Building height extraction method and system based on oblique photogrammetry model

Abstract

The invention relates to the technical field of three-dimensional modeling, and discloses a building height extraction method and system based on an oblique photogrammetry model, which can significantly improve the precision and efficiency of extracting the building height from an OSGB oblique photogrammetry model, and realize automatic accurate identification and height calculation of a building main body structure. Through fusion of space geometric feature analysis and a deep learning model, the problems of reflection interference of a glass curtain wall, misjudgment of a complex roof structure and misreckoning of accessory structures such as a balcony and an equipment room into the height of a main body are effectively solved, and the reliability of a main body height extraction result is ensured. Meanwhile, subjective links such as manual measurement, manual outline sketching and empirical parameter adjustment are completely abandoned, personal error accumulation is greatly reduced, and the processing speed is remarkably increased.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of three-dimensional modeling, and particularly relates to a building height extraction method and system based on oblique photogrammetry model. BACKGROUND

[0002] With the acceleration of urbanization, three-dimensional building models are increasingly widely used in the fields of urban planning, smart city construction and digital twinning. Building height, as an important parameter of three-dimensional models, directly affects the effect of subsequent analysis and application. Existing building height acquisition techniques mainly include manual measurement, single image analysis, point cloud segmentation and parameterized modeling methods, but these methods have different degrees of limitations and defects in practical application.

[0003] Manual measurement of building height usually relies on full-process manual operation. The measurement personnel need to carry professional equipment to the building site, and measure key points (such as the highest point of the roof and the eave position) one by one through a total station or a laser range finder. After obtaining the basic data, the technical personnel manually draw the building contour line in the CAD software, and then calculate the overall height combined with the measurement data. This method has a fundamental defect: it is difficult to contact measure high-rise buildings in densely populated urban areas, and the reflection of glass curtain walls easily leads to ranging deviation; it is difficult to accurately restore complex roof structures (such as the intersection of sloping surfaces) during manual drawing; at the same time, it is limited by weather and building access, and the measurement of a single building may take several hours, which cannot meet the time efficiency requirements of regional projects.

[0004] The single image analysis method analyzes the building height through aerial images. After constructing a three-dimensional model using Agisoft and other software, the position offset of the same building in adjacent aerial images is identified through image feature matching algorithms (such as SIFT operators), and the height value is calculated according to the principle of triangulation. There are obvious defects in the implementation: when the building surface is a glass curtain wall, the strong reflection causes the feature point matching to fail; for buildings with complex roof shapes (such as curved domes or dense skylights), the algorithm misjudges the feature points as roof structure points, resulting in overestimation or underestimation of the building main body height. More seriously, this method completely ignores the three-dimensional spatial structure of the building, and cannot distinguish between roof equipment (cooling towers, elevator rooms) and the building main body, resulting in serious distortion of the height data.

[0005] The point cloud segmentation method relies on laser point cloud to segment building structures. First, a ground laser scanner is used to collect building point clouds, and then a region growing algorithm is used to automatically separate the roof, wall and auxiliary structure. In actual operation, the difference in point cloud density causes the algorithm to fail - the sparse point cloud at the top of the low-rise building leads to incorrect roof boundary identification; the point cloud at the bottom of the high-rise building is incomplete due to nearby occlusion; and the point cloud in the vegetation-covered area penetrates to form "ghost points". The technical personnel are forced to manually repair the point cloud holes in tools such as CloudCompare, and redefine the building boundary in an artificial box selection manner. This poor repair fit causes the connection between the main building and the auxiliary structure to be damaged, resulting in a defective building model.

[0006] The parametric modeling method generates buildings through preset rules. In platforms such as Cityengine, the building base contour is imported, and the building block is extruded in bulk by combining with the preset height parameter (such as the number of floors x floor height). This way of deviating from real scene data has systematic defects: the top details of the building (such as the roof and parapet) are simplified to a uniform height value; the special shape of the historical building is completely lost; when the real scene conflicts with the preset rules (such as the bottom commercial floor height being greater than the standard floor), the model height deviates from the actual value by more than 20%. Engineers need to manually correct each building against aerial images, which improves efficiency by only about 30% compared to full manual modeling, but the precision is greatly reduced.

[0007] In addition, the existing technology also has the following problems: aerial images and three-dimensional point clouds are processed separately, lacking of collaborative fusion; manual splicing of roof and main body height often causes misalignment, damaging the integrity of the building; multiple links rely on manual operation, the measurement and drawing process is affected by subjective experience, complex structures are difficult to accurately restore, and errors are continuously accumulated; local building changes require re-shooting and rebuilding the entire area, causing serious waste of computing resources and time; the recognition ability of key features is insufficient, and a single data source cannot accurately identify glass curtain walls, roof equipment, etc., leading to height misjudgment and model distortion.

[0008] In summary, the existing building height extraction technology cannot meet the needs of high precision and high efficiency in complex urban environments, and a more accurate new intelligent computing method is urgently needed. SUMMARY

[0009] The present application provides a building height extraction method and system based on oblique photogrammetry model to solve the problem that the existing technology cannot meet the needs of high precision and high efficiency in complex urban environments. To solve the above technical problems, the embodiments of the present application disclose the following technical solutions: One aspect of the present application provides a building height extraction method based on an oblique photogrammetry model, comprising: obtaining contour line vector data and an OSGB oblique photogrammetry model of a target building; selecting, based on the contour line vector data, associated patches representing the target building in the OSGB model; For each associated patch, a structure type is determined according to a normal direction, and the building structure types include a horizontal floor, an inclined surface and a wall surface; According to the building structure type of the associated patch and the spatial attribute data, a subject patch of the target building is determined based on a pre-trained subject determination model; The patch elevation data is subjected to cluster analysis, and all subject patches are divided into multiple layers; The elevations of the lowest layer and the highest layer are calculated respectively, and the building height of the target building is obtained according to the elevation difference between the two.

[0010] Optionally, the selecting, based on the contour line vector data, associated patches representing the target building in the OSGB model comprises: Extending outward along the normal direction of the contour line vector by a preset length to generate a closed buffer boundary; Selecting, in the OSGB model, a grid patch whose centroid coordinate is projected within the buffer boundary as an associated patch corresponding to the target building.

[0011] Optionally, the determining, for each associated patch, a building structure type according to a normal direction comprises: The normal-zenith angle of each associated patch is calculated according to the following formula : wherein, are the components of the normal vector of the associated patch in x, y and z directions respectively; is the modulus of the normal vector; the building structure type of the associated patch with is determined as a horizontal floor; the building structure type of the associated patch with is determined as an inclined surface; the building structure type of the associated patch with is determined as a wall surface.

[0012] Optionally, the method further comprises: The subject determination model is trained in the following manner: A training sample set is constructed based on the network patches in the OSGB model, and the sample attributes include spatial attribute data, building structure type and determination result of the network patches, wherein the spatial attribute data includes centroid coordinates and normal vector data, and the determination result includes building subject and non-building subject; The network patches are taken as nodes, and the edges shared between the network patches are taken as connection line segments to construct a graph structure; The model structure is constructed in the following manner: The first and second graph convolution layers adopt a 3*3 equivalent receptive field convolution kernel to extract local geometric features and neighborhood relationships in the graph structure, and form a high-dimensional feature mapping; The third graph convolution layer introduces a skip connection mechanism to fuse shallow detail features and deep abstract features, and performs enhancement processing on the building edge; The features output by the graph convolution layer are compressed to a single dimension through a fully connected layer, and a Sigmoid activation function is used for mapping to output the confidence results of the facets belonging to the building main body.

[0013] Optionally, according to the building structure type and spatial attribute data of the associated facets, the main body facet of the target building is determined based on a pre-trained main body determination model, including: The spatial attribute data of each associated facet is obtained, including the coordinates (x, y, z) of the facet centroid and the components of the facet normal vector in the x, y, and z directions ; The building structure type and spatial attribute data of each associated facet are input into the main body determination model to obtain the corresponding confidence results; A main body mask is constructed based on the facets with confidence results exceeding a preset threshold; The main body mask is optimized; The main body facet is determined using the optimized main body mask.

[0014] Optionally, the optimization of the main body mask includes: Holes in the main body mask are filled using a morphological closing operation; Noise in the main body mask is removed using a connected component analysis method.

[0015] Optionally, the clustering analysis of the facet elevation data divides all main body facets into multiple layers, including: For each main body facet, the average value of the three vertices in the elevation direction is calculated as the initial elevation data; Adjacent facet elevation features are fused to obtain the fusion elevation data of each main body facet; Adjacent main body facets with fusion elevation data difference less than a preset difference are merged into a facet cluster; Facet clusters with building structure type as horizontal floor and projection area less than a preset projection area are removed; Different facet clusters are divided into different layers, and each layer is sorted in ascending order according to the fusion elevation data value of the facets in the facet cluster.

[0016] Optionally, the fusion of adjacent facet elevation features to obtain the fusion elevation data of each main body facet includes: For each main surface patch, the initial elevation data is weighted and averaged with the initial elevation data of the adjacent surface patch, and the calculation result is taken as the fusion elevation data of the main surface patch.

[0017] Optionally, the elevation of the lowest layer and the highest layer are calculated respectively, and the building height of the target building is obtained according to the elevation difference between the two, comprising: Based on the initial elevation data of each main surface patch in the lowest layer, the weighted median algorithm is used to calculate the elevation of the lowest layer; The maximum initial elevation data of the main surface patch contained in the highest layer is taken as the elevation of the highest layer; The difference between the highest layer elevation and the lowest layer elevation is calculated as the building height of the target building.

[0018] Another aspect of the present application provides a building height extraction system based on a tilt photogrammetry model, which applies the building height extraction method based on the tilt photogrammetry model provided in the foregoing aspect.

[0019] The summary section is provided to introduce selected concepts of the disclosure in a simplified form, which will be further described below in the detailed description. The summary section is not intended to identify key or essential features of the disclosure, nor is it intended to limit the scope of the disclosure.

[0020] The building height extraction method and system based on the tilt photogrammetry model disclosed in the present application can significantly improve the precision and efficiency of extracting the building height from the OSGB tilt photogrammetry model, and realize the automatic and accurate identification and height calculation of the main structure of the building. By fusing spatial geometric feature analysis and deep learning model, the problems of glass curtain wall reflection interference, complex roof structure misjudgment, and balcony, equipment room and other accessory structure miscounting into the main height are effectively overcome, ensuring the reliability of the main height extraction result. At the same time, completely abandoning subjective links such as manual measurement, manual contouring and experience parameter adjustment, greatly reducing the accumulation of human error and significantly improving the processing speed. BRIEF DESCRIPTION OF DRAWINGS

[0021] The above and other objects, features and advantages of the present disclosure will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which like reference characters refer to the like parts throughout the several views, and in which the exemplary embodiments of the present disclosure are shown.

[0022] Figure 1 A flowchart of a building height extraction method based on a tilt photogrammetry model provided by an embodiment of the present application is shown in the figure; Figure 2 A flowchart of step S200 in the embodiment of the present application is shown in the figure; Figure 1 ​Figure 3 An implementation provided for an embodiment of the present application Figure 1 A flowchart of step S400 in the implementation provided for an embodiment of the present application is shown in FIG. 4. Figure 4 An implementation provided for an embodiment of the present application Figure 1 A flowchart of step S500 in the implementation provided for an embodiment of the present application is shown in FIG. 5. Figure 5 An implementation provided for an embodiment of the present application Figure 1 A flowchart of step S600 in the implementation provided for an embodiment of the present application is shown in FIG. 6. DETAILED DESCRIPTION

[0023] Embodiments of the present disclosure will be described in more detail with reference to the drawings. Although embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure is more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0024] The term "comprising" and variations thereof as used herein are intended to mean "including but not limited to". The term "or" as used herein is intended to mean "and / or". The term "based on" means "based, at least in part, on". The terms "one example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. can refer to different or same objects. Other explicit or implicit definitions can also be included below.

[0025] Figure 1 A flowchart of a building height extraction method based on a oblique photogrammetry model provided for an embodiment of the present application is shown in FIG. 1, which includes the following steps: Figure 1 Step 100: Obtain the contour line vector data of the target building and the OSGB oblique photogrammetry model.

[0026] First, obtain the contour line vector data of the target building. The data source can be the building base, as-built drawing vectorization results provided by the real estate / planning department, or obtained through orthophoto / 3D scene manual digitization. The format includes SHP, GeoJSON, GDB, etc. This embodiment uses the first floor outer wall contour as the reference contour line vector data.

[0027] OSGB is a 3D scene data format, which is usually generated by a drone or aerial camera from multiple angles (orthophoto + oblique) to take images, through aerial triangulation, dense matching and 3D reconstruction algorithm.

[0028] ​After obtaining the OSGB oblique photogrammetry model of the target building, the contour line vector is superimposed into the OSGB model, so that they have the same coordinate system and are registered, and it is ensured that the target building is matched and corresponding in both.

[0029] Step 200: selecting the associated facet representing the target building in the OSGB model based on the contour line vector data.

[0030] In one embodiment of the present disclosure, as shown in Figure 2 Step S200 can be implemented in the following manner: Step S201: expanding outward along the normal direction of the contour line vector by a preset length to generate a closed buffer boundary.

[0031] The outward expansion operation is performed along the normal direction of the contour line to generate a closed buffer boundary. The preset length of the outward expansion can be set according to the actual situation of the building, for example, 5-10 meters, forming a three-dimensional action range similar to the building contour, which can effectively cover the real facade of the target building and avoid missing selection due to modeling errors or occlusion.

[0032] Step S202: selecting the grid facet in the OSGB model whose centroid coordinate is projected within the buffer boundary as the associated facet corresponding to the target building.

[0033] The centroid coordinate of each triangular grid facet in the OSGB model is projected to the ground plane, and it is judged whether it falls within the buffer range generated in step S201. If the centroid projection point is located within the buffer boundary, the facet is selected as the associated facet corresponding to the target building.

[0034] Since the OSGB model contains a large number of scene elements, such as adjacent buildings, roads, vegetation, etc., through the above manner, non-target building facets (such as other buildings, streets, greenery or vehicles) outside the buffer zone can be effectively removed, ensuring that the grid data used in the subsequent processing steps mainly comes from the target building itself.

[0035] Step 300: determining the structure type according to the normal direction for each associated facet.

[0036] This step further determines the corresponding building structure type of each associated facet selected from the OSGB model through its normal direction. The normal vector of the facet is an important geometric property describing its orientation, which can intuitively reflect the relative inclination relationship of the facet with the ground. Building surfaces are mainly composed of three types of components: horizontal floors (such as floor slabs, roof slabs), vertical walls (such as facades, curtain walls), and structures with a certain angle of inclination (such as pitched roofs, balcony floorboards, etc.). Therefore, based on the classification of the normal direction, fine annotation of the geometric features of the building can be achieved.

[0037] In one embodiment of the present disclosure, step S300 can be implemented in the following way: The normal zenith angle of each associated facet is calculated according to the following formula wherein, are the components of the normal vector of the associated facet in the x, y, z directions, respectively; is the norm of the normal vector.

[0038] If , the building structure type of the associated facet is determined as a horizontal floor.

[0039] If , the building structure type of the associated facet is determined as an inclined surface.

[0040] If , the building structure type of the associated facet is determined as a wall surface.

[0041] For example, if the normal vector of a facet is n=(0.05, 0.02, 0.99), it is calculated that , the facet should be marked as a horizontal floor; if the normal vector of a facet is n=(0.5, 0.1, 0.86), , it is determined as an inclined component; and if the normal is n=(0.98, 0.05, 0.1), , it is close to vertical and marked as a wall surface.

[0042] Step 400: determining the subject facet of the target building based on the building structure type and spatial attribute data of the associated facet and a pre-trained subject determination model; In one embodiment of the present disclosure, in order to realize the automatic distinction between the building subject and the non-subject component, a subject determination model based on graph convolution network (GCN) is designed and trained.

[0043] The subject determination model is trained in the following way: (1) In the data preparation stage, network facets are obtained based on the OSGB oblique photogrammetry model, and a training sample set is constructed. Each facet sample contains spatial attribute data and corresponding structure type label. Among them, the spatial attribute data includes the facet centroid coordinates (x, y, z) and the normal vector , the former represents the position of the facet in space, and the latter describes the orientation and geometric characteristics of the facet. At the same time, the sample is assigned a determination result label, i.e. building subject or non-building subject.

[0044] ​(2) In the graph structure construction stage, the network patches are abstracted as nodes in the graph, and if there is a shared boundary between the patches, a connection relationship is established between the corresponding nodes to construct a graph structure. This graph structure can explicitly model the spatial adjacency relationship between the patches.

[0045] (3) The model structure is constructed in the following manner: The first and second graph convolution layers adopt a 3*3 equivalent receptive field convolution kernel to extract local geometric features and neighborhood relationships in the graph structure, forming a high-dimensional feature mapping.

[0046] The third graph convolution layer introduces a skip connection mechanism to fuse shallow detail features and deep abstract features for enhanced processing of building edges.

[0047] The features output by the graph convolution layer are compressed to a single dimension by a fully connected layer, and a Sigmoid activation function is used for mapping to output the confidence result of the patch belonging to the building main body.

[0048] The present application adopts a multi-layer graph convolution network for feature extraction, and the first two graph convolution layers are set to a 3*3 equivalent receptive field convolution kernel to extract local geometric features in the graph structure and capture spatial relationships between adjacent patches. For example, a certain wall patch is adjacent to a floor patch, and after convolution aggregation, the model can perceive the "wall-plate" topological structure feature. The third graph convolution layer introduces a skip connection mechanism to realize the fusion of shallow detail features and deep abstract features, which can enhance the edge determination effect and avoid the loss of detail information in deep learning.

[0049] Finally, the high-dimensional features output by the graph convolution layer are compressed to a single dimension value by a fully connected layer, and a Sigmoid activation function is used for normalized mapping to obtain the confidence value of each patch belonging to the building main body. The confidence value is in the interval [0, 1], and the closer the value is to 1, the more likely the patch is a building main body.

[0050] The main body determination model can fully utilize the spatial geometric properties and graph structure features of the OSGB model, combine the graph convolution and skip connection mechanisms in deep learning, and realize high-precision building main body recognition, which not only effectively eliminates interference components such as balconies and canopies, but also provides reliable input data for subsequent floor height reconstruction and height measurement.

[0051] In one embodiment disclosed in the present application, as shown in Figure 3 the following method can be used to implement step S400: Step S401: Obtain the spatial attribute data of each associated patch, including the coordinates (x, y, z) of the patch centroid, and the components of the patch normal vector in the x, y, and z directions .

[0052] Step S402: input the building structure type and spatial attribute data of each associated facet into the subject determination model to obtain the corresponding confidence result.

[0053] The building structure type and spatial attribute of the facet constitute a seven-dimensional feature. The seven-dimensional feature of each associated facet is input into the subject determination model. The model outputs a confidence value between 0 and 1, which represents the possibility that the facet belongs to the building subject. For example, a load-bearing floor may output 0.93, while a balcony facet may output 0.25.

[0054] Step S403: construct a subject mask based on the facets whose confidence results exceed a preset threshold.

[0055] According to the comparison between the confidence result and the preset threshold (≥ 0.8), the facet set higher than the threshold is selected and marked as a subject facet, thereby preliminarily generating mask data of the building subject. The subject mask is essentially a binary distribution map for indicating which areas belong to the subject, wherein 1 represents the building subject and 0 represents the non-building subject.

[0056] Step S404: optimize the subject mask.

[0057] In one embodiment of the present disclosure, the optimization of the subject mask includes: (1) The morphological closing operation is used to repair the structure of the subject mask. The closing operation is composed of dilation operation and erosion operation in sequence, and a 3x3 structural element kernel is used in this embodiment. Through this operation, the holes and small cracks in the interior of the subject area can be effectively filled, and the gaps in the subject area caused by model determination errors or edge discontinuity can be avoided, so that the mask of the building subject is more complete and coherent.

[0058] (2) The connected domain analysis method is used to remove noise in the subject mask. The mask area is labeled by connected components, the area or pixel number of each connected region is calculated, and the isolated regions smaller than the preset threshold are deleted. These isolated noises usually come from balconies, canopies or other decorative components, and do not have the structural continuity of the subject. By removing these interference regions through connected domain analysis, the purity and accuracy of the subject mask can be significantly improved.

[0059] Step S405: determine the subject facet using the optimized subject mask.

[0060] The main body mask is corresponded to the associated facets in the original OSGB model one by one. For each associated facet, if the centroid coordinate falls in the main body mask area, it is determined that the facet belongs to the building main body; if the centroid coordinate is outside the mask, it is determined as a non-main body facet and is removed. In this way, the main body screening at the facet level is realized.

[0061] The finally output main body facet set represents the part of the target building that has a structural meaning, for example, the core tube of the podium, the standard layer wall, the floor and the roof of the residence, etc., and the decorative or accessory components (such as railings, balcony floor, awnings) unrelated to the main body are all removed.

[0062] Step 500: Cluster analysis is performed on the facet elevation data, and all main body facets are divided into multiple layers.

[0063] In an embodiment of the present disclosure, as shown in Figure 4 Step S500 can be implemented in the following way: Step 501: For each main body facet, the average value of the three vertices in the elevation direction is calculated and used as the initial elevation data.

[0064] For any main body facet, the coordinates of its three vertices are extracted, which are (x1, y1, z1), (x2, y2, z2), and (x3, y3, z3).

[0065] The average value in the z direction is calculated by the following formula: z init =z1+z2+z3 / 3 The average value z initz is used as the initial elevation data of the facet, describing the overall position of the facet in the elevation direction.

[0066] Step 502: The elevation characteristics of adjacent facets are fused to obtain the fused elevation data of each main body facet.

[0067] In actual building surfaces, there is a certain degree of noise and local protrusions (such as external wall decoration, awnings, air conditioning sites, etc.), and if only the initial elevation data is used, it will lead to discontinuous floor division. Therefore, the elevations of adjacent facets need to be fused to improve the spatial consistency.

[0068] In an embodiment of the present disclosure, step 502 can be implemented in the following way: For each main body facet, the initial elevation data and the initial elevation data of adjacent facets are weighted and averaged, and the calculation result is used as the fused elevation data of the main body facet.

[0069] For each subject patch, the patch can be taken as the center to search its adjacent patches (i.e. 1-hop neighborhood sharing edges), and the initial elevation of the subject patch and the initial elevations of the adjacent patches are weighted and averaged to obtain the fusion elevation of the subject patch. The weight can be set according to the patch area, the length of the adjacent edge, or uniform distribution.

[0070] Step 503: merging adjacent subject patches with a fusion elevation data difference less than a preset difference value into a patch cluster.

[0071] After obtaining the smooth fusion elevation data, adjacent patches with an elevation difference less than a preset difference value (such as 1.0m) are merged to form a patch cluster.

[0072] (1) Elevation difference calculation: For each subject patch i, search its adjacent patch set , and calculate the fusion elevation difference between patch i and each adjacent patch j respectively: Wherein, and represent the fusion elevation data of patch i and patch j respectively.

[0073] (2) Merging condition: If the following conditions are met: Wherein, 1m is a preset difference value, then patch i and patch j are considered to belong to the same elevation level and can be merged into the same patch cluster.

[0074] Starting from any unmarked patch, recursively or iteratively search all adjacent patches that meet the above conditions, and classify these patches into the same cluster until all patches are assigned to a patch cluster.

[0075] Step 504: removing the patch cluster whose building structure type is horizontal floor and whose projection area is less than a preset projection area.

[0076] After the formation of the patch cluster, the effectiveness of the cluster needs to be screened to avoid the interference of non-main structure such as balcony, rain shed, decorative component on floor division.

[0077] Firstly, the building structure type of the patch in the cluster is judged, and if the cluster is mainly composed of horizontal floor patches, the projection area of the cluster is further calculated.

[0078] The projection area of the patch cluster is calculated as follows: Wherein, A is the total actual area in the patch cluster, ​The angle between the normal and the vertical direction.

[0079] The actual area of the face sheet is mapped to the horizontal plane according to the normal direction, so as to obtain the real coverage range of the face sheet on the building plan.

[0080] If the projected area of a certain face sheet cluster is less than a preset threshold (such as 15 m2), it is considered that it is not enough to represent an independent floor or an important part of a floor, and the face sheet cluster is excluded from the subsequent clustering results. Through this screening process, it can be ensured that the face sheet clusters participating in the floor division are truly related to the main body.

[0081] Step 505: Divide different face sheet clusters into different layers, and sort each layer in ascending order according to the fusion elevation data value of the face sheet in the face sheet cluster.

[0082] After removing the invalid clusters, the remaining valid face sheet clusters will be further divided into different building levels. According to the fusion elevation data of the face sheet in the cluster, first, the average fusion elevation of the face sheet in all face sheet clusters is calculated as the average fusion elevation value of the corresponding face sheet cluster; then, the face sheet clusters are arranged in ascending order according to the average fusion elevation value, and floor numbers are assigned from bottom to top, for example, the number 0 represents the ground floor, 1 represents the second floor, and so on. In this way, each face sheet cluster can correspond to a specific floor position of the building.

[0083] Step 600: Calculate the elevation of the lowest layer and the highest layer respectively, and obtain the building height of the target building according to the elevation difference between the two.

[0084] In one embodiment disclosed in the present application, as shown in Figure 5 Step S600 can be implemented in the following way: Step S601: Based on the initial elevation data of each main face sheet in the lowest layer, the weighted median algorithm is used to calculate the elevation of the lowest layer.

[0085] The initial elevation data of all main face sheets in the lowest layer is counted. In order to avoid the deviation of the overall elevation calculation caused by individual outliers (such as local depressions, external components, etc.), the weighted median algorithm is used for calculation. The specific process is as follows: the main face sheets in the lowest layer are sorted in ascending order according to their initial elevation values, and the area of the face sheet is introduced as a weight factor, and the areas of the face sheets are added in turn. When the cumulative area reaches 50% of the total area, the corresponding face sheet elevation is selected as the representative elevation of the lowest layer.

[0086] Step S602: Take the largest initial elevation data of the main face sheet contained in the highest layer as the elevation of the highest layer.

[0087] Step S603: calculating the difference between the highest elevation and the lowest elevation as the building height of the target building.

[0088] The building height H of the target building is calculated by the following formula: wherein, is the elevation of the highest floor, is the elevation of the lowest floor.

[0089] The embodiments of the present application further disclose a building height extraction system based on a tilt photogrammetry model, which applies the building height extraction method based on the tilt photogrammetry model disclosed in the foregoing embodiments, and details are not repeated here.

[0090] The above has described the embodiments of the present disclosure, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical application, or technical improvement to the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A building height extraction method based on an oblique photogrammetry model, characterized in that: include: Obtain the contour vector data and OSGB oblique photogrammetry model of the target building; Based on the contour vector data, the associated facets representing the target building are selected in the OSGB model; For each associated facet, a structural type is determined according to the normal direction, where the building structural types include horizontal floors, inclined surfaces, and walls. According to the architectural structure type and spatial attribute data of the associated facets, the main facets of the target building are determined based on the pre-trained subject determination model; Perform cluster analysis on the patch elevation data and divide all the main patches into multiple layers; The elevations of the lowest and highest floors are calculated respectively, and the building height of the target building is obtained based on the elevation difference between the two.

2. The method according to claim 1, characterized in that The step of selecting associated facets representing the target building in the OSGB model based on the contour vector data includes: Expanding a preset length along the normal direction of the contour line vector to generate a closed buffer zone boundary; In the OSGB model, the mesh patches whose centroid coordinates are projected within the buffer zone boundary are selected as the associated patches corresponding to the target building.

3. The method according to claim 1, characterized in that For each associated facet, the building structure type is determined according to the normal direction, including: Calculate the normal-zenith angle of each associated patch according to the following formula : in, are the components of the normal vector of the associated patch in the x, y, and z directions respectively; , is the modulus of the normal vector; Will The building structure type of the associated surface is determined to be a horizontal floor; Will The building structure type of the associated surface is determined to be an inclined surface; Will The architectural structure type of the associated surface is determined to be a wall.

4. The method according to claim 1, wherein The method further comprises: The subject judgment model is trained in the following way: Constructing a training sample set based on the network patch in the OSGB model, wherein the sample attributes include spatial attribute data of the network patch, building structure type, and determination results, wherein the spatial attribute data includes centroid coordinates and normal vector data, and the determination results include building bodies and non-building bodies; The network patches are used as nodes and the edges shared between the network patches are used as connecting segments to construct a graph structure. The model structure is constructed as follows: The first and second graph convolution layers use convolution kernels with 3×3 equivalent receptive fields to extract local geometric features and neighborhood relationships in the graph structure to form high-dimensional feature maps; The third convolutional layer introduces a skip connection mechanism to fuse shallow detail features with deep abstract features and enhance building edges. The features output by the graph convolution layer are compressed into a single dimension through a fully connected layer, mapped using the Sigmoid activation function, and the confidence result of whether the output face belongs to the building body is output.

5. The method according to claim 1 or 4, characterized in that According to the architectural structure type and spatial attribute data of the associated facets, the main facets of the target building are determined based on the pre-trained subject determination model, including: Get the spatial attribute data of each associated patch, including the coordinates of the patch's centroid (x, y, z) and the components of the patch's normal vector in the x, y, and z directions ; Input the building structure type and spatial attribute data of each associated face into the subject judgment model to obtain the corresponding confidence result; Constructing a subject mask based on the patches whose confidence results exceed a preset threshold; Optimize the main body mask; The optimized body mask is used to determine the body patch.

6. The method according to claim 5, characterized in that The optimizing process of the main body mask includes: Use morphological closing operation to fill the holes in the subject mask; The connected component analysis method is used to remove noise from the subject mask.

7. The method according to claim 1, characterized in that The face elevation data is clustered and divided into multiple layers, including: For each main facet, the average value of the three vertices in the elevation direction is calculated and used as the initial elevation data; Fuse the elevation features of adjacent patch faces to obtain the fused elevation data of each main patch; Merge adjacent main facets whose fused elevation data difference is less than a preset difference into a facet cluster; Remove the patch clusters whose building structure type is horizontal floor and whose projected area is smaller than the preset projected area; Different patch clusters are divided into different layers, and each layer is sorted in ascending order according to the fused elevation data values ​​of the patches in the patch cluster.

8. The method according to claim 7, characterized in that The step of fusing the elevation features of adjacent patch faces to obtain fused elevation data of each main patch includes: For each main facet, a weighted average calculation is performed on the initial elevation data and the initial elevation data of the adjacent facets, and the calculation result is used as the fused elevation data of the main facet.

9. The method according to claim 7, characterized in that The method of calculating the elevations of the lowest floor and the highest floor respectively and obtaining the building height of the target building according to the elevation difference between the two floors includes: Based on the initial elevation data of each main facet in the lowest layer, the weighted median algorithm is used to calculate the elevation of the lowest layer; The maximum initial elevation data among the main facets contained in the highest layer is used as the elevation of the highest layer; The difference between the highest floor elevation and the lowest floor elevation is calculated as the building height of the target building.

10. A building height extraction system based on an oblique photogrammetry model, characterized in that: Apply the building height extraction method based on the oblique photogrammetry model described in any one of claims 1 to 9.

Images