Urban three-dimensional green space extraction method and system
Through tilted photogrammetry, the separation of point clouds and the monolithization of buildings are measured, and the urban three-dimensional green space is extracted, which solves the problem of insufficient extraction of three-dimensional green spaces in the existing technology, and accurately identify and distinguish urban three-dimensional green spaces are achieved.
Patent Information
- Application Number
- CN202411922275.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-23
- Publication Date
- 2025-04-25
AI Technical Summary
The existing technology lacks effective three-dimensional interpretation methods in urban three-dimensional green space extraction, mainly focusing on the extraction of ground green spaces of two-dimensional image data, and failing to fully solve the problem of extraction of three-dimensional green spaces.
By obtaining tilt photogrammetry point clouds, it is divided into ground point clouds and three-dimensional geological point clouds, and the three-dimensional geological point clouds are further extracted from building monolithic point clouds, the roof point clouds and facade point clouds are separated, and the urban three-dimensional green space is finally extracted based on these point clouds.
It realizes the accurate extraction of urban three-dimensional green spaces, effectively distinguishes ground green spaces from three-dimensional green spaces on buildings, and provides technical means for the refined interpretation and analysis of subsequent urban green spaces.
Smart Images
Figure CN119888485A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to remote sensing data detection technology, and in particular to an urban three-dimensional green space extraction method, extraction system, electronic equipment and computer-readable storage medium. Background Art
[0002] At present, most of the relevant research on urban 3D green space extraction is focused on ground green space extraction based on 2D image data, and there is little research on 3D green space. In terms of technology, 3D interpretation methods based on oblique photogrammetry data are mostly focused on the classification of ground object types, and there is no 3D interpretation method for 3D green space extraction. Summary of the invention
[0003] In view of this, the present application provides a method, an extraction system, an electronic device and a computer-readable storage medium for extracting urban three-dimensional green spaces, so as to achieve accurate extraction of urban three-dimensional green spaces.
[0004] To solve the above problems, this application adopts the following technical solutions:
[0005] One of the purposes of this application is to provide a method for extracting urban three-dimensional green space, comprising the following steps:
[0006] Obtain oblique photogrammetry point cloud;
[0007] Dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud;
[0008] Extracting building individual point clouds from the three-dimensional object point cloud;
[0009] Separating the building monomer point cloud into a roof point cloud and a facade point cloud;
[0010] The urban three-dimensional green space is extracted according to the roof point cloud and the facade point cloud.
[0011] In some embodiments, the step of obtaining the oblique photogrammetry point cloud specifically includes the following steps:
[0012] Read the oblique photogrammetry 3D model to obtain the oblique photogrammetry point cloud data, denoted as P = {P i :i=1,...,n}, where: i is the point cloud index, n is the number of points in the point cloud, P i =(P ix ,P iy ,P iz ,P iR ,P iG ,P iB ) is the point cloud feature vector, xyz represents the longitude, latitude and altitude of the point cloud, and RGB represents the red, green and blue band spectral characteristics of the point cloud.
[0013] In some embodiments, the step of dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud specifically includes the following steps:
[0014] According to the multi-level cloth simulation filtering algorithm, the oblique photogrammetry point cloud is divided into a ground point cloud and a three-dimensional object point cloud, wherein: the three-dimensional object point cloud is recorded as P'={P i ’ :i=1,...,n'}, wherein: the multi-level cloth simulation filtering method is specifically: by adjusting the cloth thickness, the ground point cloud and the three-dimensional object point cloud at different thickness levels are obtained, and the result obtained by taking the minimum thickness corresponding to the stable result is taken as the final result.
[0015] In some embodiments, the step of extracting building individual point clouds from the three-dimensional object point cloud specifically includes the following steps:
[0016] The three-dimensional object point cloud is subjected to data rasterization processing: the ground area covered by the three-dimensional object point cloud is assumed to be D, and D is divided into grids with a grid scale of d×d. If the grid contains multiple data points, the elevation mean of these data points is used as the elevation value of the grid; if the grid does not contain data points, the elevation mean of the neighborhood grids of the grid is used as the elevation value of the grid; according to the maximum and minimum values of the grid elevation after the above rasterization, the elevation value is quantized to the range of 0-255, and it is ensured that each grid contains at least one data point, then the data after rasterization is {(x s ,y h ,Z sh ),s=1,...,p,h=1,...,q}, where x s ,y h are the integer coordinates of the grid points, Z i,j is the grid point (x s ,y h ) is the gray value at the point (representing the elevation value of the point), p and q are the number of rows and columns of the rasterized image respectively;
[0017] The three-dimensional object point cloud is subjected to individualization processing: the image is divided into a plurality of independent and connected regions according to the spatial connectivity of the grayscale value pixels in the image, and all the point clouds in the region are regarded as a whole, thereby completing the individualization of the three-dimensional object point cloud;
[0018] The three-dimensional object point cloud is processed into patches: based on the three-dimensional model of oblique photogrammetry, the patch vertex data is obtained, which is recorded as V = {V j :j=1,...,m}, where j is the triangle index, m is the number of triangles, V j = {P i1 ,Pi2 ,P i3} is the face vertex set, P i1 ,P i2 ,P i3 are the three vertices that make up the patch, and index the vertex set in the three-dimensional object point cloud. If If the triangle is not in the correct position, it will be retained; otherwise, it will be deleted.
[0019] Extract multiple features of the patch-processed three-dimensional object point cloud, the multiple features include spectral features, texture features, normal features and geometric features; wherein the spectral feature is the average of the spectral features of the point cloud in the patch; the texture feature and the normal feature are obtained based on the original model file index; the geometric feature is described as the ratio of the sum of the vertical patch and the horizontal patch to the total number of patches of the monomerized object;
[0020] Based on the multiple features, a supervised machine learning method is used to realize individual identification of buildings to obtain individual point clouds of buildings. The supervised machine learning method includes SVM or random forest or deep learning.
[0021] In some of the embodiments, the step of separating the building monomer point cloud into roof point cloud and facade point cloud specifically includes the following steps: dividing the building point cloud into roof point cloud and facade point cloud according to the normal direction of the facets corresponding to the building monomer point cloud.
[0022] In some of the embodiments, the step of extracting urban three-dimensional green space based on the roof point cloud and the facade point cloud specifically includes the following steps: extracting the green point cloud from the roof point cloud and the facade point cloud, and then realizing the extraction of roof green space and facade green space.
[0023] In some embodiments, when the near infrared band is lacking, the vegetation index feature can be calculated based on the RGB band, as follows:
[0024] VI= 3G-2.4RB
[0025] R+G+B
[0026] Among them, RGB represents the red, green and blue spectral characteristics respectively, and the VI value range is (-1,1). The closer the value is to 1, the greater the probability of vegetation.
[0027] The second purpose of this application is to provide a system for extracting urban three-dimensional green space, including:
[0028] Data acquisition module: used to obtain oblique photogrammetry point cloud;
[0029] The first separation module is used to separate the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud;
[0030] Feature extraction module: used for extracting building individual point cloud from the three-dimensional object point cloud;
[0031] A second separation module: used for separating the building monomer point cloud into a roof point cloud and a facade point cloud;
[0032] Output module: used for extracting urban three-dimensional green space according to the roof point cloud and facade point cloud.
[0033] The second purpose of the present application is to provide an electronic device, including a processor, a memory and a communication interface, wherein the memory stores one or more programs, and the one or more programs are executed by the processor, and the one or more programs include instructions for executing the steps in any one of the methods described.
[0034] The second object of the present application is to provide a computer-readable storage medium, which stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of any one of the methods described.
[0035] This application adopts the above technical solution, and its beneficial effects are as follows:
[0036] The urban three-dimensional green space extraction method, extraction system, electronic device and computer-readable storage medium provided in the present application obtain oblique photogrammetry point cloud; divide the oblique photogrammetry point cloud into ground point cloud and three-dimensional object point cloud; extract building individual point cloud from the three-dimensional object point cloud; separate the building individual point cloud into roof point cloud and facade point cloud, and extract urban three-dimensional green space based on the roof point cloud and facade point cloud. The urban three-dimensional green space extraction method provided in the present application can effectively distinguish between ground green space and three-dimensional green space on buildings, and provide effective technical means for the subsequent fine interpretation and analysis of urban green space. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0038] Figure 1 This is a flow chart of the steps of the urban three-dimensional green space extraction method provided in Example 1 of the present application.
[0039] Figure 2 A schematic diagram of the structure of the original point cloud of oblique photogrammetry provided in Example 1 of the present application.
[0040] Figure 3 Schematic diagram of the ground point cloud filtering results provided in Example 1 of the present application.
[0041] Figure 4 This is the three-dimensional extraction result of the stereoscopic green space provided in Example 1 of the present application.
[0042] Figure 5 This is a schematic diagram of the structure of the urban three-dimensional green space extraction system provided in Example 2 of the present application.
[0043] Figure 6 This is a schematic diagram of the structure of the electronic device provided in Example 3 of the present application. DETAILED DESCRIPTION
[0044] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0045] In the description of the present application, it should be understood that the terms "upper", "lower", "horizontal", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.
[0046] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0047] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments.
[0048] Example 1
[0049] See also Figure 1 , which is a step flow chart of the urban three-dimensional green space extraction method provided in Example 1 of the present application, specifically includes the following steps S110 to S150, and the implementation method of each step is described in detail below.
[0050] Step S110: Obtaining oblique photogrammetry point cloud.
[0051] Specifically, the step of obtaining the oblique photogrammetry point cloud includes the following steps: reading the oblique photogrammetry three-dimensional model to obtain the oblique photogrammetry point cloud data, denoted as P = {P i :i=1,...,n}, where i is the point cloud index, n is the number of points in the point cloud, P i =(P ix ,P iy ,P iz ,P iR ,P iG ,P iB ) is the point cloud feature vector, where xyz represents the longitude, latitude and altitude of the point cloud, and RGB represents the red, green and blue band spectral characteristics of the point cloud. Figure 2 , which is a schematic diagram of the structure of the original point cloud of oblique photogrammetry provided in this embodiment.
[0052] Step S120: dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud.
[0053] Specifically, the tilted photogrammetry point cloud is divided into a ground point cloud and a three-dimensional object point cloud, which specifically includes the following steps: based on a multi-level cloth simulation filtering algorithm, the tilted photogrammetry point cloud is divided into a ground point cloud and a three-dimensional object point cloud, wherein the three-dimensional object point cloud is recorded as P'={P i ’ :i=1,...,n'}. The multi-level cloth simulation filtering method is described as follows: by adjusting the cloth thickness, the ground point cloud and three-dimensional object point cloud at different thickness levels are obtained, and the result obtained by taking the minimum thickness corresponding to the stable result is the final result. Please refer to Figure 3 , which is a schematic diagram of the ground point cloud filtering results provided in this embodiment.
[0054] Step S130: extracting building individual point clouds from the three-dimensional object point cloud.
[0055] Specifically, in the step of extracting building individual point clouds from the three-dimensional object point cloud, the following steps are specifically included:
[0056] Rasterize the three-dimensional point cloud data: Assume that the ground area covered by the three-dimensional point cloud data is D. In order to rasterize the point cloud data set, first divide D into grids with a grid scale of d×d. If the grid contains multiple data points, the average elevation of these data points is used as the elevation value of the grid; if the grid does not contain data points, the average elevation of the neighborhood grids of the grid is used as the elevation value of the grid. Finally, according to the maximum and minimum values of the grid elevation after the above rasterization, quantize its elevation value to the range of 0-255, and ensure that each grid contains at least one data point. Suppose the rasterized data is {(x s ,y h ,Zsh ),s=1,...,p,h=1,...,q}, where x s ,y h are the integer coordinates of the grid points, Z i,j is the grid point (x s ,y h ) (representing the elevation value of the point), p and q are the number of rows and columns of the rasterized image respectively.
[0057] Individualization of the three-dimensional object point cloud: According to the spatial connectivity of the grayscale pixels in the image, the image is divided into several independent and connected areas, and all the point clouds in the area are regarded as a whole, thereby completing the individualization of the three-dimensional object point cloud.
[0058] The three-dimensional object point cloud is processed into patches: the patch vertex data is obtained based on the three-dimensional model of oblique photogrammetry, which is denoted as V = {V j :j=1,...,m}, where j is the triangle index, m is the number of triangles, V j = {P i1 ,P i2 ,P i3} is the face vertex set, P i1 ,P i2 ,P i3 are the three vertices that make up the patch. In the three-dimensional object point cloud, if If yes, the triangle will be retained, otherwise it will be deleted.
[0059] Multi-feature extraction is performed on the patched 3D object point cloud, where the multi-features include spectral features, texture features, normal features, geometric features, etc.; the spectral features of the patch are the average of the spectral features of the point cloud within the patch; the texture features and normal features are obtained based on the original model file index; the geometric features are described as the ratio of the sum of the vertical patches and the horizontal patches to the total number of patches of the monomerized object.
[0060] Based on the multiple features, supervised machine learning methods are used to realize the individual identification of buildings to obtain individual point clouds of buildings. The supervised machine learning methods include SVM or random forest or deep learning
[0061] Step S140: Separate the building monomer point cloud into a roof point cloud and a facade point cloud.
[0062] Specifically, the separation of the building monomer point cloud into roof point cloud and facade point cloud includes the following steps: according to the normal direction of the surface corresponding to the building monomer point cloud, the building point cloud is divided into the roof point cloud and the facade point cloud with a certain robustness.
[0063] Step S150: extracting urban three-dimensional green space according to the roof point cloud and the facade point cloud.
[0064] It can be understood that this embodiment extracts green space points in roof point clouds and facade point clouds respectively by selecting patched spectral features, texture features and vegetation index features based on supervised machine learning methods (such as SVM, random forest, deep learning, etc.), thereby realizing the extraction of roof green space and facade green space. Among them, when the near-infrared band is lacking, the vegetation index feature can be calculated based on the RGB band, as follows:
[0065]
[0066] Among them, RGB represents the red, green and blue spectral characteristics respectively. The VI value range is (-1,1). The closer the value is to 1, the greater the probability of vegetation. Figure 4 , which is the three-dimensional extraction result of the stereoscopic green space provided in this embodiment.
[0067] The urban three-dimensional green space extraction method provided in this application can effectively distinguish between ground green spaces and three-dimensional green spaces on buildings, and provide an effective technical means for the subsequent detailed interpretation and analysis of urban green spaces.
[0068] Example 2
[0069] See also Figure 5 , which is a structural diagram of the urban three-dimensional green space extraction system provided in Example 2 of the present application, specifically comprising:
[0070] Data acquisition module 510: used to obtain oblique photogrammetry point cloud.
[0071] Specifically, the step of obtaining the oblique photogrammetry point cloud includes: reading the oblique photogrammetry three-dimensional model to obtain the oblique photogrammetry point cloud data, denoted as P = {P i :i=1,...,n}, where i is the point cloud index, n is the number of points in the point cloud, P i =(P ix ,P iy ,P iz ,P iR ,P iG ,P iB ) is the point cloud feature vector, where xyz represents the longitude, latitude and altitude of the point cloud, and RGB represents the red, green and blue band spectral characteristics of the point cloud, respectively.
[0072] The first separation module 520 is used to separate the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud.
[0073] Specifically, the tilted photogrammetry point cloud is divided into a ground point cloud and a three-dimensional object point cloud, which specifically includes: dividing the tilted photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud based on a multi-level cloth simulation filtering algorithm, wherein the three-dimensional object point cloud is recorded as P'={P i ’ :i=1,...,n'}. The multi-level cloth simulation filtering method is described as follows: by adjusting the cloth thickness, the ground point cloud and the three-dimensional object point cloud at different thickness levels are obtained, and the result obtained by taking the minimum thickness corresponding to the stable result is the final result.
[0074] Feature extraction module 530: used to extract building individual point cloud from the three-dimensional object point cloud.
[0075] Specifically, the step of extracting building individual point clouds from the three-dimensional object point cloud specifically includes:
[0076] Rasterize the three-dimensional point cloud data: Assume that the ground area covered by the three-dimensional point cloud data is D. In order to rasterize the point cloud data set, first divide D into grids with a grid scale of d×d. If the grid contains multiple data points, the average elevation of these data points is used as the elevation value of the grid; if the grid does not contain data points, the average elevation of the neighborhood grids of the grid is used as the elevation value of the grid. Finally, according to the maximum and minimum values of the grid elevation after the above rasterization, quantize its elevation value to the range of 0-255, and ensure that each grid contains at least one data point. Suppose the rasterized data is {(x s ,y h ,Z sh ),s=1,...,p,h=1,...,q}, where x s ,y h are the integer coordinates of the grid points, Z i,j is the grid point (x s ,y h ) (representing the elevation value of the point), p and q are the number of rows and columns of the rasterized image respectively.
[0077] Individualization of the three-dimensional object point cloud: According to the spatial connectivity of the grayscale pixels in the image, the image is divided into several independent and connected areas, and all the point clouds in the area are regarded as a whole, thereby completing the individualization of the three-dimensional object point cloud.
[0078] The three-dimensional object point cloud is processed into patches: the patch vertex data is obtained based on the three-dimensional model of oblique photogrammetry, which is denoted as V = {V j :j=1,...,m}, where j is the triangle index, m is the number of triangles, V j = {P i1 ,P i2 ,Pi3} is the face vertex set, P i1 ,P i2 ,P i3 are the three vertices that make up the patch. In the three-dimensional object point cloud, if If yes, the triangle will be retained, otherwise it will be deleted.
[0079] Multi-feature extraction is performed on the patched 3D object point cloud, where the multi-features include spectral features, texture features, normal features, geometric features, etc.; the spectral features of the patch are the average of the spectral features of the point cloud within the patch; the texture features and normal features are obtained based on the original model file index; the geometric features are described as the ratio of the sum of the vertical patches and the horizontal patches to the total number of patches of the monomerized object.
[0080] Based on the multiple features, a supervised machine learning method is used to realize individual identification of buildings to obtain individual point clouds of buildings. The supervised machine learning method includes SVM or random forest or deep learning.
[0081] The second separation module 540 is used to separate the building monomer point cloud into a roof point cloud and a facade point cloud.
[0082] It can be understood that this embodiment extracts green space points in roof point clouds and facade point clouds respectively by selecting patched spectral features, texture features and vegetation index features based on supervised machine learning methods (such as SVM, random forest, deep learning, etc.), thereby realizing the extraction of roof green space and facade green space. Among them, when the near-infrared band is lacking, the vegetation index feature can be calculated based on the RGB band, as follows:
[0083]
[0084] Among them, RGB represents the red, green and blue spectral characteristics respectively. The value range of VI is (-1,1). The closer the value is to 1, the greater the probability of vegetation.
[0085] The urban three-dimensional green space extraction system provided in this application can effectively distinguish between ground green spaces and three-dimensional green spaces on buildings, and provide effective technical means for the subsequent detailed interpretation and analysis of urban green spaces.
[0086] Example 3
[0087] See also Figure 6 , Figure 6 It is a structural diagram of an electronic device provided in an embodiment of the present application, and the electronic device includes: one or more processors, one or more memories, one or more communication interfaces, and one or more programs; the one or more programs are stored in the memories and are configured to be executed by the one or more processors.
[0088] The above program includes instructions for performing the following steps:
[0089] Obtain oblique photogrammetry point cloud;
[0090] Dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud;
[0091] Extracting building individual point clouds from the three-dimensional object point cloud;
[0092] Separating the building monomer point cloud into a roof point cloud and a facade point cloud;
[0093] The urban three-dimensional green space is extracted according to the roof point cloud and the facade point cloud.
[0094] Among them, all relevant contents of each scenario involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.
[0095] It should be understood that the above-mentioned memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.
[0096] In the embodiment of the present application, the processor of the above-mentioned device may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0097] It should be understood that the "at least one" involved in the embodiments of the present application refers to one or more, and "plurality" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: the existence of A alone, the existence of A and B at the same time, and the existence of B alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.
[0098] Furthermore, unless otherwise specified, ordinal numbers such as "first" and "second" mentioned in the embodiments of the present application are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects. For example, the first information and the second information are only used to distinguish different information, and do not indicate the difference in content, priority, sending order or importance of the two types of information.
[0099] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software units in a processor for execution. The software unit can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor executes the instructions in the memory, and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.
[0100] Example 4
[0101] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiments.
[0102] The present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any method described in the above method embodiment. The computer program product may be a software installation package.
[0103] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.
[0104] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0105] In the several embodiments provided in the present application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.
[0106] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.
[0107] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0108] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, server or TRP, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.
[0109] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing the relevant hardware through a program, and the program may be stored in a computer-readable memory, which may include a flash drive, ROM, RAM, a magnetic disk or an optical disk, etc.
[0110] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of the present application. At the same time, for general technical personnel in this field, according to the idea of the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.
[0111] The above are only preferred embodiments of the present application, and only specifically describe the technical principles of the present application. These descriptions are only for explaining the principles of the present application and cannot be interpreted as limiting the scope of protection of the present application in any way. Based on the explanation here, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application, and other specific implementation methods of the present application that can be associated with the technicians in this field without creative work, should be included in the scope of protection of the present application.
Claims
1. A method for extracting urban three-dimensional green space, characterized in that: The steps include: Obtain oblique photogrammetry point cloud; Dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud; Extracting building individual point clouds from the three-dimensional object point cloud; Separating the building monomer point cloud into a roof point cloud and a facade point cloud; The urban three-dimensional green space is extracted according to the roof point cloud and the facade point cloud.
2. The urban three-dimensional green space extraction method according to claim 1, characterized in that: The steps of obtaining the oblique photogrammetry point cloud specifically include the following steps: Read the oblique photogrammetry 3D model to obtain the oblique photogrammetry point cloud data, denoted as P = {P i :i=1,...,n}, where: i is the point cloud index, n is the number of points in the point cloud, P i =(P ix ,P iy ,P iz ,P iR ,P iG ,P iB ) is the point cloud feature vector, xyz represents the longitude, latitude and altitude of the point cloud, and RGB represents the red, green and blue band spectral characteristics of the point cloud.
3. The urban three-dimensional green space extraction method according to claim 2, characterized in that: The step of dividing the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud specifically includes the following steps: According to the multi-level cloth simulation filtering algorithm, the oblique photogrammetry point cloud is divided into a ground point cloud and a three-dimensional object point cloud, wherein: the three-dimensional object point cloud is recorded as P'={P i ’ :i=1,...,n'}, wherein: the multi-level cloth simulation filtering method is specifically: by adjusting the cloth thickness, the ground point cloud and the three-dimensional object point cloud at different thickness levels are obtained, and the result obtained by taking the minimum thickness corresponding to the stable result is taken as the final result.
4. The urban three-dimensional green space extraction method according to claim 3, characterized in that: The step of extracting building individual point clouds from the three-dimensional object point cloud specifically includes the following steps: The three-dimensional object point cloud is subjected to data rasterization processing: the ground area covered by the three-dimensional object point cloud is assumed to be D, and D is divided into grids with a grid scale of d×d. If the grid contains multiple data points, the elevation mean of these data points is used as the elevation value of the grid; if the grid does not contain data points, the elevation mean of the neighborhood grids of the grid is used as the elevation value of the grid; according to the maximum and minimum values of the grid elevation after the above rasterization, the elevation value is quantized to the range of 0-255, and it is ensured that each grid contains at least one data point, then the data after rasterization is {(x s ,y h ,Z sh ),s=1,...,p,h=1,...,q}, where x s ,y h are the integer coordinates of the grid points, Z i,j is the grid point (x s ,y h ) is the gray value at the point (representing the elevation value of the point), p and q are the number of rows and columns of the rasterized image respectively; The three-dimensional object point cloud is subjected to individualization processing: the image is divided into a plurality of independent and connected regions according to the spatial connectivity of the grayscale value pixels in the image, and all the point clouds in the region are regarded as a whole, thereby completing the individualization of the three-dimensional object point cloud; The three-dimensional object point cloud is processed into patches: based on the three-dimensional model of oblique photogrammetry, the patch vertex data is obtained, which is recorded as V = {V j :j=1,...,m}, where j is the triangle index, m is the number of triangles, V j = {P i1 ,P i2 ,P i3 } is the face vertex set, P i1 ,P i2 ,P i3 are the three vertices that make up the patch, and index the vertex set in the three-dimensional object point cloud. If If the triangle is not in the correct position, it will be retained; otherwise, it will be deleted. Extract multiple features of the patch-processed three-dimensional object point cloud, the multiple features include spectral features, texture features, normal features and geometric features; wherein the spectral feature is the average of the spectral features of the point cloud in the patch; the texture feature and the normal feature are obtained based on the original model file index; the geometric feature is described as the ratio of the sum of the vertical patch and the horizontal patch to the total number of patches of the monomerized object; Based on the multiple features, a supervised machine learning method is used to realize individual identification of buildings to obtain individual point clouds of buildings. The supervised machine learning method includes SVM or random forest or deep learning.
5. The urban three-dimensional green space extraction method according to claim 4, characterized in that: The step of separating the building monomer point cloud into a roof point cloud and a facade point cloud specifically includes the following steps: dividing the building point cloud into a roof point cloud and a facade point cloud according to the normal direction of the surface corresponding to the building monomer point cloud.
6. The urban three-dimensional green space extraction method according to claim 5, characterized in that: The step of extracting urban three-dimensional green space according to the roof point cloud and the facade point cloud specifically includes the following steps: extracting the green point cloud from the roof point cloud and the facade point cloud, thereby realizing the extraction of the roof green space and the facade green space.
7. The urban three-dimensional green space extraction method according to claim 6, characterized in that: When the near-infrared band is lacking, the vegetation index characteristics can be calculated based on the RGB band as follows: Among them, RGB represents the red, green and blue spectral characteristics respectively, and the VI value range is (-1,1). The closer the value is to 1, the greater the probability of vegetation.
8. An urban three-dimensional green space extraction system, characterized in that: include: Data acquisition module: used to obtain oblique photogrammetry point cloud; The first separation module is used to separate the oblique photogrammetry point cloud into a ground point cloud and a three-dimensional object point cloud; Feature extraction module: used for extracting building individual point cloud from the three-dimensional object point cloud; A second separation module: used for separating the building monomer point cloud into a roof point cloud and a facade point cloud; Output module: used for extracting urban three-dimensional green space according to the roof point cloud and facade point cloud.
9. An electronic device, characterized in that: The method comprises a processor, a memory and a communication interface, wherein the memory stores one or more programs, and the one or more programs are executed by the processor, and the one or more programs include instructions for executing the steps in the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute the steps of the method according to any one of claims 1 to 7.