A method for displaying house types by integrating three-dimensional models with aerial oblique photography

Abstract

The present invention relates to the technical field of three-dimensional modeling, and specifically to a method for displaying house types by integrating a three-dimensional model with aerial oblique photography. The method comprises: extracting architectural outline features of a target house using aerial image data, and performing preliminary spatial registration on the architectural outline features of the target house; setting feature points based on the spatial position of the target house, performing house attribute verification using each feature point and the architectural outline features, extracting image partition information, dividing the structure of the target house using the image partition information, and obtaining an iterative segmentation result; setting the iterative segmentation result to multiple image units, and setting an extended progressive graph corresponding to the image units; counting overlapping parts of each image unit according to the extended progressive graph to obtain an amount of house type difference; performing three-dimensional stretching on the amount of house type difference to obtain a difference distribution of each house type, correcting the target house using the difference distribution of each house type, and recursively updating a display category of the target house, thereby improving the accuracy and efficiency of house display.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional modeling, and in particular to a method for displaying house types by integrating a three-dimensional model with aerial oblique photography. Background Art

[0002] Traditionally, 3D house modeling is primarily accomplished manually. Specifically, feature points are manually set on the floor plans of the house to be modeled; manual calibration is performed based on the feature points in the floor plans; and the 3D model of the house is generated using this manually calibrated data. Traditional methods are susceptible to the influence of the surrounding physical environments in the collected data, resulting in insufficient modeling accuracy. Furthermore, it is difficult to distinguish modeling differences between different houses, resulting in the need to generate 3D models multiple times during house display, reducing display efficiency and accuracy.

[0003] For example, Chinese patent publication number CN116580162A discloses a method, device, equipment and storage medium for establishing a three-dimensional model of a house. First, the hierarchical household diagrams of each layer of the house to be modeled are automatically calibrated. Specifically, the initial hierarchical household diagram of the bottom layer is used as the initial calibration basis. Then, for each current layer that is not the bottom layer, the initial hierarchical household diagram of the current layer is calibrated based on the target hierarchical household diagram of the reference layer corresponding to the current layer to obtain the target hierarchical household diagram of the current layer, wherein the basis for the calibration is: any line segment in the target hierarchical household diagram of the current layer coincides with the associated line segment of the line segment in the target hierarchical household diagram of the reference layer; by superimposing the target hierarchical household diagrams in sequence, the automatic superposition task of the hierarchical household data of the house to be modeled can be achieved; according to the attribute data of each layer, the target hierarchical household diagram of each layer is stretched to generate a three-dimensional model of the house to be modeled.

[0004] For example, Chinese patent publication number CN118608381A discloses a method for splicing point cloud data of a house type, comprising: segmenting the house type according to a preset station segmentation rule to obtain multiple measurement stations; constructing design feature data based on the design drawing of each measurement station and constructing scan feature data based on the point cloud data of the measurement station, then matching the design feature data and the scan feature data of each measurement station to obtain the house scaling ratio and the relative position relationship of the points in the point cloud data of each measurement station, and then using the relative position relationship of the measurement stations in the design drawing to perform point cloud splicing on the multiple measurement stations to obtain a point cloud model of the house type.

[0005] The prior art describes displaying layered and household-based images based on line segments on each floor, and combining house types based on the locations of measurement stations. These methods mainly display static models of houses and are unable to show the differences in the display of each house in the three-dimensional model due to changes over time or other forms. As a result, the display category of each house cannot be dynamically updated, which reduces the efficiency and relative accuracy of house display. Summary of the Invention

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is: a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography, including: S1, obtaining aerial image data of a target house, using the aerial image data to extract the architectural outline features of the target house, and performing preliminary spatial alignment on the architectural outline features of the target house.

[0007] S2 sets feature points based on the spatial location of the target house, uses each feature point to compare the house attributes with the building outline features, extracts multiple image partition information under oblique photography, divides the structure of the target house based on the image partition information, and determines the iterative segmentation results of the target house structure at adjacent positions.

[0008] S3, sets the iterative segmentation result to multiple graph units, queries the priority of the graph units, selects the graph units according to the priority, number of occurrences and area of the graph units, builds a connection channel between the selected graph units, and sets the extended progressive graph corresponding to the graph units.

[0009] S4, counting the overlapping parts of each graph unit according to the extended progressive graph to obtain the difference in apartment types.

[0010] S5, three-dimensionally stretching the apartment type difference to obtain the apartment type difference distribution, using the apartment type difference distribution to correct the three-dimensional point cloud data of the target house, and recursively updating the display category of the target house.

[0011] The beneficial effects of the present invention are: 1. After aligning the target house, the present invention uses feature points and image partition information to describe the differences in the structure of the current house and the iterative segmentation results, thereby quickly generating a house structure diagram and improving the reusability of the three-dimensional model when generating the house structure.

[0012] 2. The present invention sets the iterative segmentation results as multiple graph units, queries the priority of the graph units, and selects the graph units according to the priority, number of occurrences and area of the graph units; it can query the relationship between multiple iteratively segmented house structures, as well as the connection relationship between different house structures, and generate an extended progressive graph corresponding to the graph units, which can comprehensively display the relationship between the corresponding house structures and whether the house structures have changed within a preset period, reduce the problem of reduced accuracy of the house model in describing the house structure due to environmental occlusion during generation, and improve the display effect of the image of the house itself.

[0013] 3. The present invention constructs an extended progressive graph, counts the overlapping parts of each graph unit, and quickly calculates the difference in housing types through hierarchical division and overlapping data analysis. It can quickly quantify the difference in housing types and display each housing type in a different connection method. In addition to improving the accuracy of three-dimensional model setting, it also improves the comprehensiveness of house type display, facilitates tracking of house status, and prevents the occurrence of house risk situations. Finally, through the distribution of its housing type differences, the house display category is updated, which can describe the dynamic update effect of the house in disaster warning and other aspects, realize the real-time display of each house type, and adapt to the requirements of house type display in various scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below with reference to the accompanying drawings and examples.

[0015] Figure 1 The present invention is a flowchart of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography.

[0016] Figure 2 The present invention is a flowchart of step S1 of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography.

[0017] Figure 3 The present invention is a flow chart of step S2 of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography.

[0018] Figure 4 The present invention is a flowchart of step S3 of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography.

[0019] Figure 5 The present invention is a flow chart of step S4 of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography.

[0020] Figure 6 The present invention is a flowchart of step S5 of a method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography. DETAILED DESCRIPTION

[0021] The following embodiments of the present invention are described in detail. The embodiments described below are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention. Where specific techniques or conditions are not specified in the embodiments, the techniques or conditions described in the literature in the art or in the product specifications shall be followed.

[0022] See Figure 1 A method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography includes: S1, obtaining aerial image data of a target house, extracting architectural outline features of the target house using the aerial image data, and performing preliminary spatial registration on the architectural outline features of the target house.

[0023] S2 sets feature points based on the spatial location of the target house, uses each feature point to compare the house attributes with the building outline features, extracts multiple image partition information under oblique photography, divides the structure of the target house based on the image partition information, and determines the iterative segmentation results of the target house structure at adjacent positions.

[0024] S3, sets the iterative segmentation result to multiple graph units, queries the priority of the graph units, selects the graph units according to the priority, number of occurrences and area of the graph units, builds a connection channel between the selected graph units, and sets the extended progressive graph corresponding to the graph units.

[0025] S4, counting the overlapping parts of each graph unit according to the extended progressive graph to obtain the difference in apartment types.

[0026] S5, three-dimensionally stretching the apartment type difference to obtain the apartment type difference distribution, using the apartment type difference distribution to correct the three-dimensional point cloud data of the target house, and recursively updating the display category of the target house.

[0027] After obtaining the aerial image data of the target house, the aerial image data is preprocessed, the lens distortion is corrected using the camera calibration parameters, and the aerial image data is converted into a sparse point cloud under the same spatial coordinates. Then, for the target house corresponding to the sparse point cloud, the three-dimensional expression of the target house is determined.

[0028] During preliminary spatial calibration, multiple image control points and check points are extracted from the aerial image data based on the actual environment of the building. The spatial position of the image control points is used to perform preliminary spatial position calibration on the target house. At the same time, the check points are used to evaluate whether the calibrated data corresponds to the current house layout.

[0029] At this time, you can quickly verify the display results of the aerial image data by adjusting the image control points and check points. At the same time, you can further reduce the influence of obstructions such as vegetation on the accuracy of model generation by adjusting the position of the image control points on the ground. The check points will represent the feature points marked when the image is processed multiple times. These feature points will represent the main identification parts in the image processing. This part explains the focus and differences of the current house display, which makes it easier to identify whether the current house is a dangerous house, etc., and improve the accuracy of the house under multi-source integration.

[0030] like Figure 2 As shown, the implementation method of step S1 also includes: S11, determining the viewing angle of the aerial image data of the target house, dividing the aerial image data of the target house into multiple viewing angle partition information based on the various viewing angles during shooting, and deploying multiple image control points and check points in each viewing angle partition information; the viewing angle partition information represents the image information of the current target house at the corresponding viewing angle at each viewing angle. Image control points are generally located at the edge of the building outline or at significant feature points, such as corners, roof edges, ground markers, etc., where the house outline and relative position can be directly viewed. Furthermore, the set image control points select the environmental entities around the target house and the house outline to set image control points. The number of settings can be 48, so as to obtain sufficient information about the surrounding environment at the current location of the target house and identify the relative position of the target house and the surrounding area.

[0031] Verification points are generally located in key building structures, such as the centers of doors and windows, stair turning points, or areas with different apartment types, such as cracks and tilted walls. Verification points represent prominent points in the house. Directly identifying this location can help us understand the structural entity of the entire house relative to the surrounding environment. Generally, verification points are selected at points different from the image control points. The number of points selected can be 29 to obtain the integrity of the spatial alignment of the current target house under oblique shooting.

[0032] S12: Obtain the locations of image control points and check points for each viewing area partition information, and align each viewing area partition information according to the image control point locations. Position alignment calculates the distance between the pixels in the viewing area partition information and the image control points to ensure that the calculated distances are consistent across multiple viewing angles after angle changes, preventing partial positional offsets. When aligning the check points, the distance between the image control points and the check points is used to determine the relative position of the current house. This requires that the distance between the image control points and the check points is consistent across images captured from multiple viewing angles.

[0033] S13, using the aligned image control points and check points as the architectural outline features of the target house, completes the preliminary spatial alignment of the target house. When the calculated distances between the image control points and the check points are the same under multiple viewing angles, the alignment is considered complete. These points that can maintain relative spatial consistency during multiple calibrations will be used for subsequent data verification of the target house. For example, the feature points used subsequently will identify the spatial consistency distribution of each defect according to the defects of the target house, and then identify the locations of these defects to find the differences between the current target house and the standard house model, so that the house layout can be quickly displayed based on these differences.

[0034] In one embodiment of the present invention, the iterative segmentation result illustrates what partition intersection points exist at the structural position of the building outline of the target house and the pre-generated three-dimensional point cloud, to illustrate what forms of partitions and intersection points exist on the subject structure outside the house, so as to illustrate whether there are risk factors when the house changes, whether the house is a dangerous building, and to display the display forms of house types in multiple scenarios.

[0035] When a target house presents risks, the internal and external representations differ, leading to misidentification when generating its image structure. For example, when a wall crack is caused by a foundation problem, the primary focus is on the interior of the wall base. In this case, only oblique photography of the exterior can lead to misidentification of risks, causing the existing issues to be overlooked when displaying the house layout, leading to related issues. Feature points are used to assist in identifying defects during house identification. These points facilitate subsequent identification of differences between the current house and other houses within the building. These differences can quickly identify differences between different house layouts, reducing the waste of resources caused by multiple 3D model generation when the differences between house layouts are too small. House attributes represent various characteristic information about a house, such as its specific coordinates, geometric shape, and outline area.

[0036] Preferably, feature points are set based on the spatial location of the target house. Feature points represent the unique features of the house in terms of shape, size, height, structure, etc. For example, some houses may have unique roof shapes, such as pointed roofs, flat roofs, arched roofs, etc., or special window layouts, balcony styles, etc., which can be used as feature points to distinguish different houses.

[0037] like Figure 3As shown, the implementation of step S2 includes: S21, creating an equal number of split layers based on the set number of feature points, each of which contains at least one feature point and a building outline feature. The split layers are then compared to see if there are any differences between the multiple layers of the target house after the feature points have been set. These differences may manifest in aspects such as the house's shape, size, or structure. For example, the house outline in one layer may differ from that in another layer, or the position and form of a feature point within a layer may differ. After restoring the split layers, a determination is made as to whether the information displayed by each layer is consistent with that before the split, to determine if there are any structural differences between the target house within each split layer. If there are any differences, multiple image partitions are extracted to indicate significantly different locations within the target house. If the restored information is inconsistent with the original information, or if significant differences are found between the split layers, it is determined that there are structural differences between the target house within each split layer. Differences indicate that the structure of the house varies between the layers corresponding to different regions or feature points. These changes may be due to damage to the house, renovations, measurement errors, and other factors. By extracting this information, the specific areas in the house where differences exist can be located more accurately; ultimately, the process of repeated modeling of house types can be reduced. By simply inputting the differences between each house into a standard model, the differences between each house type can be displayed.

[0038] S22, based on the feature points and building outline features of the split layer, obtain the target difference points under the corresponding split layer in reverse order, search the split layer with the target difference points, and determine the house attributes corresponding to the target difference points. The house attributes represent the position, shape and area of the target difference points.

[0039] S23, performing split layer restoration based on the house attributes of the target difference point to obtain adjacent positions of the target difference point, and recording the adjacent positions of the target difference point in the split layer to obtain image partition information.

[0040] Assume that the split layers are stored as a hierarchical data structure, such as a tree structure or a list, and each layer corresponds to a layer. The feature points and building outline features of each layer are compared with the feature points and building outline features of other time points. When the difference exceeds 10%, it is recorded as a target difference point. These are compared in sequence by comparing each image in the aerial image data after sorting according to time; then the position and identification of the target difference point of each split layer are annotated.

[0041] Starting from the topmost split layer (the last split layer), the process traverses downward layer by layer. Within each layer, information about target difference points is extracted, represented by coordinates, shape features, and a unique identifier. During the traversal process, target difference points are filtered based on their characteristics, resulting in multiple target difference points of different types. When filtering target difference points, they must be displayed one by one in the order in which they were filtered.

[0042] Next, using a spatial index or mapping table, target difference points are associated with building attributes to describe their location distribution and corresponding coordinates on the building. Finally, based on the location and shape of the target difference points, their positions in the original complete layer are determined. Image processing algorithms, such as image fusion and boundary smoothing, are used to merge the split layers to restore the area surrounding the target difference points. During the merging process, topological relationships between buildings, such as adjacency and inclusion, are maintained and maintained. For each target difference point, its neighboring areas are analyzed, such as in the form of 8- or 4-neighborhoods, to determine adjacent locations. Adjacent locations are marked in the split layer, and their attributes, such as whether they are other difference points and their distance from the target difference point, are recorded. Based on the marked information in the difference layer, the image is divided into different regions. Each region corresponds to one or more target difference points and their adjacent locations. The attributes of each region, such as region ID, number of difference points included, and total area, form image partition information, completing the acquisition of image partition information. This image partition information includes the boundaries of each split layer and building attributes, such as building ID, area, and adjacency.

[0043] After the image partition information is acquired, the structure of the image partition information within the target house is used for division to determine the significant differences of each image partition within the target house under the iterative region growing method based on the content covered by the image partition information.

[0044] That is, the implementation method of the iterative segmentation result includes: dividing the house structure within each image partition information, taking the components of the target house and the center point within the image partition information as the starting point, and identifying the components of the target house according to the similarity criterion; at this time, the recognition is based on similarity criteria such as color, texture, gradient, etc., which can represent the characteristics of a single house structure, and identifying the house structure identified in the aerial image data. At this time, the house structure includes doors and windows, balconies, pipes, house boundary lines, adjacent relationships between structures, layout and other information. At this time, the identified house structure will be iterated to determine the house structure under the image partition information with differences, and the corresponding image data with the minimum boundary width of adjacent positions after iteration will be used to obtain relatively accurate difference point positions.

[0045] The components of the target house are retrieved using the image partition information to obtain the target house structure. The target house structure is used to characterize the content of the house structure with differences in the image partitions. Scene integration is performed based on the target house structure to obtain an iterative segmentation result.

[0046] After identifying the target house structure, the corresponding location of the house structure is described using multiple aerial image data sets. The system then obtains differences in adjacent locations when the target house structure is different. These differences generally indicate offsets in the house coordinates, such as cracks. Differences between certain house features and the previous image capture are considered discrepancies. These differences are combined to illustrate the form of problems that may occur in adjacent locations when the house structure is faulty. Examples include wall cracks, cracked layers, and the direction of cracks. These defects, which are indicative of house defects when using drone oblique photography, will be represented in the output iterative segmentation results.

[0047] In one embodiment of the present invention, after obtaining multiple problematic areas through iterative segmentation results, the images included in each image unit are used to identify images of houses where risks, hidden dangers, and progressive associations of various structures occur, and the visible areas and shadow parts in these images are recorded to further identify the differences between the houses, so as to obtain the difference distribution and expanded images under progressive identification of different houses.

[0048] like Figure 4 As shown, step S3 also includes the following when implemented: S31, counting the data of each image unit according to a preset time period, and querying the priority of each image unit according to the position of the image unit; the priority of the image unit will be set according to the position of the image unit corresponding to the image, and selected according to the number of occurrences of the position and the area corresponding to the image unit. When there is a house structure that appears in each difference statistics, and the identified area gradually increases, it means that there is a corresponding risk for the house type, or the same house may be partially blocked when shooting at different times, resulting in inaccurate identified house structure and other problems.

[0049] S32, select the image units according to their priority, number of occurrences and area, and set at least one connection relationship between the selected image units, which represents the spatial relationship, semantic relationship, etc. between the selected image units. The spatial relationship will represent the adjacent, inclusive and other relationships between the selected image units to illustrate whether there is spatial correlation between the image units; the semantic relationship will illustrate whether the house structure positions corresponding to the image units belong to the same type, to illustrate what category the different parts of the target house belong to.

[0050] S33: Construct a connection channel based on the connection relationship of the selected graph units, and count the graph units into an extended progressive graph according to the connection sequence of the connection channel.

[0051] The extended progressive diagram shows the connection relationship between the house structures that have differences or risks within a preset time period under the connection relationship, and explains the problematic structures. It can assist in the subsequent display of the differences in the houses and carry out comprehensive management of the houses.

[0052] Preferably, when selecting a graph unit based on its priority, number of occurrences and area, the number of occurrences of the graph unit within a preset time period is queried, and then the areas identified each time for the graph unit are combined according to the number of occurrences, and then the connection relationship between the graph units is utilized to select each graph unit according to time and priority to complete the connection of the graph units.

[0053] Preferably, the implementation method of selecting the graphic units includes: querying the priorities of all graphic units, and determining the selection range according to the priorities and areas of the graphic units; the selection range describes the range of values of different priorities and areas where the graphic units represent different structures, and retrieving the number of times the same or similar house structures appear to identify the occurrence of corresponding problems.

[0054] According to the number of times the graph unit appears in the selection range, the candidate graph unit is set, and the selection probability of the candidate graph unit is determined; at this time, the selection probability represents the probability value corresponding to the number of times the graph unit appears in the selection range, and the probability value can be calculated by Gaussian distribution.

[0055] According to the number of candidate image units, the area allocation ratio of each candidate unit is calculated. The area allocation ratio represents the ratio of the area of the corresponding image of the corresponding image unit to the total area at the time of statistics. The candidate image unit is selected in real time according to the area allocation ratio and the selection probability.

[0056] Preferably, the implementation method of selecting alternative map units in real time based on the area allocation ratio and the selection probability includes: comparing the selection probability with the selection probability threshold; when the selection probability is less than the selection probability threshold, classifying the area allocation ratio according to the selection probability difference; and selecting the alternative map unit corresponding to the area allocation ratio after classification as the selected alternative map unit.

[0057] When the selection probability is greater than the selection probability threshold, candidate graph units with connection relationships are selected in order from large to small according to the selection probability values to complete the selection of graph units.

[0058] The above content shows that when a house's difference points appear less frequently, the area of the difference is output. If the number of occurrences is large, it is selected according to its connection relationship to illustrate the corresponding relationship of the target house when differences occur.

[0059] In one embodiment of the present invention, Figure 5 As shown, the implementation of step S4 includes: S41, counting the overlapping frequency of each graph unit, and dividing the extended progressive graph into multiple levels based on the overlapping frequency.

[0060] S42, determining the amount of apartment type differences based on the amount of overlapping data in each layer.

[0061] At this time, after obtaining an extended progressive diagram of the changes in the locations of different houses over time and priority, the overlapping parts of these continuously updated parts are calculated, and the differences between different house types are explained according to the overlapping parts. The values corresponding to the pixel points when overlapping can also be used to count the number of pixels with different values to explain the amount of difference in house types.

[0062] In the extended progressive diagram, each unit represents a specific housing area or apartment type. These housing areas will be displayed using multiple images connected progressively according to the pre-selected time period, number of occurrences of the unit, area, priority, etc.

[0063] The overlapping areas are then counted, and the amount of overlapping data is used as the main output for subsequent analysis. For example, metrics such as the overlapping area and overlap ratio between units are calculated to quantify the degree of overlap between units. The differences in unit layouts are calculated by comparing changes in overlapping area and unit structure between house layers at different time points or from different sources. The final unit layout differences can be expressed as the change in overlapping area or the change ratio to describe the overlap between multiple units.

[0064] By counting overlapping chart units, it's possible to identify changes in house layouts at different points in time or across different sources; this helps understand information such as a house's usage and renovation history. By also counting overlapping data, it's also possible to verify data consistency between different data sources. If discrepancies are found, corrections or adjustments can be made promptly to improve data accuracy and reliability. This discrepancy can be used to assist with subsequent house repairs and, by incorporating these discrepancies, quickly generate a 3D model of the current house, improving the accuracy of subsequent model generation.

[0065] In one embodiment of the present invention, when performing three-dimensional stretching, as shown in FIG. Figure 6As shown, step S5 includes the following implementation: S51, extracting the spatial semantic information of each unit difference, obtaining at least one external boundary line corresponding to the target house, combining the external boundary line with the spatial semantic information to perform multiple position comparisons with the house interior, and determining the position comparison direction of each external boundary line. The external boundary line represents the lines of the overall house outline and defines the overall outline of the unit difference within the house area. The spatial semantic information of the unit difference described herein represents the connection relationship between the corresponding graph units when extracting the unit difference. The spatial and semantic relationships within the graph unit connection relationship are used as spatial semantic information to obtain the outline of the house area corresponding to the unit difference. The position comparison direction of each external boundary line indicates the position of the boundary corresponding to the house interior. For the data of the house interior, a pre-set three-dimensional standard model can be used, or after scanning the house interior data, the position of the external boundary line relative to the interior is identified to indicate its position comparison direction. The position comparison direction converts its coordinates into vector form to indicate the direction.

[0066] S52 compares the positions of each external boundary line to the internal boundary line of the target house, identifies the relative positional relationship between each external boundary line and the internal boundary line of the target house, determines a distance matrix based on the relative positional relationship, and verifies the relative distances and distribution of each boundary line using the distance matrix. The verified relative distances and distribution of each boundary line are output as the house type difference distribution. The internal boundary lines represent the dividing lines between the various structures in the house and are used to describe the line segments that separate the structures in the corresponding area of the house type difference. This indicates whether the spatial positions of the identified house type differences are accurate and consistent.

[0067] When correcting the three-dimensional point cloud data of the target house, S53, the internal boundary lines and the external boundary lines are spliced to generate a three-dimensional model corresponding to the target house, and the three-dimensional point cloud data is corrected according to the difference distribution of each house type under the three-dimensional model, so as to recursively update the display category of the target house based on the plane where the target house is located during the correction.

[0068] After the above content is stitched together using the internal and external boundary lines to form a 3D model, the data corresponding to the distribution of differences between the apartment types is mapped onto the 3D model to illustrate the corresponding differences in the 3D model. The current house type displayed is then continuously updated according to the revised content.

[0069] Preferably, the relative distance and distribution of each boundary line is verified by using a distance matrix, which is expressed as whether the distance between each internal boundary line and each external boundary line is consistent when using data from multiple images for identification, to determine whether the currently acquired data is accurate. When the data is verified to be accurate, the distance matrix and the corresponding external boundary lines and internal boundary lines are used as the output of the household type difference distribution.

[0070] Preferably, the implementation method of recursively updating the display category of the target house based on the plane where the target house is located during correction includes: extracting the apartment boundary of the plane where the target house is located, and verifying whether the connection position is consistent with the orientation of the target house based on the connection position of all walls on the apartment boundary; if consistent, determining the overlapping line segments of the internal boundary line and the external boundary line of the target house, and recursively updating the display category of the target house according to the overlapping line segments.

[0071] The above content explains that when recursively updating the current target house, the target house's plane is used to identify the wall information that needs to be updated or queried. The position of the line segments on these walls is compared with the current house to identify the data that needs to be updated. At the same time, it is verified whether these areas that need to be updated belong to the internal or external boundaries. Finally, the parts of the house that differ are updated to show the differences in the different planes when displaying the apartment layout. This eliminates the impact of the environment on the multi-scale mapping of aerial images, realizes the spatial semantic association and matching of each apartment layout, and ultimately improves the speed of 3D model generation and the accuracy of house layout display.

[0072] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention, which are still covered by the scope of protection of the present invention.

Claims

1. A method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography, characterized in that: include: S1, obtaining aerial image data of a target house, extracting the building outline features of the target house using the aerial image data, and performing preliminary spatial registration on the building outline features of the target house; S2, setting feature points based on the spatial location of the target house, using each feature point to check the house attributes with the building outline features, extracting multiple image partition information under oblique photography, dividing the structure of the target house based on the image partition information, and determining the iterative segmentation results of the target house structure at adjacent locations; S3, setting the iterative segmentation result to multiple graph units, querying the priority of the graph units, selecting graph units based on their priority, number of occurrences, and area, building a connection channel between the selected graph units, and setting an extended progressive graph corresponding to the graph units; the priority of the graph units is set according to the position of the graph units corresponding to the image, and the graph units are selected based on the number of occurrences of the position and the area corresponding to the graph units; Step S3, when implemented, further includes: S31, collecting statistics of the data of each graph unit according to a preset time period, and querying the priority of each graph unit according to the position of the graph unit; S32, selecting graph units according to their priority, number of occurrences, and area, and setting at least one connection relationship between the selected graph units; The implementation of the selected graph unit includes: Query the priorities of all graph units and determine the selection range based on the priority and area of the graph units; According to the number of occurrences of the graph unit in the selection range, the candidate graph unit is set, and the selection probability of the candidate graph unit is determined; According to the number of candidate map units, the area allocation ratio of each candidate unit is calculated, and the candidate map unit is selected in real time according to the area allocation ratio and selection probability; S33, constructing a connection channel based on the connection relationship of the selected graph units, and counting the graph units into an extended progressive graph according to the connection order of the connection channel; S4, counting the overlapping parts of each unit according to the extended progressive graph to obtain the difference in apartment types; S5, three-dimensionally stretching the apartment type difference to obtain the apartment type difference distribution, using the apartment type difference distribution to correct the three-dimensional point cloud data of the target house, and recursively updating the display category of the target house.

2. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: The implementation of step S1 further includes: S11, determining the viewing angle of the aerial image data of the target house, dividing the aerial image data of the target house into a plurality of viewing angle partition information based on the viewing angles during shooting, and arranging a plurality of image control points and check points in each viewing angle partition information; S12, obtaining the layout positions of the image control points and the check points under each viewing angle partition information, and aligning the position of each viewing angle partition information according to the layout positions of the image control points; S13, using the aligned image control points and check points as the building outline features of the target house, completing the preliminary spatial alignment of the target house.

3. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: The implementation of step S2 includes: S21, creating an equal number of split layers based on the set number of feature points, wherein each split layer contains at least one feature point and a building outline feature; S22, based on the feature points of the split layer and the building outline features, obtaining the target difference points under the corresponding split layer in reverse order, searching the split layer with the target difference points, and determining the house attributes corresponding to the target difference points; S23, performing split layer restoration based on the house attributes of the target difference point to obtain adjacent positions of the target difference point, and recording the adjacent positions of the target difference point in the split layer to obtain image partition information.

4. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: The implementation methods of iterative segmentation results include: The house structure in each image partition information is divided, and the components of the target house and the center point in the image partition information are used as starting points to identify the components of the target house according to the similarity criterion; The components of the target house are retrieved using the image partition information to obtain the target house structure. The target house structure is used to characterize the content of the house structure with differences in the image partitions. Scene integration is performed based on the target house structure to obtain an iterative segmentation result.

5. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: The implementation methods for selecting candidate map units in real time based on area allocation ratio and selection probability include: Compare the selection probability with the selection probability threshold. When the selection probability is less than the selection probability threshold, classify the area allocation ratio according to the selection probability difference, and select the candidate map unit corresponding to the area allocation ratio after classification as the selected candidate map unit. When the selection probability is greater than the selection probability threshold, candidate graph units with connection relationships are selected in descending order according to the selection probability values.

6. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: The implementation of step S4 includes: S41, counting the overlapping frequency of each graph unit, and dividing the extended progressive graph into multiple levels based on the overlapping frequency; S42, determining the amount of apartment type differences based on the amount of overlapping data in each layer.

7. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 1, characterized in that: Step S5 includes the following implementations: S51, extracting spatial semantic information of the differences between each apartment type, obtaining at least one external boundary line corresponding to the target house, and performing multiple position comparisons of the external boundary line with the interior of the house in combination with the spatial semantic information to determine the position and direction of each external boundary line; S52, comparing the positions and directions of the external boundary lines, identifying the relative positional relationship between each external boundary line and the internal boundary line of the target house, determining a distance matrix based on the relative positional relationship, and verifying the relative distances and distribution of the boundary lines using the distance matrix. The verified relative distances and distribution of the boundary lines are output as the house type difference distribution; S53, splicing the internal boundary lines and the external boundary lines to generate a three-dimensional model corresponding to the target house, and correcting the three-dimensional point cloud data based on the difference distribution of each house type under the three-dimensional model, so as to recursively update the display category of the target house based on the plane where the target house is located during the correction.

8. The method for displaying house layouts by integrating a three-dimensional model with aerial oblique photography according to claim 7, characterized in that: Methods for recursively updating the display category of the target house based on the plane where the target house is located during correction include: Extract the target house's floor plan boundary and use the connection locations of all walls on the floor plan boundary to verify whether the connection locations are consistent with the target house's orientation. If they are consistent, determine the overlapping line segments of the target house's internal and external boundary lines and recursively update the target house's display category based on the overlapping line segments.

Images