Embedded urban design scene emulation method and system

Abstract

The present invention discloses an embedded urban design scene emulation method and system. The method includes the following steps: constructing a status quo urban three-dimensional model scene according to collected oblique photography data; loading a three-dimensional model of urban design to a scene, and extracting geometric attributes for generation of buildings; unifying a space coordinate system of models and scenes, and automatically determining a space matching degree by taking buildings as a basic unit, and marking matched buildings with Y and mismatched buildings with N for distinction; for a region with the buildings marked with N, performing a local flattening operation in a three-dimensional model scene of oblique photography to flatten stereo data; for a region with the buildings marked with Y, performing real-time space editing in the three-dimensional model of urban design to hide the marked buildings; and opening two sets of processed space data to implement mosaic display. The present invention can conveniently embed a three-dimensional model of urban design into a status quo three-dimensional model of oblique photography for scene emulation, and provides technical and method supports for digital presentation and management of urban design achievements.

Claims

1. An embedded urban design scene emulation method, wherein the method comprises the following steps: step 1: processing oblique photography data in an established range obtained by collection to construct a three-dimensional model of oblique photography, performing object management on buildings in the model, and extracting geometric attributes for generation of each building; step 2: loading a three-dimensional model of urban design in an established range into a three-dimensional model scene of oblique photography, and extracting geometric attributes for generation of each building; step 3: performing, for the three-dimensional model scene of oblique photography and three-dimensional model data of urban design in a unified space coordinate system, automatic determination of space matching by taking buildings as a basic unit, matched buildings being marked with Y, and mismatched buildings being marked with N; step 4: performing, for the buildings marked with N, a flattening operation in the three-dimensional model of oblique photography, so that stereo data of the region is leveled off; step 5: performing, for the buildings marked with Y, real-time space editing in the three-dimensional model of urban design to hide the marked buildings; and step 6: simultaneously opening two sets of data processed in step 4 and step 5 to implement mosaic display.

2. The embedded urban design scene emulation method according to claim 1, wherein a specific method for step (1) is as follows: (1.1) collecting and acquiring oblique photography data not less than an established range, that is, oblique data in an urban design range; (1.2) generating, for the oblique photography data, the three-dimensional model of oblique photography based on real image texture through automatic modeling software; (1.3) loading the three-dimensional model of oblique photography through a SuperMap platform; (1.4) constructing a triangulated irregular network (TIN), and mapping high-resolution images taken from different angles onto a TIN model; and (1.5) extracting a two-dimensional basal surface of a building, to implement object management on a building model.

3. The embedded urban design scene emulation method according to claim 1, wherein a specific method for step (2) is as follows: (2.1) editing the three-dimensional model of urban design, clearing the geographic position of the model, and setting latitude and longitude information to zero; (2.2) importing the three-dimensional model of urban design to the SuperMap platform; (2.3) adding a coordinate system consistent with the three-dimensional model of oblique photography to load a source of the three-dimensional model data of urban design into the scene; and (2.4) adding an element attribute table through layer attribute editing, to implement storage and management of geometric information and attribute information of each building.

4. The embedded urban design scene emulation method according to claim 3, wherein a specific method for step (3) is as follows: (3.1) matching building objects with a spatial overlapping relationship in three-dimensional models of the three-dimensional model of oblique photography and the three-dimensional model of urban design in the unified coordinate system, generating a corresponding building basal surface in the three-dimensional model of oblique photography and a building basal surface in the three-dimensional model of urban design, and calculating the following three indexes: a basal surface shape similarity, SS for short: $SS = \frac{2 * .Math. {LSI}_{p} - {LSI}_{q} .Math.}{{LSI}_{p} + {LSI}_{q}}$ LSI.sub.p and LSI.sub.q being calculated through the following formulas: $L S I_{p} = \frac{0.2 5 * E 1}{\sqrt{A 1}}$ ${LSI}_{q} = \frac{0.2 5 * E 2}{\sqrt{A 2}}$ wherein LSI.sub.p denotes a building landscape shape index in the three-dimensional model of oblique photography, and LSI.sub.q denotes a building landscape shape index in the three-dimensional model of urban design; E1 denotes a polygon perimeter of the building basal surface in the three-dimensional model of oblique photography, and A1 denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography; E2 denotes a polygon perimeter of the building basal surface in the three-dimensional model of urban design, and A2 denotes a polygon area of the building basal surface in the three-dimensional model of urban design; an overlapping area ratio, OAR for short: $OAR = 1 - \frac{2 * A 3}{A_{p} + A_{q}}$ wherein A.sub.p denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography, A.sub.q denotes a polygon area of the building basal surface in the three-dimensional model of urban design; and A3 denotes an overlapping area between a polygon of the building basal surface in the three-dimensional model of oblique photography and a polygon of the building basal surface in the three-dimensional model of urban design; and a height similarity, HS for short: $HS = \frac{2 * .Math. H_{p} - H_{q} .Math.}{H_{p} + H_{q}}$ wherein H.sub.p denotes a building height in the three-dimensional model of oblique photography, and H.sub.q denotes a building height in the three-dimensional model of urban design; and determining, through comparison, whether the three indexes, SS, OAR, and HS meet preset conditions:
SS<K1
OAR<K2
HS<K3 wherein K1, K2, and K3 are all preset constants, and are generally in a value range of (0, 0.1]; and (3.2) making automatic judgment by taking buildings as a basic unit, marking the building objects with Y if the three all meet matching conditions, and marking the building objects with N if one feature does not meet the matching conditions.

5. The embedded urban design scene emulation method according to claim 3, wherein a method for step (4) is as follows: (4.1) for a region of the buildings marked with N, firstly generating a two-dimensional basal vector surface corresponding to the buildings, taking the basal surfaces as a model flattening range, querying OpenSceneGraph Binary (OSGB) data of an oblique model in the region through a flattened surface, extracting an irregular triangular grid corresponding to the oblique model in the region, keeping the position (X, Y) of the plane unchanged, obtaining a terrain height z′ in the region through an interpolation algorithm, and modifying a height value Z of the triangular grid to z′; and (4.2) performing affine transformation processing on a texture image corresponding to an original triangular grid, changing space positions of original pixels, and linearly transforming three-dimensional coordinates of each pixel in the image, so that the pixels are all vertically projected to the triangular grid with the modified height value and original texture images are all attached to a new triangular grid.

6. The embedded urban design scene emulation method according to claim 3, wherein a method for step (5) is as follows: (5.1) selecting all the building objects marked with Y and editing the building objects into a group, so that the objects as a whole are directly selected by clicking any object; and (5.2) editing the space of the group of the selected marked buildings, and clicking a hide option to hide the marked buildings.

7. The embedded urban design scene emulation method according to claim 6, wherein the two sets of data processed in step 4 and step 5 are connected to a virtual reality device or an external somatosensory device.

8. An embedded urban design scene emulation system, wherein the system comprises a processor that is configured to function as: an oblique photography scene construction module, configured to process oblique photography data in an established range obtained by collection to construct a three-dimensional model of oblique photography, perform object management on buildings in the model, and extract geometric attributes for generation of each building; an urban design model loading module, configured to load a three-dimensional model of urban design in an established range into a three-dimensional model scene of oblique photography, and extract geometric attributes for generation of each building; a building space matching module, configured to perform, for the three-dimensional model scene of oblique photography and three-dimensional model data of urban design in a unified space coordinate system, automatic determination of space matching by taking buildings as a basic unit, matched buildings being marked with Y, and mismatched buildings being marked with N; an oblique photography flattening module, configured to perform, for the buildings marked with N, a flattening operation in the three-dimensional model of oblique photography, so that stereo data of the region is leveled off; an urban design model hiding module, configured to perform, for the buildings marked with Y, real-time space editing in the three-dimensional model of urban design to hide the marked buildings; and a mosaic interaction display module, configured to simultaneously open two sets of data processed by the oblique photography flattening module and the urban design model hiding module to implement mosaic display.

9. The embedded urban design scene emulation system according to claim 8, wherein specific functions of the oblique photography scene construction module are as follows: (1.1) collecting and acquiring oblique photography data not less than an established range, that is, oblique data in an urban design range; (1.2) generating, for the oblique photography data, the three-dimensional model of oblique photography based on real image texture through automatic modeling software; (1.3) loading the three-dimensional model of oblique photography through a SuperMap platform; (1.4) constructing a triangulated irregular network (TIN), and mapping high-resolution images taken from different angles onto a TIN model; and (1.5) extracting a two-dimensional basal surface of a building, and implementing object management on a building model.

10. The embedded urban design scene emulation system according to claim 8, wherein specific functions of the urban design model loading module are as follows: (2.1) editing the three-dimensional model of urban design, clearing the geographic position of the model, and setting latitude and longitude information to zero; (2.2) importing the three-dimensional model of urban design to the SuperMap platform; (2.3) adding a coordinate system consistent with the three-dimensional model of oblique photography to load a source of the three-dimensional model data of urban design into the scene; and (2.4) adding an element attribute table through layer attribute editing, to implement storage and management of geometric information and attribute information of each building.

11. The embedded urban design scene emulation system according to claim 8, wherein specific functions of the building space matching module are as follows: (3.1) matching building objects with a spatial overlapping relationship in three-dimensional models of the three-dimensional model of oblique photography and the three-dimensional model of urban design in the unified coordinate system, generating a corresponding building basal surface in the three-dimensional model of oblique photography and a building basal surface in the three-dimensional model of urban design, and calculating the following three indexes: a basal surface shape similarity, SS for short: $SS = \frac{2 * .Math. {LSI}_{p} - L S I_{q} .Math.}{{LSI}_{p} + {LSI}_{q}}$ LSI.sub.p and LSI.sub.q being calculated through the following formulas: ${LSI}_{p} = \frac{0.2 5 * E 1}{\sqrt{A 1}}$ $L S I_{q} = \frac{0.2 5 * E 2}{\sqrt{A 2}}$ wherein LSI.sub.p denotes a building landscape shape index in the three-dimensional model of oblique photography, and LSI.sub.q denotes a building landscape shape index in the three-dimensional model of urban design; E1 denotes a polygon perimeter of the building basal surface in the three-dimensional model of oblique photography, and A1 denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography; E2 denotes a polygon perimeter of the building basal surface in the three-dimensional model of urban design, and A2 denotes a polygon area of the building basal surface in the three-dimensional model of urban design; an overlapping area ratio, OAR for short: $OAR = 1 - \frac{2 * A 3}{A_{p} + A_{q}}$ wherein A.sub.p denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography, A.sub.q denotes a polygon area of the building basal surface in the three-dimensional model of urban design; and A3 denotes an overlapping area between a polygon of the building basal surface in the three-dimensional model of oblique photography and a polygon of the building basal surface in the three-dimensional model of urban design; and a height similarity, HS for short: $HS = \frac{2 * .Math. H_{p} - H_{q} .Math.}{H_{p} + H_{q}}$ wherein H.sub.p denotes a building height in the three-dimensional model of oblique photography, and H.sub.q denotes a building height in the three-dimensional model of urban design; and determining, through comparison, whether the three indexes, SS, OAR, and HS meet preset conditions:
SS<K1
OAR<K2
HS<K3 wherein K1, K2, and K3 are all preset constants, and are generally in a value range of (0, 0.1); and (3.2) making automatic judgment by taking buildings as a basic unit, marking the building objects with Y if the three all meet matching conditions, and marking the building objects with N if one feature does not meet the matching conditions.

12. The embedded urban design scene emulation system according to claim 11, wherein specific functions of the oblique photography flattening module are as follows: (4.1) for a region of the buildings marked with N, firstly generating a two-dimensional basal vector surface corresponding to the buildings, taking the basal surfaces as a model flattening range, querying OpenSceneGraph Binary (OSGB) data of an oblique model in the region through a flattened surface, extracting an irregular triangular grid corresponding to the oblique model in the region, keeping the position (X, Y) of the plane unchanged, obtaining a terrain height z′ in the region through an interpolation algorithm, and modifying a height value Z of the triangular grid to z′; and (4.2) performing affine transformation processing on a texture image corresponding to an original triangular grid, changing space positions of original pixels, and linearly transforming three-dimensional coordinates of each pixel in the image, so that the pixels are all vertically projected to the triangular grid with the modified height value and original texture images are all attached to a new triangular grid.

13. The embedded urban design scene emulation system according to claim 11, wherein specific functions of the urban design model hiding module are as follows: (5.1) selecting all the building objects marked with Y and editing the building objects into a group, so that the objects as a whole are directly selected by clicking any object; and (5.2) editing the space of the group of the selected marked buildings, and clicking a hide option to hide the marked buildings.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a technical flowchart of an embedded urban design scene emulation method according to the present invention;

(2) FIG. 2 is a diagram illustrating loading of a three-dimensional model of oblique photography to a SuperMap platform according to the present invention;

(3) FIG. 3 is a diagram illustrating conversion of oblique photography data into a grid surface model according to the present invention;

(4) FIG. 4 is a profile of an extracted two-dimensional basal surface of a building according to the present invention;

(5) FIG. 5 is a diagram illustrating building model object management in an oblique photography scene according to the present invention;

(6) FIG. 6 is a diagram illustrating a three-dimensional model of urban design according to the present invention;

(7) FIG. 7 is a diagram illustrating space matching on buildings according to the present invention;

(8) FIG. 8 is a comparison diagram of a three-dimensional model of oblique photography before and after pressing according to the present invention; and

(9) FIG. 9 is a diagram illustrating mosaic display of a three-dimensional model of oblique photography and a three-dimensional model of urban design according to the present invention.

DETAILED DESCRIPTION

(10) The technical solutions of the present invention are further described below with reference to the accompanying drawings and embodiments.

(11) The present invention provides an embedded urban design scene emulation method, the method including the following steps.

(12) In step 1, for oblique photography data in an established range obtained by collection, scene construction is performed through a series of technologies such as platform loading, geometric correction, and data thinning to form a three-dimensional model of oblique photography, and object management and feature extraction on a building model are implemented. The feature extraction means extracting geometric attributes for generation of each building.

(13) In step 2, coordinate system transformation is performed on a three-dimensional model of urban design in an established range, three-dimensional model data of urban design is loaded into a three-dimensional model scene of oblique photography, and geometric attributes for generation of each building are extracted. The three-dimensional model of urban design is one of the achievement forms of the urban design business. For a region corresponding to a design range, elements such as road, blocks, and buildings therein are modeled. The three-dimensional model of urban design referred to in this application refers in particular to buildings.

(14) In step 3, for the three-dimensional model scene of oblique photography and three-dimensional model data of urban design in a unified space coordinate system, automatic determination of space matching is performed by taking buildings as a basic unit, matched buildings are marked with Y, and mismatched buildings are marked with N.

(15) In step 4, for the buildings marked with N, a flattening operation is performed in the three-dimensional model of oblique photography, so that stereo data of the region is leveled off.

(16) In step 5, for the buildings marked with Y, real-time space editing is performed in the three-dimensional model of urban design to hide the marked buildings.

(17) In step 6, two sets of data processed in step 4 and step 5 are simultaneously opened to implement mosaic display. Moreover, an effect of urban design scene emulation may be further enhanced through interaction between virtual reality devices and external somatosensory devices.

(18) In step 1, for oblique photography data in an established range obtained by collection, scene construction is performed through a series of technologies such as platform loading, geometric correction, and data thinning to form a three-dimensional model of oblique photography, and object management and feature extraction on a building model are implemented.

(19) (1.1) Oblique photography data not less than an established range, that is, oblique data in an urban design range, is collected and acquired. Oblique photogrammetry refers to simultaneously collecting images at 1 vertical angle and 4 oblique angles by carrying a multi-lens camera unit on a flight platform. The flight platform is, for example, a multi-rotor unmanned aerial vehicle, a fixed-wing unmanned aerial vehicle, or a vertical take-off and landing unmanned aerial vehicle.

(20) (1.2) The three-dimensional model of oblique photography based on real image texture is generated for the oblique photography data through automatic modeling software. The generated three-dimensional model of oblique photography is constructed by obtaining omni-directional information data of ground objects by performing a series of processing, such as geometric correction, joint adjustment, and multi-view image matching, on the photography data through the automatic modeling software. The modeling software may be VirtualGeo software developed by DIGINEXT in France, or EFSElectronic Field Study software from Pictometry in the US.

(21) (1.3) Load the three-dimensional model of oblique photography through a SuperMap platform

(22) The SuperMap platform uses LOD to optimize scheduling, which only takes up fewer hardware resources to ensure a stable capacity to bear massive data, and meanwhile, supports direct loading of oblique photography models of any subdivision type, including formats such as .osg/.osgb, .x, .dae, and .obj. The platform may generate, through a configuration file generation function, files in a *.scp format according to a plurality of pieces of oblique photography model data in a *.osgb format that are stored in a plurality of folders. The files record model configuration contents such as relative paths, names, insertion point positions, and coordinate system information of oblique photography model files. The platform implements direct batch loading and browsing of OSGB model data by loading three-dimensional model cache files in a *.scp format.

(23) (1.4) A TIN is constructed, and high-resolution images taken from different angles are mapped onto a TIN model.

(24) A DSM including various ground information such as terrain and buildings by using the three-dimensional model of oblique photography loaded to the scene. Then, all eligible contour lines in the digital surface model are extracted by inputting a specified height value. After acquisition of the contour lines, line data is converted into surface data through data transformation. A Douglas-Peucker node thinning algorithm is performed on the surface data to simplify a boundary line of the surface to form a grid surface model.

(25) Point cloud data of the grid surface model is thinned and a continuous TIN is constructed by using a Delaunay triangulation algorithm. High-resolution images taken from different angles are mapped onto a TIN model through multi-view image matching. The high-resolution images are image data taken by oblique photography. Texture refers to features on the images, such as buildings.

(26) (1.5) Building model object management:

(27) A two-dimensional basal surface of a building is extracted, and object management is implemented on a building model. A specific method is as follows.

(28) In a TIN-format grid surface model, a two-dimensional basal surface of each building is outlined by using a multi-segment line command to successively connect vector points at a junction between the building and the ground, that is, endpoints in a triangulation network. For a building, the two-dimensional basal surface of the building and a vector surface of the triangulation network projected by the plane within the basal surface are combined together to form a single unit on which object management can be performed, so as to implement storage and management of geometric attribute information of each single building unit.

(29) To implement the object management on a building model, the platform implements single-unit processing and attribute connection on all kinds of ground objects by generating a vector surface matching the model.

(30) In step 2, coordinate system transformation is performed on a three-dimensional model of urban design in an established range, a three-dimensional model of urban design is located into a three-dimensional model of oblique photography, and geometric attributes for generation of each building are extracted.

(31) (2.1) Edit the three-dimensional model of urban design

(32) The three-dimensional model of urban design is one of the achievement forms of the urban design business. For a region corresponding to a design range, elements such as road, blocks, and buildings therein are modeled. The three-dimensional model of urban design referred to in this application refers in particular to buildings. A high-precision three-dimensional model of urban design achievements in an established range is edited by using SketchUp modeling software, the geographic position of the model is cleared, and latitude and longitude information is set to zero.

(33) (2.2) Import the three-dimensional model of urban design to the SuperMap platform

(34) The SuperMap platform supports import of mainstream model data, including formats such as *.osg, *.osgb, *.dae, *.obj, *.ifc, *.3ds, *.dxf, *.fbx, *.x, and *.flt. The mainstream model data is directly imported to a model dataset and then is converted into system-supported UDB format data.

(35) (2.3) Coordinate system transformation

(36) A coordinate system consistent with the three-dimensional model of oblique photography, which is generally a 2000 Chinese Geodetic Coordinate System, is added to load a source of the three-dimensional model data of urban design into the scene. Through a model editing tool, a model and a reference point in the corresponding scene are selected, a coordinate offset of the reference point is inputted, and the model is entirely moved to the actual position in the scene.

(37) (2.4) Add geometric attributes of buildings

(38) An element attribute table is added through layer attribute editing, to implement storage and management of geometric information and attribute information of each building.

(39) In step 3, for the scene of oblique photography and three-dimensional model data of urban design in a unified space coordinate system, automatic determination of space matching is performed by taking buildings as a basic unit, matched buildings are marked with Y, and mismatched buildings are marked with N.

(40) (3.1) Index calculation

(41) For three-dimensional models of the three-dimensional model of oblique photography and the three-dimensional model of urban design in the unified coordinate system, oblique models and three-dimensional model objects (such as buildings) with a spatial overlapping relationship are matched, a corresponding building basal surface in the three-dimensional model of oblique photography and a building basal surface in the three-dimensional model of urban design are generated, and three indexes, which are a basal surface shape similarity, an overlapping area ratio, and a building height feature similarity respectively, are calculated.

(42) Basal surface shape similarity, SS for short: A value closer to 0 indicates higher shape similarity, and the SS describes the complexity of shape features by calculating a degree of deviation between the shape of a polygon and a square of the same area.

(43) $SS = \frac{2 * .Math. {LSI}_{p} - {LSI}_{q} .Math.}{{LSI}_{p} + {LSI}_{q}}$

(44) LSI.sub.p and LSI.sub.q are calculated through the following formulas:

(45) 0 ${LSI}_{p} = \frac{0.2 5 * E 1}{\sqrt{A 1}}$ ${LSI}_{q} = \frac{0.2 5 * E 2}{\sqrt{A 2}}$

(46) where LSI.sub.p denotes a building landscape shape index in the three-dimensional model of oblique photography, and LSI.sub.q denotes a building landscape shape index in the three-dimensional model of urban design; E1 denotes a polygon perimeter of the building basal surface in the three-dimensional model of oblique photography, and A1 denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography; E2 denotes a polygon perimeter of the building basal surface in the three-dimensional model of urban design, and A2 denotes a polygon area of the building basal surface in the three-dimensional model of urban design. The landscape shape index is LSI for short.

(47) Overlapping area ratio, OAR for short: In the following formula, the OAR is obtained by calculating a percentage of a polygon space superposed overlapping area on the basal surface, and A value closer to 0 indicates closer positions,

(48) $OAR = 1 - \frac{2 * A 3}{A_{p} + A_{q}}$

(49) where A.sub.p denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography, A.sub.q denotes a polygon area of the building basal surface in the three-dimensional model of urban design; and A.sub.p∩A.sub.q is used to denote an overlapping area between a polygon of the building basal surface in the three-dimensional model of oblique photography and a polygon of the building basal surface in the three-dimensional model of urban design, and the overlapping area is denoted as A3.

(50) Height similarity, HS for short: A value closer to 0 indicates higher height similarity;

(51) $HS = \frac{2 * .Math. H_{p} - H_{q} .Math.}{H_{p} + H_{q}}$

(52) where H.sub.p denotes a building height in the three-dimensional model of oblique photography, and H.sub.q denotes a building height in the three-dimensional model of urban design; and

(53) A space matching determination method is as follows: determining, through comparison, whether the three indexes, SS, OAR, and HS meet preset conditions:
SS<K1
OAR<K2
HS<K3
where K1, K2, and K3 are all preset constants, and are generally in a value range of (0, 0.1].

(54) Automatic judgment is made by taking buildings as a basic unit, the building objects are marked with Y if the three all meet matching conditions, and the building objects are marked with N if one feature does not meet the matching conditions.

(55) In step 4, for the buildings marked with N, a local flattening operation is performed on the three-dimensional model scene of oblique photography, so that stereo data of the region is leveled off.

(56) (4.1) Modify the z value of a TIN grid on a two-dimensional basal surface of a building

(57) For a region of the buildings marked with N, a two-dimensional basal vector surface corresponding to the buildings is firstly generated, the basal surfaces are taken as a model flattening range, OSGB data of an oblique model in the region is queried through a flattened surface, a TIN corresponding to the oblique model in the region is extracted, the position (X, Y) of the plane is kept unchanged, a terrain height z′ in the region is obtained through an interpolation algorithm, for example, a Kriging interpolation algorithm, and a height value Z of the TIN grid is modified to z′.

(58) (4.2) Attach original texture images to a new TIN grid through affine transformation

(59) Affine transformation processing in geometric transformation is performed on a texture image corresponding to an original TIN network, space positions of original pixels are changed, and three-dimensional coordinates of each pixel in the image are linearly transformed, so that the pixels are all vertically projected to the TIN grid with the modified height value. Therefore, original texture images can be all attached to the new TIN grid.

(60) Through the foregoing steps, a flattening operation on an oblique photography model of the region can be implemented, and a scheme model newly added to the design is displayed. The scheme model herein refers to buildings in an urban design model.

(61) In step 5, for the buildings marked with Y, real-time space editing is performed in the three-dimensional model of urban design to hide the marked buildings.

(62) (5.1) Establish an object group

(63) All the building objects marked with Y are selected and edited into a group, so that the objects as a whole may be directly selected by clicking any object.

(64) (5.2) The space of the group of the selected marked buildings is edited, and a hide option is clicked to hide the marked buildings.

(65) In step 6, two sets of data processed in step 4 and step 5 are simultaneously opened to implement mosaic display, and an effect of urban design scene emulation can be further enhanced through interaction between virtual reality devices and external somatosensory devices.

(66) (6.1) Mosaic display

(67) The two sets of data processed in step 4 and step 5 are simultaneously opened to implement mosaic display.

(68) (6.2) Optional connection to a virtual reality device

(69) A scene is set, a stereo mode is started, a virtual reality (VR) device such as HTC Vive or Oculus Rift is connected, and free browsing is performed in a manner such as keyboard operation, automatic walking, or automatic running, to create a real three-dimensional immersive experience.

(70) (6.3) Optional connection to an external somatosensory device

(71) A scene is set, a stereo mode is started, and human body movement changes are dynamically captured in real time through connection to an external somatosensory device such as Microsoft Kinect or Leap Motion, and are automatically converted into three-dimensional operation instructions, so as to control traveling directions and attitudes of movement of observation points in the three-dimensional scene.

(72) In addition, the present invention further provides an embedded urban design scene emulation system, the system including the following modules:

(73) an oblique photography scene construction module, configured to process oblique photography data in an established range obtained by collection to construct a three-dimensional model of oblique photography, perform object management on buildings in the model, and extract geometric attributes for generation of each building;

(74) an urban design model loading module, configured to load a three-dimensional model of urban design in an established range into a three-dimensional model scene of oblique photography, and extract geometric attributes for generation of each building;

(75) a building space matching module, configured to perform, for the three-dimensional model scene of oblique photography and three-dimensional model data of urban design in a unified space coordinate system, automatic determination of space matching by taking buildings as a basic unit, matched buildings being marked with Y, and mismatched buildings being marked with N;

(76) an oblique photography flattening module, configured to perform, for the buildings marked with N, a local flattening operation in the scene of oblique photography, so that stereo data of the region is leveled off;

(77) an urban design model hiding module, configured to perform, for the buildings marked with Y, real-time space editing in the three-dimensional model of urban design to hide the marked buildings; and

(78) a mosaic interaction display module, configured to simultaneously open two sets of data processed by the oblique photography flattening module and the urban design model hiding module to implement mosaic display.

(79) Specific functions of the oblique photography scene construction module are as follows:

(80) (1.1) collecting and acquiring oblique photography data not less than an established range, that is, oblique data in an urban design range;

(81) (1.2) generating, for the oblique photography data, the three-dimensional model of oblique photography based on real image texture through automatic modeling software;

(82) (1.3) loading the three-dimensional model of oblique photography through a SuperMap platform;

(83) (1.4) constructing a TIN, and mapping high-resolution images taken from different angles onto a TIN model; and

(84) (1.5) extracting a two-dimensional basal surface of a building, further to implement object management on a building model.

(85) Specific functions of the urban design model loading module are as follows:

(86) (2.1) editing the three-dimensional model of urban design, clearing the geographic position of the model, and setting latitude and longitude information to zero;

(87) (2.2) importing the three-dimensional model of urban design to the SuperMap platform;

(88) (2.3) adding a coordinate system consistent with the three-dimensional model of oblique photography to load a source of the three-dimensional model data of urban design into the scene; and

(89) (2.4) adding an element attribute table through layer attribute editing, to implement storage and management of geometric information and attribute information of each building.

(90) Specific functions of the building space matching module are as follows:

(91) (3.1) matching building objects with a spatial overlapping relationship in three-dimensional models of the three-dimensional model of oblique photography and the three-dimensional model of urban design in the unified coordinate system, generating a corresponding building basal surface in the three-dimensional model of oblique photography and a building basal surface in the three-dimensional model of urban design, and calculating the following three indexes:

(92) a basal surface shape similarity, SS for short:

(93) $SS = \frac{2 * .Math. {LSI}_{p} - {LSI}_{q} .Math.}{{LSI}_{p} + {LSI}_{q}}$

(94) LSI.sub.p and LSI.sub.q being calculated through the following formulas:

(95) ${LSI}_{p} = \frac{0.2 5 * E 1}{\sqrt{A 1}}$ ${LSI}_{q} = \frac{0.2 5 * E 2}{\sqrt{A 2}}$

(96) where LSI.sub.p denotes a building landscape shape index in the three-dimensional model of oblique photography, and LSI.sub.q denotes a building landscape shape index in the three-dimensional model of urban design; E1 denotes a polygon perimeter of the building basal surface in the three-dimensional model of oblique photography, and A1 denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography; E2 denotes a polygon perimeter of the building basal surface in the three-dimensional model of urban design, and A2 denotes a polygon area of the building basal surface in the three-dimensional model of urban design; an overlapping area ratio, OAR for short:

(97) $OAR = 1 - \frac{2 * A 3}{A_{p} + A_{q}}$

(98) where A.sub.p denotes a polygon area of the building basal surface in the three-dimensional model of oblique photography, A.sub.q denotes a polygon area of the building basal surface in the three-dimensional model of urban design; and A3 denotes an overlapping area between a polygon of the building basal surface in the three-dimensional model of oblique photography and a polygon of the building basal surface in the three-dimensional model of urban design; and

(99) a height similarity, HS for short:

(100) $HS = \frac{2 * .Math. H_{p} - H_{q} .Math.}{H_{p} + H_{q}}$

(101) where H.sub.p denotes a building height in the three-dimensional model of oblique photography, and H.sub.q denotes a building height in the three-dimensional model of urban design; and

(102) determining, through comparison, whether the three indexes, SS, OAR, and HS meet preset conditions:
SS<K1
OAR<K2
HS<K3

(103) where K1, K2, and K3 are all preset constants, and are generally in a value range of (0, 0.1]; and

(104) (3.2) making automatic judgment by taking buildings as a basic unit, marking the building objects with Y if the three all meet matching conditions, and marking the building objects with N if one feature does not meet the matching conditions.

(105) Specific functions of the oblique photography flattening module are as follows:

(106) (4.1) for a region of the buildings marked with N, firstly generating a two-dimensional basal vector surface corresponding to the buildings, taking the basal surfaces as a model flattening range, querying OSGB data of an oblique model in the region through a flattened surface, extracting an irregular triangular grid corresponding to the oblique model in the region, keeping the position (X, Y) of the plane unchanged, obtaining a terrain height z′ in the region through an interpolation algorithm, and modifying a height value Z of the triangular grid to z′; and

(107) (4.2) performing affine transformation processing on a texture image corresponding to an original triangular grid, changing space positions of original pixels, and linearly transforming three-dimensional coordinates of each pixel in the image, so that the pixels are all vertically projected to the triangular grid with the modified height value and original texture images are all attached to a new triangular grid.

(108) Specific functions of the urban design model hiding module are as follows:

(109) (5.1) selecting all the building objects marked with Y and editing the building objects into a group, so that the objects as a whole may be directly selected by clicking any object; and

(110) (5.2) editing the space of the group of the selected marked buildings, and clicking a hide option to hide the marked buildings.

Embedded urban design scene emulation method and system

Assignee

Inventors

Cpc classification

Classification Explorer

G06T7/60

PHYSICS

Classification Explorer

G06T7/40

PHYSICS

Classification Explorer

G06T19/20

PHYSICS

Classification Explorer

G06T2200/08

PHYSICS

Classification Explorer

Y02A30/60

GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS

Classification Explorer

G06T17/05

PHYSICS

Classification Explorer

G06T17/00

PHYSICS

Classification Explorer

G06T2219/028

PHYSICS

Classification Explorer

G06T2219/2004

PHYSICS

Classification Explorer

G06T2219/2021

PHYSICS

International classification

Classification Explorer

G06T17/05

PHYSICS

Classification Explorer

G06T19/20

PHYSICS

Classification Explorer

G06T7/40

PHYSICS

Classification Explorer

G06T7/60

PHYSICS

Classification Explorer

G06T17/00

PHYSICS

Abstract

Claims

Description