information includes the relative direction between the nose tip point and the reference center line.

[0034] Optionally, in the embodiments of the present application, the guidance information generation module further includes: an aircraft speed judgment sub-module, configured to judge whether the current speed of the target aircraft is greater than the speed threshold; and a speed reminder guidance sub-module, configured to generate the overspeed reminder information in the target guidance information if the current speed of the target aircraft is greater than the speed threshold, otherwise, generate the deceleration reminder information in the target guidance information, where the overspeed reminder information and the deceleration reminder information include the current speed of the target aircraft.

[0035] The embodiment of the present application further provides an airport berth guiding device, including: a point cloud feature extraction module, configured to extract features from the airport three-dimensional point cloud data to be processed to obtain multiple feature points of a target aircraft; a coordinate position determination module, configured to determine the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data; a feature distance determination module, configured to determine the distances between the multiple feature points according to the coordinate positions of the multiple feature points; an aircraft model identification module, configured to identify the identification model of the target aircraft from an aircraft model database according to the distances between the multiple feature points, and the corresponding relationship between the distances between the feature points and the identification models of the aircraft is stored in the aircraft model database; a berth information determination module, configured to query the target berth information corresponding to the identification model of the target aircraft in the airport berth database; a guiding information generation module, configured to generate target guiding information according to the target berth information and the multiple feature points of the target aircraft, and the target guiding information is used to guide the target aircraft to enter and park at the airport berth.

[0036] The embodiment of the present application further provides an electronic device, including: a processor and a memory, the memory stores machine-readable instructions executable by the processor, and when the machine-readable instructions are run by the processor, the method described above is executed.

[0037] The embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is run by a processor, the method described above is executed.

[0038] The embodiment of the present application further provides a computer program product, including: a computer program or computer instructions, and when the computer program or computer instructions are run by a processor, the method described above is executed. Description of the Drawings

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained according to these drawings.

[0040] Figure 1 A front view showing the aircraft turning from the flight area to the ground area when preparing to land; Figure 2 A schematic diagram of a terminal device provided by the embodiment of the present application installed on a vertical pole beside a movable aircraft position; Figure 3 A schematic flowchart of the aircraft model identification method provided by the embodiment of the present application; Figure 4Schematic diagram of the process of guiding an aircraft of model C919 to a ground parking position provided by an embodiment of the present application; Figure 5 Schematic diagram of the feature points and the distances between multiple feature points provided by an embodiment of the present application; Figure 6 Schematic diagram of a terminal device for target guidance information provided by an embodiment of the present application; Figure 7 Schematic diagram of the flow of an airport berth guidance method provided by an embodiment of the present application; Figure 8 Schematic diagram of the structure of an aircraft model identification device provided by an embodiment of the present application; Figure 9 Schematic diagram of the structure of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. It should be understood that the accompanying drawings in the embodiments of the present application are only for the purposes of illustration and description, and are not used to limit the protection scope of the embodiments of the present application. Additionally, it should be understood that the schematic drawings are not drawn to actual scale. The flowcharts used in the embodiments of the present application illustrate the operations implemented according to some embodiments of the embodiments of the present application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without a logical context relationship may be reversed in order or implemented simultaneously. Moreover, those skilled in the art may add one or more other operations to the flowchart or remove one or more operations from the flowchart under the guidance of the content of the embodiments of the present application.

[0042] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and illustrated in the accompanying drawings here may be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed embodiments of the present application, but merely represents the selected embodiments of the present application.

[0043] It is understandable that the "first" and "second" in the embodiments of the present application are used to distinguish similar objects. Those skilled in the art can understand that the terms such as "first" and "second" do not limit the quantity and execution order, and the terms such as "first" and "second" do not necessarily limit differences. In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after. The term "plural" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups).

[0044] Please refer to Figure 1 The front view of the aircraft shown when preparing to land and transferring from the flight area to the ground area; in the related art, when an aircraft is preparing to land, it needs to transfer from the flight area to the ground area and from air traffic control to ground takeover. In order to achieve a smooth transition from air traffic control to ground takeover, it is necessary to identify the model of the aircraft so as to guide the aircraft to accurately park at the airport berth. Traditional technologies for identifying aircraft models usually rely on two-dimensional images or videos. For example, machine learning technologies or deep learning network models are used to perform object detection on two-dimensional images or videos to obtain the identified aircraft model as auxiliary information for the airport berth guidance system (also known as the airport berthing assistance system). In the specific practice process, it is found that traditional technologies for identifying aircraft models mostly identify through the front view of the aircraft and an angle slightly deviated from the front view (such as an angle deviated by -10% to 10%). This method of relying on two-dimensional images or videos to identify aircraft models is easily affected by complex lighting and weather (such as fog, rain, and snow), resulting in the deep learning network model being unable to clearly extract the contour features of the aircraft, leading to a low accuracy rate for identifying aircraft models. At this time, communication tools are usually used to confirm the aircraft model with the captain or pilot of the aircraft, and then the aircraft model is input into the airport berth guidance system (also known as the airport berthing assistance system).

[0045] To improve the problems described above, in the embodiments of the present application, the distance between multiple feature points is determined based on the coordinate positions of multiple feature points determined from airport three-dimensional point cloud data. Even when some feature points or the area between feature points are occluded, the identification model of the target aircraft can be identified from the aircraft model database based on the distance between multiple feature points. The key feature point spacing matching method is used to improve the low accuracy rate of identifying the overall contour features from images to identify aircraft models, thereby improving the accuracy rate of identifying aircraft models.

[0046] Please refer to Figure 2Schematic diagram of a terminal device installed on a vertical pole beside a movable aircraft stand provided by an embodiment of the present application; in some application scenarios, the aircraft model identification method provided by the embodiment of the present application can be executed by an electronic device, where the electronic device refers to a terminal device or a server with the function of executing a computer program. The terminal device can be installed on the wall of the airport terminal building or on a vertical pole beside a movable aircraft stand. Examples of the above terminal devices include: client devices of the airport parking assistance system, desktop computers, etc. A server refers to a device that provides computing services through a network. Examples of servers include: x86 servers and non-x86 servers. Non-x86 servers include: mainframes, minicomputers, and UNIX servers.

[0047] Please refer to Figure 3 Schematic flowchart of the aircraft model identification method provided by the embodiment of the present application; the main idea of this aircraft model identification method is to determine the distances between multiple feature points based on the coordinate positions of the multiple feature points determined from the airport three-dimensional point cloud data. Even when some feature points or the areas between feature points are occluded, the identification model of the target aircraft can be identified from the aircraft type database based on the distances between the multiple feature points. The implementation manners of the above aircraft model identification method may include: Step S110: Perform feature extraction on the airport three-dimensional point cloud data to be processed to obtain multiple feature points of the target aircraft.

[0048] The airport three-dimensional point cloud data is a set of three-dimensional coordinate points representing the geometric features of the surfaces of objects in the airport scene. The airport three-dimensional point cloud data can be data obtained by performing an omnidirectional scan of the airport area using a three-dimensional lidar. For example, it may include millions of spatial point coordinates generated by scanning airport taxiways, parking stands beside the terminal building, and ground parking stands.

[0049] A feature point is a locally key position point with distinctiveness in the three-dimensional point cloud of the target aircraft. The process of extracting feature points is, for example, to identify and extract points from the airport three-dimensional point cloud data that can represent the significant geometric shape features of the key structure of the target aircraft. For example, extract the contour points of the nose, wingtips, and tail of the aircraft, the vertex of the left wingtip, the top point of the vertical tail, the landing gear connection points, etc.

[0050] Please refer to Figure 4Schematic diagram of the process of guiding an aircraft of model C919 to a ground parking position provided by the embodiments of the present application; the target aircraft is the aircraft whose model needs to be identified in the airport three-dimensional point cloud data. After identifying the model of the aircraft, it can also guide the aircraft to accurately park at the parking position next to the terminal building or the ground parking position. For example, the figure shows the process of guiding an aircraft of model C919 to a ground parking position. It can be understood that the feature points extracted from the airport three-dimensional point cloud data, due to using the geometric features of the three-dimensional point cloud rather than the texture features of the optical image, the system fundamentally gets rid of the dependence on the lighting conditions. The detection beam actively emitted by the lidar is not affected by the intensity of the ambient light, enabling the system to maintain stable recognition performance in complex optical environments such as day and night alternation and backlighting. At the same time, the laser of a specific wavelength has penetrability to atmospheric particles such as rain and fog, overcoming the inherent problem of the traditional vision system failing in bad weather.

[0051] Step S120: Determine the coordinate positions of multiple feature points according to the airport three-dimensional point cloud data.

[0052] The coordinate position of a feature point is the mathematical representation of the feature point in three-dimensional space (usually the X / Y / Z axis coordinate values). For example, based on the coordinate system of the airport three-dimensional point cloud data, the coordinate of the nose feature point can be (120.5m, 35.2m, 8.7m). By replacing the overall contour matching with the distributed detection of discrete feature points, the system has the ability to "infer the whole from the part". Even if some feature points are occluded, the spatial constraint relationship between the remaining feature points can still form an effective recognition basis. This recognition logic based on topological relationships makes the system show reliability beyond the reach of traditional methods in the complex environment of the apron.

[0053] Step S130: Determine the distances between multiple feature points according to the coordinate positions of the multiple feature points.

[0054] The distance between feature points is the distance calculated from the coordinate positions of multiple feature points. There are many types of distances here, such as the Euclidean geometric length, horizontal distance, vertical distance (i.e., height) between two feature points, etc. The three-dimensional space distance measurement adds depth dimension information compared with two-dimensional image analysis, making the subtle differences between similar aircraft models (such as a difference of a few centimeters in the distance between engine pylons) become distinguishable features. This recognition method based on solid geometry essentially has higher discriminability than the image analysis of plane projection.

[0055] Step S140: Identify the identification model of the target aircraft from the aircraft model database according to the distances between multiple feature points. The corresponding relationship between the distances between feature points and the identification models of aircraft is stored in the aircraft model database.

[0056] The identification model of an aircraft is the unique model code assigned by the aircraft manufacturer to distinguish different aircraft models, such as the A320 of Airbus, the B787 of Boeing, and the C919 of Commercial Aircraft Corporation of China, etc.

[0057] The aircraft model database is an associated data set that stores the distances between standard feature points of different aircraft models and the corresponding identification models of the aircraft. For example: in the aircraft model database, it can be recorded that the wingspan of the A320 is 34.1 meters, the wingspan of the B787 is 60.1 meters, the engine spacing of a certain aircraft model is X meters, and the ground clearance of the nose tip of the aircraft is Y meters, etc.

[0058] In the implementation process of the above solution, the distances between multiple feature points are determined by the coordinate positions of multiple feature points determined from the airport three-dimensional point cloud data, converting the recognition problem into a geometric distance matching problem, avoiding the complex deep learning inference process. The simple distance comparison operation enables the algorithm to run in real time on low-power embedded devices while retaining scalability. In addition, the three-dimensional space distance measurement adds depth dimension information compared to two-dimensional image analysis, making the subtle differences between similar aircraft models (such as a difference of a few centimeters in the engine pylon spacing) distinguishable features. Therefore, even when some feature points or the areas between feature points are occluded, the identification model of the target aircraft can be identified from the aircraft model database based on the distances between multiple feature points. The method of matching the key feature point spacings improves the low accuracy of identifying aircraft models by the overall contour features, effectively converting the traditional overall contour matching problem into a geometric constraint problem of feature points, thereby improving the accuracy of identifying aircraft models.

[0059] As an alternative implementation of the above aircraft model identification method, before extracting features from the airport three-dimensional point cloud data to be processed, the preset background model point cloud can also be deleted. This implementation can include: Step S101: Obtain the preset background model point cloud and the calibration reference model.

[0060] The above-mentioned preset background model point cloud is obtained by using a lidar to scan the airport apron from multiple angles during the period when there are no aircraft parked. The dynamic scanning strategy of a 32-line lidar can be used. For example: the horizontal resolution is 0.1°, the vertical resolution is 0.2°, the point cloud density in the near-field area (<100m) is 1000 points / ㎡, and the density in the far-field area (≥100m) is 200 points / ㎡.

[0061] The spatial alignment of the above-mentioned calibration reference model establishes a unified coordinate system benchmark for the entire airport. This benchmark transfer mechanism ensures the comparability of point cloud data collected at different positions and different times, overcomes systematic deviations caused by sensor installation errors, ground settlement, etc., and logically guarantees the long-term stability of key distance measurements such as the nose and engines, enabling the aircraft model database to have continuous cumulative value.

[0062] For example, the implementation of the above step S101 is as follows: After obtaining the background model point cloud by scanning the airport apron from multiple angles using a lidar, statistical outlier removal can be applied to remove dynamic interference points such as birds and temporary obstacles, and a three-dimensional background surface model with normal vectors can be generated through Poisson surface reconstruction, so as to obtain the above-mentioned preset background model point cloud. The above calibration reference model can use a pre-deployed L-shaped calibration target, and the surface of the calibration target can be covered with a diffuse reflection material with a reflectivity of 60% to facilitate the acquisition of the point cloud data of the calibration reference model. The above calibration reference model can be installed at the front end of the center line of the parking bay (coordinate P0), or at the left boundary marking point (coordinate P1, 5 m away from the center line), or at the right boundary marking point (coordinate P2, 5 m away from the center line), and can be installed according to specific requirements.

[0063] Step S102: Align the calibration reference model with the calibration object model found in the airport three-dimensional point cloud data to obtain the aligned point cloud data.

[0064] For example, the implementation of the above step S102 is as follows: First, based on the technology of reflection intensity threshold segmentation, a point cloud region with a reflectivity of 50%-70% is screened from the airport three-dimensional point cloud data. Then, the RANSAC algorithm is used to fit the L-shaped plane, which can meet the following constraint conditions: the included angle between adjacent sides is 90°±0.5° and the side length is 2 m±0.1 m, and the least squares method is used to fit the plane equation to calculate the corner coordinates of the calibration object. Then, the local features of the calibration object and the reference model are coarsely matched by using the FPFH feature descriptor, and the improved ICP algorithm is used for iterative optimization, so as to complete the accurate matching and alignment of the calibration reference model and the calibration object model found in the airport three-dimensional point cloud data, and thus obtain the aligned point cloud data. Optionally, after obtaining the aligned point cloud data, the accuracy can be verified by calculating the residuals of the corner points of the calibration object after registration. If the maximum residual of the corner points of the calibration object after registration > 5 mm, an artificial review process can be triggered.

[0065] Step S103: Delete the preset background model point cloud from the aligned point cloud data.

[0066] For example, the implementation of the above step S103 is as follows: Convert the real-time scanned and aligned point cloud data into the background model coordinate system. Then, establish a voxel grid of 0.1m × 0.1m × 0.1m, and mark the voxels that meet the voxel-level comparison conditions as the background. The point clouds in the areas that have not changed within a preset duration (30 minutes) can be fused into the background model point cloud according to a preset weight (such as 5%) through an incremental learning mechanism. Finally, apply MLS (Moving Least Squares) surface reconstruction to repair the boundary defects caused by occlusion, and use a curvature-based filter to suppress the noise in the processed point cloud data.

[0067] Please refer to Figure 5 the schematic diagram of the distances between the feature points provided by the embodiments of the present application shown in the figure; the multiple feature points of the target aircraft shown in the figure include: three feature points, namely, the nose tip point, the first engine center point, and the second engine center point, and the straight-line distances between any two of the nose tip point, the first engine center point, and the second engine center point. As an alternative implementation of the above step S110, for example: First, preprocess the airport three-dimensional point cloud data to obtain the preprocessed point cloud data. Then, use a geometric feature analysis algorithm to identify the preprocessed point cloud data, and calculate the nose tip point of the aircraft nose cone structure in combination with the curvature based on the identified sharp area of the aircraft head. The preprocessing of the airport three-dimensional point cloud data includes: eliminating the abnormal points generated by environmental interference (such as shuttle buses, ground management personnel, temporary obstacles, etc.) through spatial density analysis and reflection intensity screening, and retaining the effective point clouds related to the aircraft body. Or, separate the ground point clouds using a plane fitting algorithm and only retain the three-dimensional data of the aerial targets (such as aircraft) to ensure that subsequent analysis focuses on the aircraft body. Or perform bilateral filtering on the airport three-dimensional point cloud data to retain the edge features.

[0068] As an alternative implementation of the above step S120, for example: Convert the airport 3D point cloud data into the airport global coordinate system, correct the position offset through pre-calibrated sensor parameters (such as installation height, angle), and obtain the corrected continuous multi-frame point cloud data. The coordinate positions of multiple feature points can be found from the corrected continuous multi-frame point cloud data. Alternatively, a motion prediction model (such as a Kalman filter model) can also be applied to the corrected continuous multi-frame point cloud data, and combined with the real-time motion state of the aircraft (speed, direction) to infer the coordinate positions of the feature points. It can be understood that during the process of correcting the position offset through pre-calibrated sensor parameters, multiple sensor data (such as lidar and vision) can also be fused to correct the position offset. For example, the positioning results of different sensors are dynamically weighted according to the target distance to correct the position offset. In addition, for instantaneous occlusion or noise, historical trajectory interpolation can also be used to fill in the missing coordinates. After obtaining the coordinate positions of multiple feature points, a probability model can also be used to evaluate the credibility of the coordinate positions of each feature point, and low-confidence data (such as transient noise points) can be removed.

[0069] As an alternative implementation of the above step S130, the above multiple feature points may include: the nose tip point of the aircraft, the center point of the first engine, and the center point of the second engine; the implementation of determining the distances between multiple feature points according to the coordinate positions of the multiple feature points may include: Step S131: Calculate the engine spacing between the center point of the first engine and the center point of the second engine according to the coordinate positions of the center point of the first engine and the center point of the second engine.

[0070] The first alternative implementation of the above step S131 is, for example: Assume that the coordinate position of the center point of the first engine is represented as , and the coordinate position of the center point of the second engine is represented as . Calculate the coordinate positions of the center point of the first engine and the center point of the second engine in the plane rectangular coordinate system through the formula to obtain the engine spacing between the center point of the first engine and the center point of the second engine. This implementation ignores the height difference and focuses more on the engine spacing in the horizontal direction. Among them, represents the engine spacing between the center point of the first engine and the center point of the second engine, and are respectively the axis value and axis value in the coordinate position of the center point of the first engine, and are respectively the axis value and axis value in the coordinate position of the center point of the second engine.

[0071] The second alternative implementation of the above step S131 is, for example: Assume that the coordinate position of the center point of the first engine is represented as , and the coordinate position of the center point of the second engine is represented as . By using the formula , calculate the engine spacing between the center point of the first engine and the center point of the second engine in the plane rectangular coordinate system, obtaining the engine spacing between the center point of the first engine and the center point of the second engine. This implementation ignores the height difference and focuses more on the engine spacing in the horizontal direction. Among them, represents the engine spacing between the center point of the first engine and the center point of the second engine, , and are respectively the axis value, axis value, and axis value in the coordinate position of the center point of the first engine, , and are respectively the axis value, axis value, and axis value in the coordinate position of the center point of the second engine.

[0072] Step S132: Calculate the first straight-line distance between the nose tip point and the center point of the first engine according to the coordinate positions of the nose tip point and the center point of the first engine.

[0073] The first implementation of the above step S132 is, for example: Use the formula to calculate the coordinate positions of the nose tip point and the center point of the first engine, obtaining the first straight-line distance between the nose tip point and the center point of the first engine. Among them, represents the first straight-line distance between the nose tip point and the center point of the first engine, , are respectively the axis value, axis value in the coordinate position of the center point of the first engine, , are respectively the axis value, axis value in the coordinate position of the nose tip point.

[0074] The second implementation of the above step S132 is, for example: Use the formula to calculate the coordinate positions of the nose tip point and the center point of the first engine, obtaining the first straight-line distance between the nose tip point and the center point of the first engine. Among them, represents the first straight-line distance between the nose tip point and the center point of the first engine, , and are respectively the axis value, axis value and axis value in the coordinate position of the first engine center point, , and are respectively the axis value, axis value and axis value in the coordinate position of the nose tip point.

[0075] Step S133: Calculate the second straight-line distance between the nose tip point and the second engine center point according to the coordinate positions of the nose tip point and the second engine center point.

[0076] For example, the first implementation manner of the above step S133: Use the formula to calculate the coordinate positions of the nose tip point and the second engine center point, and obtain the second straight-line distance between the nose tip point and the second engine center point. Among them, represents the second straight-line distance between the nose tip point and the second engine center point, , are respectively the axis value, axis value in the coordinate position of the second engine center point, , are respectively the axis value, axis value in the coordinate position of the nose tip point.

[0077] For example, the second implementation manner of the above step S133: In the specific practice process, the following formula can be used to calculate the coordinate positions of the nose tip point and the second engine center point, and obtain the second straight-line distance between the nose tip point and the second engine center point.

[0078] ; Among them, represents the second straight-line distance between the nose tip point and the second engine center point, , and are respectively the axis value, axis value and axis value in the coordinate position of the second engine center point, , and are respectively the axis value, axis value and axis value in the coordinate position of the nose tip point.

[0079] As an alternative implementation of the above step S130, the above-mentioned multiple feature points may further include: a ground reference point; the above-mentioned implementation of determining the distances between multiple feature points based on the coordinate positions of the multiple feature points may further include: Step S134: Calculate the height of the nose tip point above the ground based on the coordinate position of the ground reference point and the coordinate position of the nose tip point of the aircraft.

[0080] For example, the implementation of the above step S134 is as follows: Assume that the coordinate position of the ground reference point is represented as and the coordinate position of the nose tip point of the aircraft is represented as , the formula can be used to calculate the coordinate positions of the ground reference point and the nose tip point of the aircraft to obtain the height of the nose tip point above the ground. Among them, represents the height of the nose tip point above the ground, represents the z-axis coordinate value in the vertical direction in the coordinate position of the nose tip point of the aircraft, represents the vertical direction in the coordinate position of the ground reference point z axis coordinate value.

[0081] And / or, the above-mentioned implementation of determining the distances between multiple feature points based on the coordinate positions of the multiple feature points may further include: Step S135: Calculate the height of the center point of the first engine above the ground based on the coordinate position of the ground reference point and the coordinate position of the center point of the first engine.

[0082] For example, the implementation of the above step S135 is as follows: Assume that the coordinate position of the ground reference point is represented as and the coordinate position of the center point of the first engine is represented as , the formula can be used to calculate the coordinate positions of the ground reference point and the center point of the first engine to obtain the height of the center point of the first engine above the ground. Among them, represents the height of the center point of the first engine above the ground, represents the z-axis coordinate value in the vertical direction in the coordinate position of the center point of the first engine, represents the vertical direction in the coordinate position of the ground reference point z axis coordinate value.

[0083] And / or, the above-mentioned implementation of determining the distances between multiple feature points based on the coordinate positions of the multiple feature points may further include: Step S136: Calculate the height of the center point of the second engine above the ground based on the coordinate position of the ground reference point and the coordinate position of the center point of the second engine.

[0084] For example, the implementation of the above step S136 is as follows: Assume that the coordinate position of the ground reference point is expressed as , and the coordinate position of the center point of the second engine is expressed as . The formula can be used to calculate the height of the center point of the second engine above the ground based on the coordinate positions of the ground reference point and the center point of the second engine. Among them, represents the height of the center point of the second engine above the ground, represents the z-axis coordinate value in the vertical direction in the coordinate position of the center point of the second engine, represents the vertical direction in the coordinate position of the ground reference point z axis coordinate value.

[0085] In the implementation process of the above solution, the real-time monitoring of the above-mentioned height above the ground enables the system to automatically adapt to the attitude changes of the aircraft under different load conditions. When the pitch angle of the aircraft changes due to fuel consumption or cargo trimming, the combined measurement of the height parameter and the horizontal distance can accurately infer the true aircraft model characteristics, overcoming the problem of feature distortion caused by attitude changes in traditional two-dimensional recognition. In addition, through the reference function of the ground reference point, the system effectively eliminates ground interference factors such as uneven road surfaces and tire compression. This height measurement value directly reflects the true spatial position of the aircraft structure, enabling the recognition results of the same aircraft model at different airports and under different road surface conditions to be highly consistent, solving the problem of recognition parameter drift caused by ground effects in traditional methods, thereby improving the accuracy of identifying the aircraft model.

[0086] As an alternative implementation of the above step S140, the implementation of identifying the identification model of the target aircraft from the aircraft model database based on the distances between multiple feature points may include: Step S141: Match multiple candidate aircraft models in the aircraft model database according to the distances between multiple feature points.

[0087] For example, the implementation of the above step S141 is as follows: Search and match in the aircraft model database according to the engine spacing between the center point of the first engine and the center point of the second engine, the first straight-line distance between the nose tip point and the center point of the first engine, and the second straight-line distance between the nose tip point and the center point of the second engine to obtain multiple candidate aircraft models that are matched.

[0088] Step S142: Calculate the error value between the preset distance of the candidate aircraft model and the distances between multiple feature points.

[0089] For example, in the implementation of step S142 above: Considering some minor errors that may exist in actual measurements, a certain error is allowed between the preset distance of the candidate aircraft model and the distances between multiple feature points. At this time, an executable program written in a preset programming language can be used to calculate the error value between the preset distance of the candidate aircraft model and the distances between multiple feature points. Among them, the candidate aircraft model is matched from the aircraft model database, and multiple feature points are determined from the point cloud data. The preset programming language can be C, Java, Python, Lisp, PHP, JavaScript, etc.

[0090] Step S143: Determine the identification model of the target aircraft by selecting the candidate aircraft model with the smallest error value from multiple candidate aircraft models.

[0091] For example, in the implementation of step S143 above: Since the error value between the preset distance of the candidate aircraft model and the distances between multiple feature points is calculated, each candidate aircraft model among multiple candidate aircraft models has an error value. At this time, the candidate aircraft model with the smallest error value can be selected from multiple candidate aircraft models according to the error value. If the error value is the smallest and within the error threshold range, the candidate aircraft model with the smallest error value can be determined as the identification model of the above-mentioned target aircraft. Specifically, taking the ground clearance height of the nose tip of the aircraft as an example for illustration, the other distance error values are similar, and the formula can be used to determine whether the error value of the ground clearance height of the nose tip of the aircraft is within the error threshold range. Among them, represents the ground clearance height of the nose tip of the aircraft, represents the preset distance of the ground clearance height of the nose tip of the aircraft, represents the error threshold.

[0092] In the implementation process of the above solution, the above screening logic of minimizing errors essentially constitutes a dynamic decision threshold. For aircraft models with obvious features (such as wide-body aircraft), the airport berthing system will automatically tighten the error tolerance, while for aircraft models with similar features, a relatively loose matching window will be maintained. This adaptive adjustment of discrimination accuracy realizes the intelligent balance between recognition accuracy and recall rate.

[0093] As an alternative implementation of the above aircraft model recognition method, after identifying the identification model of the target aircraft from the aircraft model database according to the distances between multiple feature points, berthing guidance can also be performed. The implementation of this berthing guidance can include: Step S150: Query the target berthing information corresponding to the identification model of the target aircraft in the airport berthing database.

[0094] It can be understood that the above-mentioned three-dimensional point cloud data of the airport can include real-time spatial information such as airport taxiways, parking positions beside the terminal building, ground parking positions, obstacles, etc., and the identification model of the target aircraft is identified through steps S110 to S140 in the above-mentioned aircraft model identification method.

[0095] Step S160: Generate target guidance information based on the target berth information and multiple feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0096] As an optional implementation manner of the above step S150, for example: It can be understood that since the airport berth database stores the berth information corresponding to each aircraft model, it is possible to query among the multiple berth information in the airport berth database and automatically screen the candidate berth information that is currently free at the airport and meets the size requirements according to the identification model of the target aircraft. Then, perform three-dimensional point cloud analysis on the candidate berth areas in the candidate berth information, such as: detecting the clarity of ground markings (stop lines, guiding lines), scanning surrounding obstacles (such as ground service vehicles, temporary equipment), calculating the path curvature from the berth entrance to the stop line (ensuring that the turning radius ≥ the aircraft model requirements), etc., and comprehensively evaluate the path safety (such as the minimum lateral safety distance), operation efficiency (such as taxiing distance, number of turns), ground service support status (such as fuel truck, gangway preparation status), etc. of each candidate berth information. Finally, output the target berth information corresponding to the identification model of the target aircraft. Among them, the above-mentioned berth information can include berth number, minimum wingspan space, maximum height limit, clearance height, jet bridge docking position, three-dimensional coordinates of the standard stop point, and key nodes of the recommended taxiing path, etc.

[0097] As an alternative implementation of the above aircraft model identification method, before querying the target berth information corresponding to the identification model of the target aircraft in the airport berth database, preliminary guidance can also be performed. The implementation of this preliminary guidance can include: The electronic device can receive the current flight model sent by the server. Here, the current flight model is the identification model of the target aircraft obtained by the server before the target aircraft lands, and the server sends this identification model to the electronic device. It can be understood that if the server fails to successfully send the identification model of the target aircraft to the electronic device due to various reasons (such as network failure or network delay), or the aviation management system changes the model of the target aircraft or schedules other aircraft to perform the current flight task, the electronic device needs to identify the identification model of the aircraft when landing through the airport three-dimensional point cloud data by itself. Then, the electronic device can query the preliminary berth information corresponding to the current flight model in the airport berth database. Here, the implementation is similar to that of step S150 above, so it will not be elaborated. Finally, the electronic device can generate preliminary guidance information based on the current flight model and the preliminary berth information. There are many types of this preliminary guidance information, such as distance guidance information, deviation guidance information, overspeed guidance information, deceleration guidance information, and / or over-line guidance information, etc. The detailed processes of these guidance information will be described in detail below.

[0098] As an alternative implementation of the above step S160, the above-mentioned multiple feature points can include: the nose tip point; the implementation of generating the target guidance information based on the target berth information and the multiple feature points of the target aircraft can include: Step S161: Determine whether the nose tip point exceeds the stop line in the berth information.

[0099] It can be understood that the position of the nose tip point reflects the overall movement trend of the aircraft in advance. When it is detected that the nose is approaching but has not reached the stop line, the system can predict the final stop position based on the current deceleration curve. If there is a risk of overstepping the line (such as brake response delay) in the prediction, a hierarchical warning can be triggered in advance. This control strategy based on kinematic feedforward provides an additional 2 - 3 seconds of reaction time window for the pilot compared to the traditional position feedback control.

[0100] Step S162: If the nose tip point does not exceed the stop line in the berth information, generate the distance reminder information in the target guidance information, and this distance reminder information includes the distance between the current position and the stop line.

[0101] Please refer to Figure 6Schematic diagram of a terminal device for target guidance information provided by an embodiment of the present application; the target guidance information may include: the identification model of the aircraft (the aircraft identification model in the figure is A380), distance reminder information (the distance between the current position of the aircraft and the stop line in the figure is 17.0 meters), deviation reminder information (the deviation reminder mark for turning to the right in the figure), and speed reminder information (the current speed of the aircraft is 3.2 m / s meters per second). The implementation manners of the above steps S161 to S162 are, for example: using an executable program written in a preset programming language to determine whether the nose tip point exceeds the stop line in the berth information. The preset programming language can be C, Java, Python, Lisp, PHP, or JavaScript, etc. When it is detected that the coordinate of the nose tip point exceeds the plane boundary of the stop line, the over-line determination mechanism is immediately triggered. At this time, the system comprehensively calculates parameters such as the current over-line distance, the movement speed of the aircraft, and the heading angle, and generates multi-level warning information including a precise digital distance reminder. The distance reminder information can be presented through a red-yellow dual-color dynamic warning symbol on the terminal guidance display screen, and at the same time, a voice prompt for recommending corrective operations is sent to the pilot's headset. All data is synchronized to the airport ground control system in real time. Among them, the above distance reminder information may include the distance between the current position and the stop line (the distance between the current position of the aircraft and the stop line in the figure is 17.0 meters).

[0102] Optionally, the above airport berth guidance system can control the above distance reminder information to automatically switch the expression dimension according to the spatial relationship between the nose of the aircraft and the stop line: display "remaining distance" at a long distance, switch to "recommended deceleration curve" at a medium distance, and switch to "centimeter-level fine-tuning guidance" at a short distance. This information hierarchical presentation method that conforms to the law of human attention distribution enables the pilot to obtain the best decision-making support at different stages. In addition, it can also control the subtle abnormalities in the movement trajectory of the nose of the aircraft (such as asymmetric offset) to reflect potential failures (such as unilateral brake failure) in advance. By continuously analyzing the deviation pattern between the nose position and the theoretical path, the system can identify steering system abnormalities in the early stage of the ground taxiing phase, realizing the functional leap from simple guidance to health monitoring.

[0103] As an alternative implementation manner of the above step S160, the implementation manner of generating target guidance information according to the target berth information and multiple feature points of the target aircraft may further include: Step S163: Determine whether the distance between the nose tip point of the aircraft and the reference center line in the target berth information is greater than a preset distance.

[0104] Step S164: If the distance between the nose tip point of the aircraft and the reference center line in the target berth information is greater than the preset distance, generate deviation reminder information in the target guidance information. The deviation reminder information includes the relative direction between the nose tip point and the reference center line.

[0105] For example, in the implementation of the above steps S163 to S164: The electronic device obtains the three-dimensional coordinates of the nose tip point of the aircraft through real-time scanning, compares the three-dimensional coordinates of the nose tip point with the reference center line preset for the berth in terms of spatial position, and calculates the lateral distance between the two. When it is detected that the deviation distance between the nose tip point and the reference center line exceeds the preset safety threshold (usually 0.5 - 1.2 meters, automatically adjusted according to the aircraft type), the system immediately generates a visual deviation reminder message. This message contains the specific deviation direction (such as the arrow mark in the figure, or the text prompt "0.8 meters to the left" not shown in the figure), and the color arrow mark and digital distance are displayed through the terminal guidance screen. In addition, a voice prompt can also be sent to the flight pilot (such as "Please correct 1 meter to the right"). It can be understood that the direction information in the above deviation reminder naturally corresponds to the rudder operation logic: "Deviation to the left" directly prompts the correction requirement of "Turn right". This mirror correspondence relationship between the control instruction and the operation action forms a negative feedback closed loop that conforms to human intuition, avoiding the common "reverse operation error" in the traditional guidance system (such as misexecuting the left turn instruction as a right turn), which is especially beneficial for novice pilots to quickly establish the correct spatial control mapping.

[0106] Optionally, the airport berth guidance system on the electronic device can also adopt multi-sensor fusion technology to ensure the accuracy of direction determination, and will dynamically adjust the reminder intensity according to the distance of the aircraft approaching the stop line. The reminder frequency will be increased by 50% in the last 20-meter stage to ensure timely correction. During the reminder process, all deviation data can be recorded in real time and transmitted to the airport operation control center for subsequent operation analysis and service improvement.

[0107] As an alternative implementation of the above step S160, the implementation of generating the target guidance information based on the target berth information and multiple feature points of the target aircraft may include: Step S165: Determine whether the current speed of the target aircraft is greater than the speed threshold.

[0108] Optionally, the above speed threshold can be adjusted according to the performance of different aircraft types and the airport environment to make the system have good adaptability. For example, under complex meteorological conditions or different berth layouts, the system can flexibly adjust the guidance strategy according to the actual situation.

[0109] Step S166: If the current speed of the target aircraft is greater than the speed threshold, generate an overspeed reminder message in the target guidance information, and the overspeed reminder message includes the current speed of the target aircraft.

[0110] It can be understood that the airport berth guidance system can dynamically generate guidance information based on the real-time speed and position information of the aircraft. This real-time nature makes the guidance information always match the current state of the aircraft, improving the accuracy and timeliness of the guidance.

[0111] Step S167: If the current speed of the target aircraft is less than or equal to the speed threshold, generate a deceleration reminder message in the target guidance information, where the deceleration reminder message includes the current speed of the target aircraft.

[0112] It can be understood that the airport berth guidance system in the above-mentioned electronic device can be built with a dynamic speed threshold model, which comprehensively considers multiple factors such as aircraft type parameters (such as the braking performance differences of different aircraft types like A320 / B787, etc.), runway surface conditions (different friction coefficients such as dry / wet / icy, etc.), meteorological environment (headwind / tailwind speed), and the remaining length to the stop line, etc., to generate personalized speed control thresholds for each aircraft.

[0113] The implementation manners of the above Step S165 to Step S167 are as follows: For example, the electronic device can work in cooperation with a high-precision lidar and a millimeter-wave radar. The airport berth guidance system it operates can real-time monitor the three-dimensional motion state of the aircraft during taxiing and approach, accurately calculate the instantaneous velocity vector of the aircraft, and determine whether the current speed of the target aircraft is greater than the dynamic speed threshold through the instantaneous velocity vector. When the system detects that the current speed of the aircraft exceeds the dynamic speed threshold, it immediately triggers a multi-level warning mechanism: First, synchronously display a red flashing overspeed warning sign on the display screen of the electronic device and multiple ground guiding signs, and at the same time intuitively display the current overspeed value in both digital and analog forms (such as "overspeed 0.5m / s"); the system will calculate the recommended safe deceleration curve based on the aircraft kinematic model and real-time position data, and generate specific braking suggestions (such as "it is recommended to immediately reduce the speed to 1.2m / s"); these messages will be synchronously transmitted to the pilot and ground commanders through the airport wireless data link and the voice prompt system. For the situation where the speed is not overspeed but close to the threshold, the system will generate a preventive deceleration reminder, display a yellow slow-down prompt on the guiding screen, and give the recommended speed value to maintain the current deceleration trend. When the aircraft is less than 50 meters away from the stop line, the speed control threshold will automatically tighten by 20% to ensure precise control at the end of the approach; at the same time, it has the ability of intelligent learning and continuously optimizes the recommended speed parameters of each aircraft type by analyzing historical operation data.

[0114] Please refer to Figure 7 the flowchart of the airport berth guidance method provided by the embodiment of the present application shown in; The embodiment of the present application provides an airport berth guidance method, including: Step S210: Extract features from the to-be-processed airport three-dimensional point cloud data to obtain multiple feature points of the target aircraft.

[0115] Step S220: Determine the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data.

[0116] Step S230: Determine the distances between multiple feature points based on the coordinate positions of the multiple feature points.

[0117] Step S240: Identify the identification model of the target aircraft from the aircraft model database according to the distances between the multiple feature points. The aircraft model database stores the corresponding relationship between the distances between the feature points and the identification models of the aircraft.

[0118] The implementation manners of the above Step S210 to Step S240 are similar to those of the above Step S110 to Step S140. For unclear parts, refer to the implementation manners of the above Step S110 to Step S140, so they will not be elaborated here. In the implementation process of the above solution, by performing feature extraction on the airport 3D point cloud data to be processed, multiple feature points of the target aircraft can be obtained. The coordinate positions and the distances between these feature points can accurately describe the geometric shape and structure of the aircraft, providing rich and accurate feature information for subsequent aircraft model identification, thereby improving the accuracy of aircraft identification. In addition, the identification model of the target aircraft is identified from the aircraft model database according to the distances between the multiple feature points. This identification method based on feature point matching can quickly find the model that matches the features of the target aircraft in the aircraft model database, greatly improving the identification efficiency and reducing the waiting time of the aircraft at the airport.

[0119] Step S250: Query the target berth information corresponding to the identification model of the target aircraft in the airport berth database.

[0120] As an optional implementation manner of the above Step S250, for example: It can be understood that since the airport berth database stores the berth information corresponding to each aircraft model, it is possible to query in the multiple berth information in the airport berth database and automatically screen the candidate berth information that is currently idle at the airport and meets the size requirements according to the identification model of the target aircraft. Then, perform 3D point cloud analysis on the candidate berth areas in the candidate berth information, such as: detecting the clarity of the ground markings (stop line, guiding line), scanning the surrounding obstacles (such as ground service vehicles, temporary equipment), calculating the path curvature from the berth entrance to the stop line (ensuring that the turning radius ≥ the aircraft model requirement), etc., and comprehensively evaluating the path safety (such as the minimum lateral safety distance), operation efficiency (such as taxiing distance, number of turns), ground service support status (such as fuel truck, gangway preparation status), etc. of each candidate berth information. Finally, output the target berth information corresponding to the identification model of the target aircraft. Among them, the above berth information may include berth number, minimum wingspan space, maximum height limit, clearance height, gangway docking position, 3D coordinates of the standard stop point, and key nodes of the recommended taxiing path, etc.

[0121] Step S260: Generate target guidance information based on the target berth information and multiple feature points of the target aircraft. The target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0122] The implementation manner of the above step S260 is similar to that of the above step S160. For unclear parts, refer to the implementation manner of the above step S160, so it will not be elaborated here. It can be understood that by deeply coupling aircraft type recognition and berth guidance, after recognizing the aircraft type according to the distance of feature points, the precise berth parameter information corresponding to this aircraft type (such as the position of the stop line, the hovering height of the engine, etc.) is immediately called to form a complete spatial cognition chain of "recognition - positioning - guidance". This closed-loop design enables the guidance instruction to fully consider the structural characteristics of a specific aircraft type, avoiding the adaptation error caused by general guidance, thereby improving the accuracy of airport berth guidance.

[0123] In the implementation process of the above solution, by generating target guidance information based on the target berth information and multiple feature points of the target aircraft, this guidance information is for a specific berth of a specific aircraft, which can fully consider the size and shape of the aircraft as well as the specific position and space limitations of the berth, thereby providing more personalized and accurate guidance to help the aircraft enter and park at the airport berth more smoothly. Further, since the guidance information is generated based on real-time three-dimensional point cloud data and can timely reflect the current states of the aircraft and the berth, the system can adjust the guidance information in real time according to the actual position and attitude of the aircraft to ensure the accuracy and timeliness of the guidance and improve the success rate of aircraft parking.

[0124] Please refer to Figure 8 the structural schematic diagram of the aircraft type recognition device provided by the embodiment of the present application shown; The embodiment of the present application provides an aircraft type recognition device 300, including: A point cloud feature extraction module 310, configured to extract features from the to-be-processed airport three-dimensional point cloud data to obtain multiple feature points of the target aircraft.

[0125] A coordinate position determination module 320, configured to determine the coordinate positions of the multiple feature points according to the airport three-dimensional point cloud data.

[0126] A feature distance determination module 330 is configured to determine the distances between the multiple feature points according to the coordinate positions of the multiple feature points.

[0127] An aircraft type recognition module 340, configured to identify the identification model of the target aircraft from the aircraft type database according to the distances between the multiple feature points. The corresponding relationship between the distances between the feature points and the identification models of the aircraft is stored in the aircraft type database.

[0128] As an optional implementation manner of the above device, the aircraft type recognition device further includes: A point cloud model acquisition module, configured to acquire a preset background model point cloud and a calibration reference model.

[0129] A point cloud model alignment module, configured to align the calibration reference model with a calibration object model found in the airport three-dimensional point cloud data to obtain the aligned point cloud data.

[0130] A background model deletion module, configured to delete the preset background model point cloud from the aligned point cloud data.

[0131] As an optional implementation manner of the above device, the multiple feature points include: the nose tip point of the aircraft, the center point of the first engine, and the center point of the second engine; the feature distance determination module includes: An engine spacing calculation sub-module, configured to calculate the engine spacing between the center point of the first engine and the center point of the second engine according to the coordinate positions of the center point of the first engine and the center point of the second engine.

[0132] A first distance calculation sub-module, configured to calculate the first straight-line distance between the nose tip point of the aircraft and the center point of the first engine according to the coordinate positions of the nose tip point of the aircraft and the center point of the first engine.

[0133] A second distance calculation sub-module, configured to calculate the second straight-line distance between the nose tip point of the aircraft and the center point of the second engine according to the coordinate positions of the nose tip point of the aircraft and the center point of the second engine.

[0134] As an optional implementation manner of the above device, the multiple feature points further include: a ground reference point; the feature distance determination module further includes: A ground clearance calculation sub-module, configured to calculate the ground clearance of the nose tip point of the aircraft according to the coordinate positions of the ground reference point and the nose tip point of the aircraft.

[0135] And / or, a first height calculation sub-module, configured to calculate the ground clearance of the center point of the first engine according to the coordinate positions of the ground reference point and the center point of the first engine.

[0136] And / or, a second height calculation sub-module, configured to calculate the ground clearance of the center point of the second engine according to the coordinate positions of the ground reference point and the center point of the second engine.

[0137] As an optional implementation manner of the above device, the aircraft model identification module includes: An aircraft model matching sub-module, configured to match multiple candidate aircraft models in the aircraft model database according to the distances between the multiple feature points.

[0138] A feature error calculation sub-module, configured to calculate the error value between the preset distance of the candidate aircraft model and the distances between the multiple feature points.

[0139] The aircraft model determination sub-module is used to determine the identification model of the target aircraft by selecting the candidate aircraft model with the smallest error value from multiple candidate aircraft models.

[0140] As an alternative implementation of the above device, the aircraft model identification device further includes: The berth information determination module is used to query the target berth information corresponding to the identification model of the target aircraft in the airport berth database.

[0141] The guidance information generation module is used to generate target guidance information based on the target berth information and multiple feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0142] As an alternative implementation of the above device, the aircraft model identification device further includes: The flight model receiving module is used to receive the current flight model sent by the server.

[0143] The preliminary berth determination module is used to determine preliminary berth information based on the airport three-dimensional point cloud data and the current flight model.

[0144] The preliminary guidance determination module is used to generate preliminary guidance information based on the current flight model and the preliminary berth information.

[0145] As an alternative implementation of the above device, the multiple feature points include: the nose tip point; the guidance information generation module includes: The nose stop judgment sub-module is used to judge whether the nose tip point exceeds the stop line in the berth information.

[0146] The distance reminder guidance sub-module is used to generate the distance reminder information in the target guidance information if the nose tip point does not exceed the stop line in the berth information, and the distance reminder information includes the distance between the current position and the stop line.

[0147] As an alternative implementation of the above device, the guidance information generation module further includes: The deviation reminder guidance sub-module is used to generate the deviation reminder information in the target guidance information if the distance between the nose tip point and the reference center line in the target berth information is greater than the preset distance, and the deviation reminder information includes the relative direction between the nose tip point and the reference center line.

[0148] As an alternative implementation of the above device, the guidance information generation module further includes: The aircraft speed judgment sub-module is used to judge whether the current speed of the target aircraft is greater than the speed threshold.

[0149] The speed reminder guidance submodule is used to generate overspeed reminder information in the target guidance information if the current speed of the target aircraft is greater than the speed threshold, otherwise, to generate deceleration reminder information in the target guidance information, and the overspeed reminder information and the deceleration reminder information include the current speed of the target aircraft.

[0150] The embodiment of the present application provides an airport berth guidance device, comprising: The point cloud feature extraction module is used to extract features from the three-dimensional point cloud data of the airport to be processed and obtain multiple feature points of the target aircraft.

[0151] The coordinate position determination module is used to determine the coordinate positions of multiple feature points based on the airport three-dimensional point cloud data.

[0152] The feature distance determination module is used to determine the distance between multiple feature points according to the coordinate positions of the multiple feature points.

[0153] The aircraft model recognition module is used to identify the identification model of the target aircraft from the aircraft model database according to the distance between multiple feature points. The aircraft model database stores the corresponding relationship between the distance between the feature points and the identification model of the aircraft.

[0154] The berth information determination module is used to query the target berth information corresponding to the identification model of the target aircraft in the airport berth database.

[0155] The guidance information generation module is used to generate target guidance information according to the target berth information and multiple feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

[0156] It should be understood that the device corresponds to the above-mentioned aircraft model identification method embodiment and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be referred to in the above description, and the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or fixed in the operating system (OS) of the device.

[0157] See also Figure 9 An electronic device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores machine-readable instructions executable by the processor 410, and when the machine-readable instructions are executed by the processor 410, the above method is executed.

[0158] The embodiments of the present application also provide a computer-readable storage medium 430. A computer program is stored on the computer-readable storage medium 430. When the computer program is run by a processor 410, the above-mentioned method is executed. Among them, the computer-readable storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM for short), electrically erasable programmable read-only memory (EEPROM for short), erasable programmable read-only memory (EPROM for short), programmable read-only memory (PROM for short), read-only memory (ROM for short), magnetic memory, flash memory, magnetic disk or optical disc.

[0159] The embodiments of the present application also provide a computer program product, including: a computer program or computer instructions. When the computer program or computer instructions are run by a processor, the above-described method is executed.

[0160] It should be noted that the various embodiments in this specification are all described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, reference can be made to the partial description of the method embodiments.

[0161] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are only illustrative. For example, the flowcharts and block diagrams in the drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment or a part of the code. A module, a program segment or a part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may also occur in a different order from that marked in the drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, which mainly depends on the functions involved.

[0162] In addition, each functional module in the embodiments of the present application may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part. In addition, in the description of this specification, the descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific example", "some examples", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0163] The above description is only an alternative implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the embodiments of the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the embodiments of the present application.

Claims

1. A method for identifying an aircraft model, characterized in that: include: Perform feature extraction on the 3D point cloud data of the airport to be processed to obtain multiple feature points of the target aircraft; Determine the coordinate positions of the plurality of feature points according to the three-dimensional point cloud data of the airport; Determine the distance between the plurality of feature points according to the coordinate positions of the plurality of feature points; The identification model of the target aircraft is identified from an aircraft model database according to the distances between the plurality of feature points, wherein the aircraft model database stores a correspondence between the distances between the feature points and the identification model of the aircraft.

2. The method according to claim 1, characterized in that The plurality of feature points include: a nose tip point, a first engine center point, and a second engine center point; and determining the distances between the plurality of feature points according to the coordinate positions of the plurality of feature points includes: Calculating an engine distance between the first engine center point and the second engine center point according to the coordinate position of the first engine center point and the coordinate position of the second engine center point; Calculate a first straight-line distance between the nose tip point and the first engine center point according to the coordinate position of the nose tip point and the coordinate position of the first engine center point; A second straight-line distance between the nose tip point and the second engine center point is calculated according to the coordinate position of the nose tip point and the coordinate position of the second engine center point.

3. The method according to claim 2, characterized in that The plurality of feature points further include: a ground reference point; and the step of determining the distances between the plurality of feature points according to the coordinate positions of the plurality of feature points further includes: Calculate the height of the nose tip point from the ground according to the coordinate position of the ground reference point and the coordinate position of the nose tip point; and / or, calculating the height of the center point of the first engine from the ground according to the coordinate position of the ground reference point and the coordinate position of the center point of the first engine; And / or, the height of the center point of the second engine above the ground is calculated based on the coordinate position of the ground reference point and the coordinate position of the center point of the second engine.

4. The method according to claim 1, characterized in that: The step of identifying the model of the target aircraft from an aircraft model database according to the distances between the plurality of feature points comprises: Matching a plurality of candidate aircraft models in the aircraft model database according to the distances between the plurality of feature points; Calculating an error value between a preset distance of the candidate aircraft model and the distances between the plurality of feature points; A candidate aircraft model with the smallest error value is selected from the multiple candidate aircraft models and determined as the identification model of the target aircraft.

5. The method according to claim 1, characterized in that After identifying the identification model of the target aircraft from the aircraft model database according to the distances between the plurality of feature points, the method further includes: Query the target berth information corresponding to the identification model of the target aircraft in the airport berth database; Target guidance information is generated according to the target berth information and a plurality of feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

6. The method according to claim 5, characterized in that Before querying the target berth information corresponding to the identification model of the target aircraft in the airport berth database, the method further includes: Receive the current flight model sent by the server; Determine preliminary berth information according to the airport three-dimensional point cloud data and the current flight model; Preliminary guidance information is generated according to the current flight model and the preliminary berth information.

7. The method according to claim 5, characterized in that The multiple feature points include: a nose tip point; the generating target guidance information according to the target berth information and the multiple feature points of the target aircraft includes: Determining whether the nose tip point exceeds the stop line in the berth information; If not, distance reminder information in the target guidance information is generated, and the distance reminder information includes the distance between the current position and the stop line.

8. The method according to claim 7, characterized in that The generating target guidance information according to the target berth information and the multiple feature points of the target aircraft further includes: If the distance between the nose tip point and the reference center line in the target berth information is greater than a preset distance, deviation reminder information in the target guidance information is generated, and the deviation reminder information includes a relative direction between the nose tip point and the reference center line.

9. The method according to claim 5, characterized in that The generating target guidance information according to the target berth information and the multiple feature points of the target aircraft comprises: Determining whether the current speed of the target aircraft is greater than a speed threshold; If so, then the overspeed reminder information in the target guidance information is generated; otherwise, then the deceleration reminder information in the target guidance information is generated, and the overspeed reminder information and the deceleration reminder information include the current speed of the target aircraft.

10. An airport berth guidance method, characterized in that: include: Perform feature extraction on the 3D point cloud data of the airport to be processed to obtain multiple feature points of the target aircraft; Determine the coordinate positions of the plurality of feature points according to the three-dimensional point cloud data of the airport; Determine the distance between the plurality of feature points according to the coordinate positions of the plurality of feature points; Identifying the identification model of the target aircraft from an aircraft model database according to the distances between the plurality of feature points, wherein the aircraft model database stores a correspondence between the distances between the feature points and the identification model of the aircraft; Query the target berth information corresponding to the identification model of the target aircraft in the airport berth database; Target guidance information is generated according to the target berth information and a plurality of feature points of the target aircraft, and the target guidance information is used to guide the target aircraft to enter and park at the airport berth.

11. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions are executed by the processor to perform any method according to claims 1 to 10.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 10 is executed.

13. A computer program product, characterized in that include: A computer program or a computer instruction, wherein when the computer program or the computer instruction is executed by a processor, the method according to any one of claims 1 to 10 is executed.

Images