Airport clearance dynamic monitoring method and equipment based on parametric automatic modeling

Abstract

The embodiments of the present invention provide a method and device for dynamic monitoring of airport clearance based on parametric automatic modeling. The method includes obtaining the site parameters of the target airport and generating the clearance restriction surface of the target airport; establishing a multi-temporal remote sensing image spatial database and generating an obstacle vector surface; monitoring the multi-temporal remote sensing image spatial database and updating the obstacle vector surface; calculating the ground elevation and top elevation of the updated obstacle vector surface and generating a three-dimensional model of the obstacle; establishing a semantic thematic spatial database and obtaining a semantically objectified obstacle three-dimensional model; monitoring the semantic thematic spatial database and updating the semantically objectified obstacle three-dimensional model with the semantic attribute information in the added semantic thematic data to obtain the updated change information of the semantically objectified obstacle three-dimensional model; and obtaining the superelevation detection result of the target airport. In this way, the changes in the obstacle three-dimensional model can be accurately identified and dynamically monitored.

Description

Technical Field

[0001] The present invention generally relates to the field of computers, and more particularly, to an airport clearance dynamic monitoring method and device based on parameterized automatic modeling. Background Art

[0002] Airport clearance is designed to ensure the safety of aircraft takeoffs, landings, and go-arounds. The spatial zones around airports, which limit the terrain and height of features, are essential for safe aircraft operations. With the continuous advancement of urbanization and the development of airport-related economic zones, the precise identification and dynamic monitoring of over-height obstacles and obstacles approaching height limits within airport clearance zones is crucial.

[0003] Existing airport clearance monitoring methods and preliminary monitoring and analysis services for new airport site selection are mainly based on the analysis of remote sensing data. However, they lack the ability to aggregate multi-source data to jointly participate in the dynamic monitoring and calculation of airport clearance, resulting in the inability to achieve automated, intelligent, and automatic full-process call for multi-source data to participate in calculations. Summary of the Invention

[0004] According to an embodiment of the present invention, a dynamic airport clearance monitoring solution based on parametric automatic modeling is provided. This solution can accurately identify and dynamically monitor changes in obstacle 3D models by monitoring data changes in a multi-temporal remote sensing image spatial database and a semantic thematic spatial database.

[0005] In the first aspect of the present invention, a method for dynamic monitoring of airport clearance based on parametric automatic modeling is provided. The method includes: obtaining the site parameters of the target airport and generating the clearance restriction surface of the target airport; establishing a multi-temporal remote sensing image spatial database, obtaining the digital surface model and digital true orthogonal image within the clearance restriction surface, extracting the obstacle surface contour within the clearance restriction surface, and generating an obstacle vector surface; monitoring the multi-temporal remote sensing image spatial database, and if the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database increases, extracting the obstacle surface contour within the clearance restriction surface based on the increased digital surface model and digital true orthogonal image, and updating the obstacle vector surface; calculating the ground elevation and top elevation of the updated obstacle vector surface, and generating a three-dimensional model of the obstacle; establishing a semantic thematic spatial database, obtaining semantic thematic data within the clearance restriction surface, establishing a spatial association relationship between the semantic thematic data and the obstacle three-dimensional model, and assigning semantic attribute information in the semantic thematic data to the three-dimensional model of the obstacle. The method comprises the following steps: monitoring the semantic thematic spatial database, and if the semantic thematic data within the clearance restriction surface in the semantic thematic spatial database increases, updating the spatial association relationship between the increased semantic thematic data and the obstacle three-dimensional model, and updating the semantic thematic obstacle three-dimensional model with the semantic attribute information in the increased semantic thematic data; spatially associating the updated semantic thematic obstacle three-dimensional model with the semantic thematic obstacle three-dimensional model before the update, performing three-dimensional spatial information comparison and semantic attribute comparison on the spatially associated obstacle three-dimensional model, and obtaining the updated change information of the semantic thematic obstacle three-dimensional model; spatially intersecting the part of the change information corresponding to the updated semantic thematic obstacle three-dimensional model with the clearance restriction surface of the target airport, and obtaining the superelevation detection result of the target airport.

[0006] In the second aspect of the present invention, a device for dynamic monitoring of airport clearance based on parametric automatic modeling is provided. The device includes: a clearance restriction surface generation module, which is used to obtain the parameters of the target airport site and generate the clearance restriction surface of the target airport; a three-dimensional model generation module for obstacles, which is used to establish a multi-temporal remote sensing image spatial database, obtain the digital surface model and digital true orthogonal image within the clearance restriction surface, extract the obstacle surface contour within the clearance restriction surface, and generate an obstacle vector surface; monitor the multi-temporal remote sensing image spatial database, if the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database increases, then extract the obstacle surface contour within the clearance restriction surface based on the increased digital surface model and digital true orthogonal image, and update the obstacle vector surface; calculate the ground elevation and top elevation of the updated obstacle vector surface, and generate a three-dimensional model of the obstacle; a semantic object-oriented obstacle three-dimensional model generation module, which is used to establish a semantic thematic spatial database, obtain the semantic thematic data within the clearance restriction surface, establish a spatial association relationship between the semantic thematic data and the obstacle three-dimensional model, and transform the semantic thematic data into the three-dimensional model of the obstacle. The semantic attribute information in the semantic thematic data is assigned to the corresponding three-dimensional obstacle model to obtain a semantically objectified three-dimensional obstacle model; the semantic thematic space database is monitored, and if the semantic thematic data within the clearance restriction surface in the semantic thematic space database is increased, the spatial association relationship between the increased semantic thematic data and the three-dimensional obstacle model is updated, and the semantically objectified three-dimensional obstacle model is updated with the semantic attribute information in the increased semantic thematic data; an association and comparison module is used to spatially associate the updated semantically objectified three-dimensional obstacle model with the semantically objectified three-dimensional obstacle model before the update, and perform three-dimensional spatial information comparison and semantic attribute comparison on the spatially associated three-dimensional obstacle model to obtain the updated change information of the semantically objectified three-dimensional obstacle model; an ultra-elevation detection module is used to spatially intersect the part of the change information corresponding to the updated semantically objectified three-dimensional obstacle model with the clearance restriction surface of the target airport to obtain the ultra-elevation detection result of the target airport.

[0007] In a third aspect of the present invention, an electronic device is provided. The electronic device comprises at least one processor; and a memory communicatively connected to the at least one processor; the memory storing instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method of the first aspect of the present invention.

[0008] It should be understood that the contents described in the summary of the invention are not intended to limit the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present invention will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1 A flow chart of a method for dynamic monitoring of airport clearance based on parameterized automatic modeling according to an embodiment of the present invention is shown;

[0011] Figure 2 A flow chart of generating a clearance limiting surface according to an embodiment of the present invention is shown;

[0012] Figure 3 A schematic diagram of a clearance restriction surface generation specification according to an embodiment of the present invention is shown;

[0013] Figure 4 A flowchart of generating an obstacle vector plane according to an embodiment of the present invention is shown;

[0014] Figure 5 A schematic diagram of the U-Net network model structure according to an embodiment of the present invention is shown;

[0015] Figure 6 A flow chart of deep learning model generation according to an embodiment of the present invention is shown;

[0016] Figure 7 A flowchart of generating a three-dimensional model of an obstacle according to an embodiment of the present invention is shown;

[0017] Figure 8 A flowchart of three-dimensional spatial information comparison according to an embodiment of the present invention is shown;

[0018] Figure 9 A block diagram of an airport clearance dynamic monitoring device based on parameterized automatic modeling according to an embodiment of the present invention is shown;

[0019] Figure 10 shows a block diagram of an exemplary electronic device capable of implementing embodiments of the present invention;

[0020] Among them, 1000 is an electronic device, 1001 is a CPU, 1002 is a ROM, 1003 is a RAM, 1004 is a bus, 1005 is an I / O interface, 1006 is an input unit, 1007 is an output unit, 1008 is a storage unit, and 1009 is a communication unit. DETAILED DESCRIPTION

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0023] Figure 1 A flow chart of a method for dynamic monitoring of airport clearance based on parameterized automatic modeling according to an embodiment of the present invention is shown.

[0024] The method includes:

[0025] S101. Obtain parameters of a target airport site and generate a clearance restriction surface for the target airport.

[0026] As an embodiment of the present invention, the airport's clearance restriction surfaces include two types: civil aviation and military aviation, and civil aviation and military aviation each have different restriction surface levels and restriction surface generation parameters. For example, when civil aviation selects a type, the level of the type will also be determined based on the runway length.

[0027] The airport clearance restriction surfaces of civil aviation include: runways, lift strips, approach surfaces, take-off climb surfaces, inner horizontal surfaces, conical surfaces, transition surfaces, etc.

[0028] The airport clearance restriction surfaces for military aviation include: runway, takeoff and landing strip, end clearance, inner horizontal surface, conical surface, outer horizontal surface, transition surface, etc.

[0029] Before generating the clearance restriction surface of the target airport, you need to input the parameters of the target airport site.

[0030] As an embodiment of the present invention, the parameters of the airport site include the flight zone level (ie, runway type), runway length, width, center point coordinates, runway direction, elevation, etc.

[0031] In this embodiment, if Figure 2 As shown, the step of obtaining the target airport site parameters and generating the target airport's clearance restriction surface includes:

[0032] S201: Acquire target airport site parameters, and obtain size and slope information of a clearance restriction surface corresponding to the target airport site parameters.

[0033] Obtain the corresponding clearance restriction surface generation specifications based on the target airport type. The clearance restriction surface generation specifications are based on relevant national standards, such as Figure 3 As shown in the figure, the airports with different runway types are specified in detail, and the length, width, deployment rate, height, slope and other requirements of various restriction surfaces are generated accordingly.

[0034] The target airport site parameters are matched to the clearance restriction surface generation specification to obtain the size and slope information of the corresponding clearance restriction surface.

[0035] S202: Calculate the three-dimensional spatial coordinate point of each clearance restriction surface based on the size and slope information of the clearance restriction surface.

[0036] As an embodiment of the present invention, a vector file in the ShapeFile format is first created, the vector element type is a multipatch, and the three-dimensional space coordinates of each type of restricted surface are calculated according to the specification requirements.

[0037] The three-dimensional space coordinate calculation process includes:

[0038] Runway: According to the filled-in longitude and latitude coordinates of the center point, convert the corresponding projection to obtain the coordinate position, extend the corresponding length and width accordingly, and give the coordinate position and height according to the corresponding elevation and direction at both ends.

[0039] Liftway: Extend the height and width accordingly based on the runway and maintain the same end height as the runway.

[0040] Approach surface: divided into left and right sides; taking the left side as an example, it is divided into the first, second and third sections. The elevation of the left side of the elevator strip is used as the basic height, and the upper left point of the elevator strip is used as the starting point of the first section on the left. It extends to the left according to the length and slope of the first section and climbs as the upper left x at the end of the first section on the left. The value of y is calculated according to the dispersion rate and the length of the first section. The height of the position is calculated by the length and slope to obtain the coordinates of the upper left point at the end of the first section on the left. The rest are calculated based on this calculation. The third section is a horizontal section with an unchanged height, and the others are similar.

[0041] Takeoff climb surface: divided into left and right sides; taking the left side as an example, it is divided into the first and second sections; taking the elevation of the left side of the lift strip as the basic height, and the center point of the left side of the runway as the basis, y increases the corresponding length as the starting point of the first section on the left, and then extends, spreads and climbs in a manner similar to the approach surface.

[0042] Inner horizontal plane: divided into a middle rectangle and semicircles on the left and right sides; the rectangle is formed by increasing the corresponding values ​​of y of the four points of the lift belt, and then the center point of the left side of the runway is used as the center of the circle. The rectangle is constructed according to a certain semicircular arc, and the height is the average height of the two ends plus the corresponding height, and the right side is the same.

[0043] Conical surface: divided into upper and lower rectangles in the middle and semicircles on the left and right sides; with the left and right points above the inner horizontal rectangle, the y value increases accordingly to construct the upper rectangle, lower rectangle, etc.; secondly, with the center point on the left side of the runway as the center of the circle, the inner horizontal outer rectangle is constructed according to a certain semicircular arc, and the height is the average height of the two ends plus the corresponding height, and the same is true on the right side.

[0044] Transition surface: divided into upper and lower rectangles and triangles corresponding to the left and right sides respectively; based on the left point on the upper edge of the lift strip, this is point 1, the height of this point and the transition slope and the inner horizontal height are used to calculate the value of y, this is point 2, climbing to the inner horizontal height through the approach surface is the value of x, and the value of y is determined by the value and diffusion rate, this is point 3, and the remaining three angles are calculated in the corresponding manner.

[0045] S203: Connect the three-dimensional spatial coordinate points of each clearance restriction surface into a three-dimensional polyhedron to obtain vector data of the three-dimensional polyhedron.

[0046] The three-dimensional space coordinate points of each clearance limiting surface are connected to form a three-dimensional polyhedron, that is, not all the obtained three-dimensional space coordinate points are connected to each other. For example, two coordinate points on a diagonal line do not need to be connected because if they are connected, the connecting line passes through the interior of the three-dimensional polyhedron and the three-dimensional polyhedron cannot be finally formed.

[0047] S102: Establish a multi-temporal remote sensing image spatial database, obtain a digital surface model and a digital true orthogonal image within the clearance restriction surface, extract the obstacle surface contours within the clearance restriction surface, and generate an obstacle vector surface.

[0048] The multi-temporal remote sensing image spatial database stores remote sensing images from multiple time periods, including digital surface models (DSMs) and digital true orthogonal images (TDOMs). The DSMs and TDOMs are generated through periodic aerial photogrammetry, with data collection, processing, and storage occurring regularly. DSM and TDOM data are stored for each temporal phase, organized and stored using a pre-designed database structure.

[0049] In this embodiment, each clearance restriction surface created in S101 is first obtained. Each clearance restriction surface is stored as a vector file in the ShapeFile format. The four boundaries (maximum, minimum, longitude, and latitude) of each vector file are read separately to calculate the overall maximum four boundaries (minimum longitude, minimum latitude, maximum longitude, and maximum latitude).

[0050] Next, a multi-temporal remote sensing image spatial database is accessed and a spatial query is performed based on the maximum bounding range to select remote sensing image data within the clearance restriction surface of the target airport. This includes spatially intersecting the digital surface model and digital true orthogonal image data in the multi-temporal remote sensing image spatial database with the maximum bounding range of the vector data of the clearance restriction surface to obtain the digital surface model and digital true orthogonal image data within the clearance restriction surface.

[0051] Finally, if Figure 4 As shown, extracting the obstacle surface contour within the clearance restriction surface and generating an obstacle vector surface includes:

[0052] S401-1. Noise is filtered out of the added digital surface model, and obstacle contours are extracted based on elevation information to obtain obstacle terrain feature contour data.

[0053] As one embodiment of the present invention, noise filtering in DSM data involves filtering out ground-based features (small, isolated objects on the ground) and non-obstacles (non-obstacle features such as cars, trees, flower beds, utility poles, and walls). This clutters the DSM data and makes feature identification and extraction difficult. Therefore, noise filtering is required.

[0054] In this embodiment, noise filtering includes filtering out ground objects and non-obstacles.

[0055] Among them, filtering out ground objects includes:

[0056] Since the elevation of ground-attached objects is relatively low and there is a significant height difference between them and airport obstacles, they can be filtered out by applying a pixel threshold segmentation algorithm to the DSM. The pixel values ​​in the DSM raster data are elevation values. The threshold segmentation algorithm separates airport obstacles from ground-attached objects by setting a constant, namely the threshold. The thresholding process traverses the pixel values ​​of the input raster data, compares them with the threshold constant, and assigns new pixel values ​​to form new raster data. The formula is as follows:

[0057] Definition: in(x,y) is the pixel value of the input raster data, where x is the pixel column number and y is the pixel row number; out(x,y) is the pixel value of the output raster data, where x is the pixel column number and y is the pixel row number; T is the height threshold.

[0058] If in(x,y)≥T; then out(x,y)=1;

[0059] If in(x,y)<T; then out(x,y)=0;

[0060] Through pixel threshold segmentation calculation, airport obstacles and ground-attached objects can be separated and the ground-attached objects can be filtered out.

[0061] Filter out obstructions, including:

[0062] After filtering out ground attachments, some non-obstacles (such as cars, trees, flower beds, utility poles, and courtyard walls) that exceed the threshold still require further filtering. Because non-obstacles often have irregular top shapes, geometric rules are used to filter them out. By setting parameters such as slope, geometry, height, and area, an adaptive DSM filtering algorithm is used to remove non-obstacle information.

[0063] After noise filtering, obstacle contours are extracted based on elevation information to obtain obstacle terrain feature contour data, including:

[0064] Obstacles have relatively regular shapes, and their contours are best represented by straight lines. Therefore, the Sobel operator, a mature edge detection algorithm, is used to extract obstacle edge contours from the noise-filtered DSM. Hough transforms are then used to detect line features within the edges, yielding obstacle contour data. The Sobel operator performs well on images with grayscale gradients and high noise levels, and accurately locates edges, making it well-suited for processing DSM image data. The Sobel operator employs a weighted average, followed by a differential operation, and finally, a gradient calculation. The Hough transform is used to detect curves in images that can be described by functional relationships, such as lines, circles, parabolas, and ellipses. It has been successfully applied in many fields, including image analysis and pattern recognition. The Hough transform utilizes global image properties (such as texture information) to connect edge pixels to form the boundaries of closed regions. Its basic principle is to transform the points that constitute a curve (including a straight line) in image space into parameter space. By detecting extreme points in the parameter space, the descriptive parameters of the curve are determined, resulting in the equation of the curve (or line).

[0065] S401-2. Extract the texture features of the added digital true orthogonal image, input it into the deep learning model, and output the obstacle texture feature contour data.

[0066] As an embodiment of the present invention, based on a deep learning method, a U-Net network model is built using the Pytorch framework to identify and extract obstacles in TDOM remote sensing images.

[0067] The U-Net network model structure is as follows Figure 5As shown in the figure, the U-Net network expands on the concept of fusing low-dimensional and high-dimensional features in fully convolutional neural networks. Each layer in the left half of the network uses two 3×3 convolution operations, maintaining the same number of convolution kernels. This reduces the spatial dimensions of the input data and extracts high-level representative features. The right half uses deconvolution layers to perform multiple upsampling operations from bottom to top, gradually restoring detailed information in the lower spatial dimensions. Furthermore, batch normalization (BN) layers are introduced to address the model's proneness to vanishing and exploding gradients. Dropout layers are also introduced to address overfitting caused by excessive model parameters and a small number of training samples. Once the U-Net model is constructed, the model parameters can be set and it can be used for dataset training.

[0068] Before extraction, sample library preparation, U-Net network training, sample update and other operations have been completed in advance to obtain the final deep learning model. The process is as follows Figure 6 A brief introduction is as follows:

[0069] ① Sample Library Construction: Targeting obstacles, we constructed an obstacle sample library based on optical remote sensing imagery. The sample library includes sample images and image label maps (obstacles are filled with white and the background is filled with black). First, we overlay and analyze the existing obstacle data with the remote sensing imagery. Through manual inspection, we screen out areas containing real obstacles. After vectorization, we fill the areas inside the obstacles with white and those outside with black to construct a sample label map. The image data and label map are then simultaneously cropped to 256×256 pixels. This process standardizes the samples and increases the sample size, completing the construction of the preliminary sample library.

[0070] ②U-Net network training: The U-Net network model was implemented using the deep learning framework Pytorch. The experimental samples were augmented using methods such as rotation, mirroring, and flipping.

[0071] ③ Sample Update: Samples are crucial to experimental results. A high-quality sample library enables rapid model convergence, resulting in smaller loss values ​​and higher overall accuracy. Sample Update Process: First, the network is initially trained based on the initially constructed sample library to obtain a preliminary training model. Then, the preliminary training model is used to classify the sample library, determining the extraction accuracy of each sample. A precision threshold is then selected, and samples below the threshold are removed from the sample library. This cycle is repeated sample by sample, updating the sample library three times to obtain a purer sample library, which serves as the final network model training input.

[0072] According to the pre-trained final model, TDOM image data is input to extract obstacles, and rasterized obstacle annotation data after extraction and classification can be obtained.

[0073] As an embodiment of the present invention, after identification and extraction based on a deep learning framework, binary obstacle annotation raster data is obtained, in which the obstacles are filled with white and the background is filled with black; based on the binary raster data, the raster data conversion vector data algorithm of the open source geospatial data operation library GDAL is used to obtain the obstacle contour vector surface data.

[0074] S402: Perform vector surface fusion on the obstacle terrain feature contour data and the obstacle texture feature contour data to obtain a fused vector surface.

[0075] The obstacle vector surface extracted from TDOM based on deep learning has accurate two-dimensional plane coordinate information. However, due to the influence of shadows around the obstacles and special spectral textures, some vector surfaces cannot be well located at the obstacle edges. The obstacle contour information extracted from DSM has elevation information and a relatively complete topological relationship. The segmentation result of the structure boundary is flatter and smoother. Therefore, the two can complement each other and improve the accuracy of the obstacle vector surface through fusion.

[0076] The vector data fusion algorithm of the open source geospatial data operation library GDAL is used to achieve the fusion of two types of obstacle vector surfaces. The vector surface data at the same spatial position are fused into one vector surface data.

[0077] S403: Perform topology check and correction on the fused vector plane to obtain the obstacle vector plane.

[0078] The fused obstacle vector surface data also needs to be corrected for spatial topology. In this embodiment, the spatial topology correction is performed using the ArcGIS software platform, which specifically includes:

[0079] ① Convert multiple components to a single component. The fused obstacle vector surface may contain multiple components, meaning multiple surfaces belong to the same feature. This requires converting the multi-component data to a single component. A multi-component refers to a feature containing multiple surfaces; a single component refers to a feature containing only one surface. This is done using the "Multipart To Singlepart" tool in the ArcGIS software platform.

[0080] ②Topology check

[0081] The ArcGIS software platform is used to establish topological rules for surface features and perform topological error checks.

[0082] The topology rules are as follows:

[0083] First, faces cannot overlap. Obstacle vector faces cannot overlap. The data of overlapping parts can be obtained through topology checking.

[0084] Repair method: Use the "Clip" tool of ArcGIS software to eliminate the overlapping parts and obtain non-overlapping surfaces.

[0085] Second, there must be no tiny holes in the surface. A single obstacle's vector surface must not have tiny holes or gaps. Topology checks can be used to obtain overlapping data.

[0086] Repair method: Use the "EliminatePolygonPart" tool in ArcGIS software (creates a new output feature class containing features obtained by deleting certain parts or holes of specified sizes from the input surface), and according to the set area threshold (holes smaller than the given area will be eliminated), eliminate tiny holes or gaps to obtain a complete surface.

[0087] ③Topological error correction

[0088] Finally, design correction methods based on the different topological errors. Obtain the topological check results and identify the surface features with topological errors. Then, follow the repair method described in step 2 above, calling different repair tools in ArcGIS software based on the different topological error types to repair the topological errors.

[0089] After repair, the topology check-repair process is repeated until no topology errors exist. The final result is the refined obstacle vector surface.

[0090] Furthermore, after the initial obstacle vector surface is generated, it is necessary to monitor the multi-temporal remote sensing image spatial database, including:

[0091] The multi-temporal remote sensing image spatial database is monitored in real time, and an increase in a digital surface model (DSM) and a digital true orthogonal image (TDOM) within a clearance restriction surface in the multi-temporal remote sensing image spatial database is used as a trigger condition for the singulation process. The temporal phase corresponding to the increased data is recorded.

[0092] If the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database increases, the individualization process is executed, namely:

[0093] The obstacle surface contour within the clearance restriction surface is extracted based on the added digital surface model DSM and the digital true orthogonal image TDOM, and the obstacle vector surface is updated; the ground elevation and top surface elevation of the updated obstacle vector surface are calculated to generate a three-dimensional model of the obstacle.

[0094] As an embodiment of the present invention, Figure 7As shown, the calculation of the ground elevation and top elevation of the updated obstacle vector plane to generate a three-dimensional model of the obstacle includes:

[0095] S701: Filter the added digital surface model to obtain digital elevation data, superimpose the obstacle vector surface with the digital elevation data, calculate the ground elevation of each vertex on each obstacle vector surface, and obtain the ground elevation of the obstacle vector surface;

[0096] S702: superimposing the obstacle vector plane onto the added digital surface model, calculating the top surface elevation of each vertex on the obstacle vector plane, and obtaining the top surface elevation of the obstacle vector plane;

[0097] S703: Obtaining height information of the obstacle vector plane according to the ground elevation and the top elevation of the obstacle vector plane;

[0098] S704: Establish a three-dimensional model of the obstacle based on the obstacle vector surface and its ground elevation, top elevation, and height information.

[0099] In the above embodiment, specifically, by using an elevation interpolation algorithm, the top elevation of each corner point on the obstacle vector surface can be obtained. Then, according to the principle that the elevation of the obstacle top is the same, the least squares method is used to obtain the top elevation of the obstacle top that best matches it. The DSM is filtered to obtain a digital elevation model, that is, the ground elevation obtained after removing surface attachments. Then, similarly, the obstacle vector surface is superimposed on the newly generated DEM in the same manner as above. By using an elevation interpolation algorithm, the ground elevation of each corner point on the obstacle vector surface can be obtained. Then, according to the principle that the elevation of the obstacle top is the same, the least squares method is used to obtain the top elevation of the obstacle top that best matches it. Finally, the top elevation of the obstacle vector surface and the ground elevation are subtracted to obtain the height of the obstacle itself. Based on the top elevation, ground elevation, and self-height information of the obstacle vector surface obtained in the previous step, vertical stretching is performed according to the self-height information to obtain a three-dimensional geometric model of the obstacle.

[0100] As an embodiment of the present invention, the method further includes:

[0101] According to a preset cycle, the data within the clearance restriction surface of the target airport in the multi-temporal remote sensing image spatial database is monitored, and the increase of the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database is used as the execution trigger condition of the individualization process.

[0102] When a preset period is reached, the data of the latest phase within the clearance restriction surface of the target airport in the multi-temporal remote sensing image spatial database is obtained, and it is determined whether the obtained data has increased compared with the corresponding data of the previous phase in the multi-temporal remote sensing image spatial database. If the data has increased, the obstacle surface contour within the clearance restriction surface is extracted based on the increased digital surface model and digital true orthogonal image, and the obstacle vector surface is updated; the ground elevation and top surface elevation of the updated obstacle vector surface are calculated to generate a three-dimensional model of the obstacle.

[0103] S103. Establish a semantic thematic spatial database, obtain semantic thematic data within the clearance restriction surface, establish a spatial association relationship between the semantic thematic data and the three-dimensional obstacle model, assign semantic attribute information in the semantic thematic data to the corresponding three-dimensional obstacle model, and obtain a semantically objectified three-dimensional obstacle model.

[0104] As an embodiment of the present invention, a semantic thematic spatial database is pre-constructed and is a spatial database that stores semantic thematic information; wherein the semantic thematic information is vector spatial data (points, lines, and surfaces), and has semantic themes such as land feature types and place names and addresses, which are derived from geocoding / reverse geocoding (Internet-oriented Chinese address semantic acquisition and analysis), regularly obtain the latest place name and address data, and perform spatialization, extract semantic information through specific model analysis, and regularly update the spatial database of semantic thematic information.

[0105] By performing spatial query based on the maximum bounds of the clearance restriction surface, semantic thematic information within the clearance range can be screened out.

[0106] Obtain the above-mentioned obstacle 3D model, traverse each obstacle, and further filter out the semantic thematic information associated with the obstacle space based on the spatial query.

[0107] Semantic thematic information is extracted and organized according to specific rules and assigned to the 3D obstacle model. Attribute information is then added to the 3D obstacle model, such as by creating an attribute table and assigning values ​​to attribute fields. The resulting semantically objectified 3D obstacle model is obtained. This semantically objectified information includes feature type, place name, and address.

[0108] Furthermore, the embodiment of the present invention also includes:

[0109] The semantic thematic space database is monitored in real time, and the increase of semantic thematic data within the clearance restriction surface in the semantic thematic space database is used as a trigger condition for the semantic objectification process. The time phase corresponding to the increase of semantic thematic data is recorded.

[0110] If the semantic thematic data within the clearance restriction surface in the semantic thematic spatial database increases, then the spatial association relationship between the increased semantic thematic data and the three-dimensional obstacle model is updated, that is, the increased semantic thematic data is associated with the spatial information of the corresponding three-dimensional obstacle model;

[0111] Then, the semantically objectified three-dimensional obstacle model is updated with the semantic attribute information in the added semantic thematic data, that is, the semantic attribute information in the added semantic thematic data is assigned to the three-dimensional obstacle model associated therewith.

[0112] Semantic attribute information is descriptive information about obstacles, including place name, address, category, land use type, ownership unit, ownership nature, location unit, construction time, status, planning period, etc. Object-oriented description is achieved through the association and interaction of semantic attribute information.

[0113] Updates to semantic thematic data record the time, cause, and inter-data relationships of changes to obstacle attributes, giving the obstacle model spatial, temporal, and semantic themes and ensuring the integrity of the spatiotemporal obstacle data. Recording changes to obstacle semantic attributes allows for backtracking of historical versions and clarifies the obstacles being monitored. This semantic objectification allows for further quantitative evaluation and safety classification, assisting in the rapid analysis of obstacle data and having significant practical significance for dynamic monitoring of airport clearance and strengthening obstacle management.

[0114] As an embodiment of the present invention, the method further includes:

[0115] According to a preset cycle, the data within the clearance restriction surface of the target airport in the semantic thematic space database is monitored, and the increase of semantic thematic data within the clearance restriction surface in the semantic thematic space database is used as the execution trigger condition of the semantic objectification process.

[0116] When the preset period is reached, the semantic thematic data of the latest phase within the clearance restriction surface of the target airport in the semantic thematic space database is obtained, and it is determined whether the obtained semantic thematic data has increased compared with the semantic thematic data of the previous phase in the semantic thematic space database. If the semantic thematic data within the clearance restriction surface in the semantic thematic space database has increased, the spatial association relationship between the increased semantic thematic data and the obstacle three-dimensional model is updated, and the semantically objectified obstacle three-dimensional model is updated with the semantic attribute information in the increased semantic thematic data.

[0117] S104: Spatially associating the updated semantically objectified 3D obstacle model with the semantically objectified 3D obstacle model before the update, performing a 3D spatial information comparison and a semantic attribute comparison on the spatially associated 3D obstacle model to obtain updated change information of the semantically objectified 3D obstacle model.

[0118] The comparison of the three-dimensional obstacle models after spatial association includes spatial information comparison and semantic information comparison, that is, the changes in spatial information and semantic attributes must be detected.

[0119] As an embodiment of the present invention, Figure 8 As shown, the three-dimensional spatial information comparison of the spatially associated obstacle three-dimensional model includes:

[0120] S801: Perform overlay analysis on the updated semantically objectified obstacle 3D model and the semantically objectified obstacle 3D model before the update to establish an associated 3D model.

[0121] Overlay analysis is used to associate different versions of 3D obstacle model data based on spatial characteristics. Overlay analysis is a common method used by geographic information systems to extract spatial association information. Overlay analysis generates new layers, with some features of the input layer being split by the boundaries of the overlay layer. The new data layer generated by overlay analysis integrates all the characteristics of the features in two or more original layers, generating new spatial relationships and updating the attribute relationships of all layers.

[0122] The association relationship between elements in different versions of obstacle 3D models is realized through overlay analysis.

[0123] In this embodiment, the "3D Intersection (Intersect3D)" tool of the ArcGIS software platform can be used to calculate the intersection of three-dimensional polyhedral elements, implement overlay analysis of three-dimensional models, and obtain correlation relationships.

[0124] S802: Perform geometric feature information comparison on the associated three-dimensional models to obtain a geometric information comparison result.

[0125] In this embodiment, the geometric feature information mainly includes: volume, surface area, bottom elevation, net height, centroid coordinates, bottom four-dimensional coordinates, bottom area, etc.

[0126] The above overlay analysis was used to establish an associated three-dimensional model, and the above geometric feature information was extracted using ArcGIS software. By comparing the geometric feature information, it was determined whether the spatial information had changed.

[0127] The judgment rules for spatial information changes include: three-dimensional model height judgment, three-dimensional model coincidence degree judgment, relative surface area judgment, centroid distance judgment, and symmetric difference judgment.

[0128] Specifically, for the three-dimensional model height judgment: The height change of the obstacle three-dimensional model is the most frequent and has the greatest impact on the clearance safety. The obstacle height change detection includes the comparison of the net height of the obstacle and the elevation of the obstacle bottom surface.

[0129] The judgment formula is as follows: Definition: HB A is the elevation of the bottom surface of the three-dimensional model A; HB B is the elevation of the bottom surface of the three-dimensional model B; HR A is the net height of the three-dimensional model A; HR B is the net height of the three-dimensional model B; |HB A -HB B | is the absolute value of the bottom surface elevation difference; |HR A -HR B | is the absolute value of the net height difference; LD A is the diagonal length of the minimum circumscribed cuboid of the three-dimensional model A; LD B is the diagonal length of the minimum circumscribed cuboid of the three-dimensional model B; The three-dimensional model height change coefficient V1 is:

[0130]

[0131] It is obtained that 0 < V1 ≤ 1, and the smaller the value, the greater the change.

[0132] Specifically, for the three-dimensional model coincidence degree judgment: Definition: S A is the volume of the three-dimensional model A; S B is the volume of the three-dimensional model B; The Min function is used to return the minimum value among the given parameters; S A∩B is the volume of the overlapping and intersecting part of the three-dimensional models A and B. The three-dimensional model coincidence degree coefficient V2 is:

[0133]

[0134] It is obtained that 0 < V2 ≤ 1, and the smaller the value, the greater the change.

[0135] Specifically, for the relative surface area judgment: The relative surface area is the value obtained by dividing the surface area of the three-dimensional model by the volume. It reflects the degree of divergence of the geometric shape in space.

[0136] Definition: R A is the relative surface area of the three-dimensional model A; R B is the relative surface area of the three-dimensional model B; The Min function is used to return the minimum value among the given parameters; Then the three-dimensional model relative surface area ratio coefficient V3 is:

[0137]

[0138] It is obtained that \(0 < V3\leq1\). The smaller the value, the greater the change.

[0139] Specifically, for the centroid distance judgment: Definition: D AB is the distance between the centroids of the three-dimensional models A and B; LD A is the diagonal length of the minimum circumscribed cuboid of the three-dimensional model A; LD B is the diagonal length of the minimum circumscribed cuboid of the three-dimensional model B. The three-dimensional model height change coefficient V4 is:

[0140]

[0141] It is obtained that \(0 < V4\leq1\). The smaller the value, the greater the change.

[0142] Specifically, for the symmetric difference judgment: The symmetric difference refers to the volume of the non-overlapping parts of the two three-dimensional models A and B after aligning their centroids.

[0143] Definition: S A is the volume of the three-dimensional model A; S B is the volume of the three-dimensional model B; S A∪B is the volume of the union after aligning the centroids of the three-dimensional models A and B; S A∩B is the volume of the overlapping and intersecting part after aligning the centroids of the three-dimensional models A and B. The three-dimensional model symmetric difference coefficient V5 is:

[0144]

[0145] It is obtained that \(0 < V5\leq1\). The smaller the value, the greater the change.

[0146] Based on the above judgment rules, different weights are assigned, as follows S803.

[0147] S803. Assign weight values to the geometric information comparison results, perform three-dimensional space change detection, and obtain the three-dimensional space change results. Let the total number of judgment rules be n, the value of the i-th judgment rule be Vi, and the weight be Wi. Then the final change detection result calculated by combining multiple rules and weights is calculated as follows:

[0148]

[0149] It is obtained that \(0 < V AB \leq1\). The smaller the value, the greater the change.

[0150] Finally, compare the V AB result with the set threshold value to judge whether there is a change and obtain the part of the space change.

[0151] As one embodiment of the present invention, semantic attribute comparison is performed on the spatially associated 3D obstacle models. This includes comparing semantic attribute information to determine the similarity between the semantic attribute information. After obtaining associations between different versions of the 3D obstacle model data through 3D model overlay analysis, semantic analysis can be performed using common natural language processing (NLP) algorithms to determine whether the semantic information has changed.

[0152] S105 . Perform spatial intersection on the portion of the updated semantically objectified obstacle three-dimensional model corresponding to the change information and the clearance restriction surface of the target airport to obtain a superelevation detection result of the target airport.

[0153] As an embodiment of the present invention, S105 specifically includes:

[0154] A spatial intersection is performed on the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model with each clearance restriction surface of the target airport. If spatial intersection occurs, the maximum value of the obstacle's height exceeding each clearance restriction surface is calculated, and the obstacle three-dimensional model information and the corresponding clearance restriction surface information corresponding to the maximum value are used as the superelevation result of the target airport. Otherwise, the superelevation result of the target airport is that no obstacle exceeds the clearance restriction surface.

[0155] In an optional embodiment of the present invention, based on the above-mentioned S105, the height of each clearance restriction surface of the target airport can be reduced according to the height limit threshold to obtain a pre-restriction surface corresponding to each clearance restriction surface; the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model is spatially intersected with each pre-restriction surface; if spatial intersection occurs, the maximum value by which the obstacle exceeds the height of each pre-restriction surface is calculated, and the obstacle three-dimensional model information corresponding to the maximum value and the corresponding pre-restriction surface information are used as the pre-superelevation result of the target airport; otherwise, the pre-superelevation result of the target airport is that there is no obstacle exceeding the pre-restriction surface.

[0156] In the above embodiment, the change information corresponding to the updated semantically objectified obstacle three-dimensional model and each clearance restriction surface of the target airport is first obtained.

[0157] Based on the configured threshold for approaching the height limit, the height of each clearance restriction surface is reduced by a preset value. For example, if a threshold of 20 meters is considered approaching superelevation, the height of all clearance restriction surfaces at all target airports is reduced by 20 meters. The resulting reduced clearance restriction surface is called the pre-restriction surface. If an obstacle's 3D model exceeds the pre-restriction surface, the obstacle is considered superelevated or near superelevation.

[0158] The portion of the change information corresponding to the updated semantically objectified obstacle 3D model is spatially intersected with each clearance restriction surface of the target airport. The spatial intersection is a spatial intersection analysis between 3D polyhedra, and the geometric shape of the intersecting portion is obtained as a newly added 3D polyhedron.

[0159] Through the geometry of the intersection, obstacles that exceed the pre-limiting surface are screened out, along with the (altitude) elevation of the pre-limiting surface (the elevation is the highest elevation specified by the pre-limiting surface at that location), and the height difference (the elevation of the top surface of the obstacle minus the elevation of the pre-limiting surface).

[0160] When a pre-overheight obstacle is located in an overlapping area, the height limit is determined based on the limit surface with the lowest height limit. Therefore, the maximum height difference exceeded is further compared. For example, if the obstacle exceeds the takeoff and climb surface by 10 meters and the approach surface by 5 meters, the obstacle will be determined to be 10 meters above the takeoff and climb surface.

[0161] According to the above process, all pre-exceeded obstacle information can be obtained, including spatial information, three-dimensional geometric information, attribute information, semantic object information, pre-exceeded restriction surface name, pre-exceeded height difference, ground (altitude) elevation, obstacle height itself, obstacle top surface (altitude) elevation, etc.

[0162] Furthermore, superelevation obstacle information and near-superelevation obstacle information are distinguished from all pre-superelevation obstacle information.

[0163] In some embodiments, the above-mentioned super-high obstacle information and information close to super-high obstacle information can be loaded and displayed in a three-dimensional scene, and statistical charts can be generated and stored in a database, and the storage version can be marked. By distinguishing different storage versions through marking, historical data can be called for comparison and query to realize dynamic monitoring result display.

[0164] According to the embodiments of the present invention, by monitoring data changes in a multi-temporal remote sensing image spatial database and a semantic thematic spatial database, changes in obstacle three-dimensional models can be accurately identified and dynamically monitored.

[0165] The above is an introduction to a method embodiment. The following further illustrates the solution of the present invention through an apparatus embodiment.

[0166] like Figure 9 As shown, the apparatus 900 includes:

[0167] A clearance restriction surface generation module 910 is used to obtain parameters of the target airport site and generate a clearance restriction surface of the target airport;

[0168] The three-dimensional model generation module 920 of the obstacle includes a first initial database generation module 920-1 and a first database update module 920-2.

[0169] The first initial database generation module 920-1 is used to establish a multi-temporal remote sensing image spatial database, obtain the digital surface model and digital true orthogonal image within the clearance restriction surface, extract the obstacle surface contour within the clearance restriction surface, and generate an obstacle vector surface.

[0170] The first database updating module 920-2 is configured to monitor the multi-temporal remote sensing image spatial database, and if data within the clearance restriction surface in the multi-temporal remote sensing image spatial database is increased, extract the obstacle surface contour within the clearance restriction surface based on the increased digital surface model and digital true orthogonal image, and update the obstacle vector surface; calculate the ground elevation and top elevation of the updated obstacle vector surface, and generate a three-dimensional model of the obstacle;

[0171] The semantic object-based obstacle three-dimensional model generation module 930 includes a second initial database generation module 930 - 1 and a second database update module 930 - 2 .

[0172] The second initial database generation module 930-1 is used to establish a semantic thematic spatial database, obtain the semantic thematic data within the clearance restriction surface, establish a spatial association relationship between the semantic thematic data and the obstacle three-dimensional model, assign the semantic attribute information in the semantic thematic data to the corresponding obstacle three-dimensional model, and obtain a semantically objectified obstacle three-dimensional model.

[0173] The second database updating module 930-2 is configured to monitor the semantic thematic spatial database, and if semantic thematic data within the clearance restriction surface in the semantic thematic spatial database is increased, update the spatial association between the increased semantic thematic data and the three-dimensional obstacle model, and update the semantically objectified three-dimensional obstacle model using the semantic attribute information in the increased semantic thematic data;

[0174] An association and comparison module 940 is configured to spatially associate the updated semantically objectified 3D obstacle model with the semantically objectified 3D obstacle model before the update, compare the 3D spatial information and semantic attributes of the spatially associated 3D obstacle model, and obtain updated change information of the semantically objectified 3D obstacle model.

[0175] The superelevation detection module 950 is configured to perform spatial intersection between the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model and the clearance restriction surface of the target airport to obtain a superelevation detection result of the target airport.

[0176] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the described module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0177] According to an embodiment of the present invention, the present invention further provides an electronic device.

[0178] Figure 10 A schematic block diagram of an electronic device 1000 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0179] The device 1000 includes a computing unit 1001 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the device 1000 may also be stored in the RAM 1003. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0180] Various components in device 1000 are connected to I / O interface 1005, including an input unit 1006, such as a keyboard, mouse, etc.; an output unit 1007, such as various types of displays, speakers, etc.; a storage unit 1008, such as a magnetic disk, optical disk, etc.; and a communication unit 1009, such as a network card, modem, wireless communication transceiver, etc. The communication unit 1009 allows device 1000 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0181] The computing unit 1001 may be a variety of general-purpose and / or specialized processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The computing unit 1001 performs the various methods and processes described above, such as methods S101 to S105. For example, in some embodiments, methods S101 to S105 may be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as a storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 1000 via the ROM 1002 and / or the communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the methods S101 to S105 described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to execute methods S101 to S105 in any other appropriate manner (eg, by means of firmware).

[0182] Various implementations of the systems and techniques described above in this document can be realized in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof.

[0183] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for dynamic monitoring of airport clearance based on parameterized automatic modeling, characterized in that: include: Obtain the target airport site parameters and generate the target airport's clearance restriction surface; Establishing a multi-temporal remote sensing image spatial database, obtaining a digital surface model and a digital true orthogonal image within the clearance restriction surface, extracting the outline of the obstacle surface within the clearance restriction surface, and generating an obstacle vector surface; monitoring the multi-temporal remote sensing image spatial database, and if the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database increases, extracting the outline of the obstacle surface within the clearance restriction surface based on the increased digital surface model and digital true orthogonal image, and updating the obstacle vector surface; calculating the ground elevation and top elevation of the updated obstacle vector surface, and generating a three-dimensional model of the obstacle; Establishing a semantic thematic spatial database, obtaining semantic thematic data within the clearance restriction surface, establishing a spatial association relationship between the semantic thematic data and the three-dimensional obstacle model, assigning semantic attribute information in the semantic thematic data to the corresponding three-dimensional obstacle model, and obtaining a semantically objectified three-dimensional obstacle model; monitoring the semantic thematic spatial database, and if the semantic thematic data within the clearance restriction surface in the semantic thematic spatial database increases, updating the spatial association relationship between the increased semantic thematic data and the three-dimensional obstacle model, and updating the semantically objectified three-dimensional obstacle model with the semantic attribute information in the increased semantic thematic data; spatially associating the updated semantically objectified 3D obstacle model with the semantically objectified 3D obstacle model before the update, and performing 3D spatial information comparison and semantic attribute comparison on the spatially associated 3D obstacle model to obtain updated change information of the semantically objectified 3D obstacle model; Performing spatial intersection of the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model with the clearance restriction surface of the target airport to obtain a superelevation detection result of the target airport; The step of calculating the ground elevation and top elevation of the updated obstacle vector plane to generate a three-dimensional model of the obstacle includes: Filtering the added digital surface model to obtain digital elevation data, superimposing the obstacle vector surface with the digital elevation data, calculating the ground elevation of each vertex on each obstacle vector surface, and obtaining the ground elevation of the obstacle vector surface; superimposing the obstacle vector surface onto the added digital surface model, calculating the top surface elevation of each vertex on the obstacle vector surface, and obtaining the top surface elevation of the obstacle vector surface; Obtaining height information of the obstacle vector plane according to the ground elevation and top elevation of the obstacle vector plane; A three-dimensional model of the obstacle is established based on the obstacle vector surface and its ground elevation, top elevation and height information.

2. The method according to claim 1, characterized in that The step of obtaining the target airport site parameters and generating the target airport's clearance restriction surface includes: Acquiring target airport site parameters, and obtaining size and slope information of a clearance restriction surface corresponding to the target airport site parameters; Calculating the three-dimensional spatial coordinate point position of each clearance restriction surface based on the size and slope information of the clearance restriction surface; The three-dimensional spatial coordinate points of each clearance restriction surface are connected to form a three-dimensional polyhedron to obtain vector data of the three-dimensional polyhedron.

3. The method according to claim 1, wherein The step of extracting the obstacle surface contour within the clearance restriction surface and generating an obstacle vector surface includes: Noise filtering is performed on the added digital surface model, and obstacle contours are extracted based on elevation information to obtain obstacle terrain feature contour data; and Extract the texture features of the added digital true-orthogonal image, input it into the deep learning model, and output the obstacle texture feature contour data; Performing vector surface fusion on the obstacle terrain feature contour data and the obstacle texture feature contour data to obtain a fused vector surface; Topological checking and correction are performed on the fused vector surface to obtain the obstacle vector surface.

4. The method according to claim 1, wherein The three-dimensional spatial information comparison of the spatially associated three-dimensional obstacle model includes: Performing overlay analysis on the updated semantically objectified obstacle 3D model and the semantically objectified obstacle 3D model before the update to establish a correlated 3D model; Performing geometric feature information comparison on the associated three-dimensional models to obtain a geometric information comparison result; A weight value is assigned to the geometric information comparison result, and three-dimensional space change detection is performed to obtain a three-dimensional space change result.

5. The method according to claim 1, wherein Performing spatial intersection of the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model with the clearance restriction surface of the target airport to obtain a superelevation result of the target airport, including: Performing spatial intersection between the portion of the updated semantically objectified obstacle three-dimensional model corresponding to the change information and each clearance restriction surface of the target airport; If the spaces intersect, the maximum value of the obstacle's height exceeding each clearance restriction surface is calculated, and the three-dimensional obstacle model information and the corresponding clearance restriction surface information corresponding to the maximum value are used as the superelevation result of the target airport; otherwise, the superelevation result of the target airport is that there is no obstacle exceeding the clearance restriction surface.

6. The method according to claim 5, characterized in that Also includes: According to the height limit threshold, the height of each clearance restriction surface of the target airport is reduced to obtain a pre-restriction surface corresponding to each clearance restriction surface; Performing spatial intersection between the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model and each pre-constrained surface; If the spaces intersect, the maximum value of the obstacle's height exceeding each pre-restricted surface is calculated, and the three-dimensional obstacle model information and the corresponding pre-restricted surface information corresponding to the maximum value are used as the pre-superelevation result for the target airport. Otherwise, the pre-superelevation result for the target airport is that there is no obstacle exceeding the pre-restricted surface.

7. The method according to claim 1, wherein Also includes: Monitoring data within the clearance restriction surface of the target airport in the multi-temporal remote sensing image spatial database according to a preset period, and using an increase in data within the clearance restriction surface in the multi-temporal remote sensing image spatial database as an execution trigger condition; as well as According to a preset cycle, the data within the clearance restriction surface of the target airport in the semantic thematic space database is monitored, and the increase of the semantic thematic data within the clearance restriction surface in the semantic thematic space database is used as an execution trigger condition.

8. An airport clearance dynamic monitoring device based on parametric automatic modeling, characterized in that: include: A clearance restriction surface generation module is used to obtain the parameters of the target airport site and generate the clearance restriction surface of the target airport; The three-dimensional obstacle model generation module is used to establish a multi-temporal remote sensing image spatial database, obtain a digital surface model and a digital true orthogonal image within the clearance restriction surface, extract the obstacle surface contour within the clearance restriction surface, and generate an obstacle vector surface; monitor the multi-temporal remote sensing image spatial database, and if the data within the clearance restriction surface in the multi-temporal remote sensing image spatial database increases, extract the obstacle surface contour within the clearance restriction surface based on the increased digital surface model and digital true orthogonal image, and update the obstacle vector surface; calculate the ground elevation and top surface elevation of the updated obstacle vector surface to generate a three-dimensional model of the obstacle; A semantically objectified three-dimensional obstacle model generation module is configured to establish a semantic thematic spatial database, obtain semantic thematic data within the clearance restriction surface, establish a spatial association relationship between the semantic thematic data and the three-dimensional obstacle model, assign semantic attribute information in the semantic thematic data to the corresponding three-dimensional obstacle model, and obtain a semantically objectified three-dimensional obstacle model; monitor the semantic thematic spatial database, and if the semantic thematic data within the clearance restriction surface in the semantic thematic spatial database increases, update the spatial association relationship between the increased semantic thematic data and the three-dimensional obstacle model, and update the semantically objectified three-dimensional obstacle model with the semantic attribute information in the increased semantic thematic data; an association and comparison module for spatially associating the updated semantically objectified 3D obstacle model with the semantically objectified 3D obstacle model before the update, performing a 3D spatial information comparison and a semantic attribute comparison on the spatially associated 3D obstacle model, and obtaining updated change information of the semantically objectified 3D obstacle model; an ultra-elevation detection module, configured to perform spatial intersection between the portion of the change information corresponding to the updated semantically objectified obstacle three-dimensional model and the clearance restriction surface of the target airport to obtain an ultra-elevation detection result of the target airport; The step of calculating the ground elevation and top elevation of the updated obstacle vector plane to generate a three-dimensional model of the obstacle includes: Filtering the added digital surface model to obtain digital elevation data, superimposing the obstacle vector surface with the digital elevation data, calculating the ground elevation of each vertex on each obstacle vector surface, and obtaining the ground elevation of the obstacle vector surface; superimposing the obstacle vector surface onto the added digital surface model, calculating the top surface elevation of each vertex on the obstacle vector surface, and obtaining the top surface elevation of the obstacle vector surface; Obtaining height information of the obstacle vector plane according to the ground elevation and top elevation of the obstacle vector plane; A three-dimensional model of the obstacle is established based on the obstacle vector surface and its ground elevation, top elevation and height information.

9. An electronic device comprising at least one processor; and A memory communicatively connected to the at least one processor; characterized in that The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

Images