CN113706698B - Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment - Google Patents

Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment Download PDF

Info

Publication number
CN113706698B
CN113706698B CN202111242316.8A CN202111242316A CN113706698B CN 113706698 B CN113706698 B CN 113706698B CN 202111242316 A CN202111242316 A CN 202111242316A CN 113706698 B CN113706698 B CN 113706698B
Authority
CN
China
Prior art keywords
road
point cloud
dimensional
area
point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202111242316.8A
Other languages
Chinese (zh)
Other versions
CN113706698A (en
Inventor
张学全
吴红燕
罗云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Huancheng Jingwei Technology Co ltd
Original Assignee
Wuhan Huancheng Jingwei Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan Huancheng Jingwei Technology Co ltd filed Critical Wuhan Huancheng Jingwei Technology Co ltd
Priority to CN202111242316.8A priority Critical patent/CN113706698B/en
Publication of CN113706698A publication Critical patent/CN113706698A/en
Application granted granted Critical
Publication of CN113706698B publication Critical patent/CN113706698B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/005—Tree description, e.g. octree, quadtree
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/18—Network design, e.g. design based on topological or interconnect aspects of utility systems, piping, heating ventilation air conditioning [HVAC] or cabling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/006—Mixed reality

Abstract

The invention discloses a real-scene three-dimensional road reconstruction method and device, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model; extracting point cloud data of a road area from the point cloud model along the road vector line; extracting road geometric features of a road specific area from the point cloud data, and extracting texture features corresponding to the road geometric area from the orthographic projection image; the method comprises the steps of reconstructing a three-dimensional road, sequentially performing texture mapping on the three-dimensional road by adopting texture characteristics, performing accessory facility modeling and semantic information modeling to generate a real three-dimensional road, and performing index organization on the three-dimensional road based on a global tile pyramid structure and a quadtree subdivision. The method and the device solve the technical problems of complex processing, high cost and low semantic degree of the real three-dimensional road construction in the related technology.

Description

Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment
Technical Field
The invention relates to the field of computers, in particular to a real-scene three-dimensional road reconstruction method and device, a storage medium and electronic equipment.
Background
In the related technology, the live-action three-dimension is a digital virtual space for real, three-dimensional and time-series reflection and expression of human production, life and ecological space, is a novel basic mapping standardized product, is an important component of national novel infrastructure construction, and provides a uniform space base for economic and social development and informatization of various departments. The live-action three-dimension is constructed by bearing a geographic entity which is structured, semanticized, supports human-computer compatible understanding and real-time perception of the internet of things on a three-dimensional geographic scene.
The road is a passage for daily travel and material transportation of residents and is a traffic hub which is vital to urban operation. The construction of the urban road live-action three-dimensional model plays an important role in urban management, traffic simulation, emergency evacuation and the like. Road three-dimensional modeling based on unmanned aerial vehicle oblique photogrammetry, because the restriction of aerial photography overlook and downward looking shooting angle and the sheltering from of ground objects on two sides of the road, the image acquisition is susceptible to aerial photography blind areas and the influence of matching errors, resulting in phenomena such as partial road surface unevenness, structural adhesion, white wall hole breaking, model distortion and texture deletion. Meanwhile, the oblique photography three-dimensional scene is a whole, and the concrete elements on the road are difficult to identify and query to realize the singleness. And the problems form a challenge for high-precision real-scene three-dimensional road construction and related applications.
The three-dimensional road fine modeling method in the related art is based on vehicle-mounted laser point cloud modeling, and has the advantages of high road model precision, but also has the following defects: (1) the modeling is complex. Laser point cloud scenes usually contain millions of point data, and a series of complex feature extraction operations are required to extract road structure grids. (2) The modeling cost is high. The current laser measuring equipment and processing software are high in cost, and the application of the laser point cloud is limited. Therefore, the existing laser point cloud road modeling method is difficult to be applied in a large range due to the problems of complexity, cost and the like.
In view of the above problems in the related art, no effective solution has been found at present.
Disclosure of Invention
The embodiment of the invention provides a real-scene three-dimensional road reconstruction method and device, a storage medium and electronic equipment.
According to an aspect of an embodiment of the present application, there is provided a real-scene three-dimensional road reconstruction method, including: acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model; extracting point cloud data of a road area from the point cloud model along a road vector line; extracting road geometric features of a road specific area from the point cloud data, and extracting texture features corresponding to the road geometric area from the orthophoto map; the method comprises the steps of carrying out segmentation processing on road vector lines, carrying out triangulation network construction on each segment by adopting road geometric features, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture features, carrying out accessory facility modeling and semantic information modeling, and generating a real three-dimensional road, and carrying out index organization on the three-dimensional road based on an earth tile pyramid structure and a quadtree subdivision so as to realize the road three-dimensional visualization of a target area.
Further, extracting point cloud data of a road region from the point cloud model along a road vector line comprises: performing buffer area cutting according to the preset buffer area radius based on the road center line vector in the preset two-dimensional data, and filtering the point cloud model to obtain a first road area point cloud; acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and acquiring the road surface point cloud height of the second road area point cloud, and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road surface point cloud height.
Further, the road geometric features comprise road shoulder lines, and extracting the road geometric features of the specific road area from the point cloud data comprises: partitioning the point cloud data, taking a rectangular block on a road central line in the road vector line as a seed point, judging whether an isolation zone exists in the middle of the road according to the seed point, and marking an isolation zone area; respectively carrying out growth search to two sides perpendicular to the road center line along the seed points until the average height of the seed points is higher than the road surface height to obtain road shoulder blocks where the road shoulder lines are located; acquiring point cloud data in the road shoulder blocks and the adjacent blocks thereof, reducing the size of the blocks by half, continuing to perform growth search on the road shoulder blocks, and extracting to obtain road shoulder line point cloud; and generating a road shoulder line according to the road shoulder line point cloud.
Further, generating a road shoulder line from the road shoulder line point cloud comprises: aiming at the straight road section point cloud, the characteristic point p in the road shoulder line point cloud is processed based on the following formulaiPerforming shoulder line smoothing to generate a straight shoulder line:
Figure 557171DEST_PATH_IMAGE001
；
wherein, Pi(t) is, t is a sampling parameter, pi+1Is a characteristic point piAdjacent point of (a), pi+2Is a characteristic point pi+1Adjacent points of (a);
aiming at a turning road section point cloud, a local coordinate point P (x) of the ith sampling point P in the road shoulder line point cloud is converted based on the following conversion algorithmi，yi) Converting into a global Cartesian coordinate point P', and generating a curved road shoulder line:
Figure 694892DEST_PATH_IMAGE002
；
wherein the content of the first and second substances,
Figure 953835DEST_PATH_IMAGE003
is a global coordinate, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 36060DEST_PATH_IMAGE004
and
Figure 241914DEST_PATH_IMAGE005
is a global coordinate direction vector parameter, scaleX and scaleY are scaling parameters.
Further, the road geometric characteristics are adopted to carry out triangulation network construction on each subsection, and a three-dimensional road is reconstructed, and the method comprises the following steps: for a single straight road or curved road in the road vector line, selecting a first vertex of each segment from the road shoulder line characteristic points of the road geometric characteristics as an origin to construct a local Cartesian coordinate system; and loading the segmented road model onto the three-dimensional map by adopting a scaling matrix Mscaling, the rotation matrix Mrotation and the translation matrix Mtranslation.
Further, triangulating each segment using the road geometry includes: judging the type of an intersection road in the road vector line based on the road central point vector data; and constructing a road triangulation network by adopting shoulder line characteristic points, shoulder heights and pedestrian path widths according to the intersection types of the intersection roads, wherein the road geometric characteristics comprise the shoulder lines and the pedestrian path widths, and the shoulder heights are obtained from the digital earth surface model of the oblique photography data.
Further, the texture features are adopted to sequentially carry out texture mapping on the three-dimensional road, and accessory facility modeling and semantic information modeling are carried out, and the method comprises the following steps: determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 30878DEST_PATH_IMAGE006
；
determining the number of the instantiated auxiliary facilities, calculating the area of a target area of a corresponding example according to the number of the examples, and searching the next Level matched with the examples of the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 539351DEST_PATH_IMAGE007
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
and acquiring semantic attribute information, and overloading the semantic attribute information to the model component of the three-dimensional road.
According to another aspect of the embodiments of the present application, there is also provided a live-action three-dimensional road reconstruction apparatus, including: the system comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model; the filtering module is used for extracting point cloud data of a road area from the point cloud model along a road vector line; the extraction module is used for extracting the road geometric characteristics of a road specific area from the point cloud data and extracting the texture characteristics corresponding to the road geometric area from the orthophoto map; the reconstruction module is used for carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture characteristics, carrying out accessory facility modeling and semantic information modeling, and generating a live-action three-dimensional road; and the organization module is used for carrying out index organization on the three-dimensional roads based on the earth tile pyramid structure and the quadtree subdivision so as to realize the road three-dimensional visualization of the target area.
Further, the filtration module comprises: the first filtering unit is used for performing buffer area cutting according to the preset buffer area radius based on the road center line vector in the preset two-dimensional data, and filtering the point cloud model to obtain a first road area point cloud; the second filtering unit is used for acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and the third filtering unit is used for acquiring the road pavement point cloud height of the second road area point cloud and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road pavement point cloud height.
Further, the road geometric feature includes a road shoulder line, and the extraction module includes: the processing unit is used for partitioning the point cloud data, taking a rectangular block on a road central line in the road vector line as a seed point, judging whether an isolation zone exists in the middle of the road according to the seed point, and marking an isolation zone area; the searching unit is used for respectively carrying out growth searching on two sides perpendicular to the road center line along the seed points until the average height of the seed points is higher than the road surface height to obtain road shoulder blocks where the road shoulder lines are located; the extraction unit is used for acquiring point cloud data in the road shoulder blocks and the adjacent blocks thereof, and continuing to perform growth search on the road shoulder blocks after the sizes of the blocks are halved to extract road shoulder line point clouds; and the generating unit is used for generating the road shoulder line according to the road shoulder line point cloud.
Further, the generation unit includes: a processing subunit, configured to, for a straight road segment point cloud, pair the feature points p in the road shoulder line point cloud based on the following formulaiPerforming shoulder line smoothing to generate a straight shoulder line:
Figure 46556DEST_PATH_IMAGE008
；
wherein, Pi(t) is, t is a sampling parameter, pi+1Is a characteristic point piAdjacent point of (a), pi+2Is a characteristic point pi+1Adjacent point of (2)；
A conversion subunit, configured to, for a turning road segment point cloud, convert a local coordinate point P (x) of an ith sampling point P in the shoulder line point cloud based on the following conversion algorithmi，yi) Converting into a global Cartesian coordinate point P', and generating a curved road shoulder line:
Figure 524941DEST_PATH_IMAGE002
；
wherein the content of the first and second substances,
Figure 761888DEST_PATH_IMAGE003
is a global coordinate, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 362633DEST_PATH_IMAGE004
and
Figure 294817DEST_PATH_IMAGE005
is a global coordinate direction vector parameter, scaleX and scaleY are scaling parameters.
Further, the reconstruction module includes: the first construction unit is used for selecting the first vertex of each segment from the road shoulder line characteristic points of the road geometric characteristics as an origin to construct a local Cartesian coordinate system for a single straight road or curved road in a road vector line; and the loading unit is used for loading the road model of each segment onto the three-dimensional map by adopting a scaling matrix, a rotation matrix and a translation matrix.
Further, the reconstruction module includes: the judging unit is used for judging the type of the intersection on the basis of the road central point vector data for the intersection road in the road vector line; and the second construction unit is used for constructing a road triangulation network by adopting the feature points of the road shoulder lines, the road shoulder height and the width of the pedestrian path according to the type of the intersection road, wherein the geometrical features of the road comprise the road shoulder lines and the width of the pedestrian path, and the road shoulder height is obtained from the digital earth surface model of the oblique photography data.
Further, the reconstruction module includes: a calculating unit for determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 639211DEST_PATH_IMAGE009
；
the searching unit is used for determining the number of the instantiated auxiliary facilities, calculating the area of the target area of the corresponding example according to the number of the examples, and searching the next Level matched with the examples of the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 481396DEST_PATH_IMAGE007
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
and the overload unit is used for acquiring semantic attribute information and overloading the semantic attribute information to the model component of the three-dimensional road.
According to another aspect of the embodiments of the present application, there is also provided a storage medium including a stored program that executes the above steps when the program is executed.
According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus; wherein: a memory for storing a computer program; a processor for executing the steps of the method by running the program stored in the memory.
Embodiments of the present application also provide a computer program product containing instructions, which when run on a computer, cause the computer to perform the steps of the above method.
By the invention, an oblique photography image of a target area is obtained, an oblique three-dimensional model is generated based on the oblique photography image, a point cloud model and an orthographic projection image are generated according to the oblique three-dimensional model, point cloud data of a road area is extracted from the point cloud model along a road vector line, road geometric characteristics of a specific area of the road are extracted from the point cloud data, texture characteristics corresponding to the road geometric area are extracted from the orthographic projection image, the road vector line is processed in a segmentation mode, triangulation is carried out on each segmentation by adopting the road geometric characteristics, a three-dimensional road is reconstructed, texture mapping is carried out on the three-dimensional road by adopting the texture characteristics in sequence, subsidiary facility modeling and semantic information modeling are carried out, a real three-dimensional road is generated, and index organization is carried out on the three-dimensional road based on an earth tile pyramid structure and a quadtree subdivision so as to realize the road three-dimensional visualization of the target area, the method has the advantages that the real-scene three-dimensional road is reconstructed on the point cloud model and the orthophoto map corresponding to the oblique photography image, laser point cloud road modeling and vector road automatic modeling are combined, the technical problems of high complexity and high cost of real-scene three-dimensional road construction in the related technology are solved, the processing speed is high, the modeling cost is low, and the modeling effect is good.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
FIG. 1 is a block diagram of a hardware configuration of a computer according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for reconstructing a live-action three-dimensional road according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of data preprocessing according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a road surface region extraction geometry according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a single road partition geometric modeling of an embodiment of the present invention;
FIG. 6 is a schematic diagram of a road intersection zoning geometric modeling of an embodiment of the present invention;
FIG. 7 is a schematic diagram of a road partition texture map in accordance with an embodiment of the present invention;
FIG. 8 is a block organization diagram of a large-scale road model according to an embodiment of the present invention;
FIG. 9 is an effect diagram of the embodiment of the invention after the three-dimensional reconstruction of the live-action road;
fig. 10 is a block diagram of a real three-dimensional road reconstruction apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
The method provided by the first embodiment of the application can be executed in a computer, a mobile phone, an unmanned aerial vehicle or a similar operation device. Taking an example of the present invention running on a computer, fig. 1 is a block diagram of a hardware structure of a computer according to an embodiment of the present invention. As shown in fig. 1, the computer may include one or more (only one shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and optionally, a transmission device 106 for communication functions and an input-output device 108. It will be appreciated by those of ordinary skill in the art that the configuration shown in FIG. 1 is illustrative only and is not intended to limit the configuration of the computer described above. For example, a computer may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be used to store a computer program, for example, a software program and a module of an application software, such as a computer program corresponding to a real three-dimensional road reconstruction method in an embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to a computer through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.
In this embodiment, a real three-dimensional road reconstruction method is provided, and fig. 2 is a flowchart of a real three-dimensional road reconstruction method according to an embodiment of the present invention, as shown in fig. 2, the flowchart includes the following steps:
step S202, obtaining an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model;
the oblique photography image of the embodiment is oblique photography measurement data, multiple sensors are adopted for multi-view image information and data acquisition, high-precision earth surface three-dimensional data achievement can be rapidly obtained, multi-source surveying and mapping geographic information such as a real-scene three-dimensional model, an ortho-image, a digital earth surface model and point cloud data is covered, the oblique photography image can be obtained based on unmanned aerial vehicle oblique photography measurement, and an air-to-three modeling software (such as ContextCapture) can be adopted for generating the oblique three-dimensional model.
Step S204, extracting point cloud data of a road area from the point cloud model along the road vector line;
step S206, extracting road geometric characteristics of a specific road area from the point cloud data, and extracting texture characteristics corresponding to the road geometric area from the orthographic projection image;
and the ground object point cloud is noise in the point cloud model, and the point cloud data of the road area is finally obtained by filtering the ground object point cloud.
The road geometric characteristics comprise spatial distribution and position characteristics of road shoulder lines, green belts, trees and the like, and the road texture characteristics comprise the color, the shape and the like of a road surface.
Step S208, carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting texture characteristics, carrying out accessory facility modeling and semantic information modeling, and generating a real three-dimensional road;
and S210, performing index organization on the three-dimensional road based on the earth tile pyramid structure and the quadtree subdivision to realize the road three-dimensional visualization of the target area.
Through the steps, an oblique photography image of a target area is obtained, an oblique three-dimensional model is generated based on the oblique photography image, a point cloud model and an orthographic projection image are generated according to the oblique three-dimensional model, point cloud data of a road area are extracted from the point cloud model along a road vector line, road geometric characteristics of a specific road area are extracted from the point cloud data, texture characteristics corresponding to the road geometric area are extracted from the orthographic projection image, the road vector line is processed in a segmentation mode, triangulation is carried out on each segmentation by adopting the road geometric characteristics, a three-dimensional road is reconstructed, texture mapping is carried out on the three-dimensional road by adopting the texture characteristics in sequence, subsidiary facility modeling and semantic information modeling are carried out, a real three-dimensional road is generated, and index organization is carried out on the three-dimensional road based on an earth tile pyramid structure and a quadtree subdivision so as to realize the road three-dimensional visualization of the target area, the method has the advantages that the real-scene three-dimensional road is reconstructed on the point cloud model and the orthophoto map corresponding to the oblique photography image, laser point cloud road modeling and vector road automatic modeling are combined, the technical problems of high complexity and high cost of real-scene three-dimensional road construction in the related technology are solved, the processing speed is high, the modeling cost is low, and the modeling effect is good.
The method comprises the steps of extracting parameter information such as road shoulder lines, green belts, trees and the like through a live-action three-dimensional model and a point cloud model generated by oblique photography image data, realizing large-range urban road three-dimensional reconstruction based on a partition modeling and fusion method, and describing in detail in the embodiment below.
Fig. 3 is a schematic diagram of data preprocessing according to an embodiment of the present invention, which is to acquire an oblique photography model, a dense point cloud, an orthoimage, and a digital surface model based on oblique photography image data to perform data preprocessing for preparing for road three-dimensional reconstruction, where the data preprocessing includes point cloud filtering, geometric feature extraction, and texture feature extraction.
In one embodiment of this embodiment, extracting point cloud data of a road region from the point cloud model along a road vector line includes: performing buffer area cutting according to the preset buffer area radius based on the road center line vector in the preset two-dimensional data, and filtering the point cloud model to obtain a first road area point cloud; acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and acquiring the road pavement point cloud height of the second road area point cloud, and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road pavement point cloud height.
In order to extract the road characteristics, it is necessary to classify the ground features around the road and filter out irrelevant ground features such as vehicles. The filtering steps of the ground object point cloud are as follows:
and based on the vector data of the road center line, performing buffer area cutting with a certain buffer area radius, and filtering out the road area point cloud from the urban area point cloud data.
And acquiring the height H of the highest point of the road area point cloud, gradually reducing the filtering height by a fixed step length d, filtering the road area point cloud data, and correcting the road area. Let the road surface height be h0Then the filter height can be calculated as:
Figure 456305DEST_PATH_IMAGE010
；
wherein n is a positive integer, and
Figure 672523DEST_PATH_IMAGE011
and n is the number of decrements.
And acquiring the point cloud height of the road surface, filtering the point cloud data based on the height, and removing the point clouds of ground objects such as vehicles and the like on the ground.
In an example of extracting geometric features of the present embodiment, the road geometric features include a road shoulder line, and extracting road geometric features of a specific area of a road from the point cloud data includes:
s11, partitioning the point cloud data, taking a rectangular block on a road center line in the road vector line as a seed point, judging whether an isolation zone exists in the middle of the road according to the seed point, and marking an isolation zone area;
s12, respectively carrying out growth search to two sides perpendicular to the road center line along the seed points until the average height of the seed points is higher than the road surface height, and obtaining the road shoulder blocks where the road shoulder lines are located;
s13, acquiring point cloud data in the road shoulder blocks and the adjacent blocks, reducing the size of the blocks by half, continuing to perform growth search on the road shoulder blocks, and extracting to obtain road shoulder line point cloud;
and S14, generating a road shoulder line according to the road shoulder line point cloud.
In one implementation of the above embodiment, generating a road shoulder line from the road shoulder line point cloud comprises:
s14-1, aiming at the point cloud of the straight road section, aiming at the characteristic point p in the point cloud of the shoulder line based on the following formulaiPerforming shoulder line smoothing to generate a straight shoulder line:
Figure 882925DEST_PATH_IMAGE012
；
wherein, Pi(t) is, t is a sampling parameter, pi+1Is a characteristic point piAdjacent point of (a), pi+2Is a characteristic point pi+1Adjacent points of (a);
s14-2, aiming at the point cloud of the turning road section, the local coordinate point P (x) of the ith sampling point P in the point cloud of the road shoulder line is converted based on the following conversion algorithmi，yi) Converting into a global Cartesian coordinate point P', and generating a curved road shoulder line:
Figure 376354DEST_PATH_IMAGE002
；
wherein the content of the first and second substances,
Figure 787744DEST_PATH_IMAGE013
is a global coordinate point, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 773148DEST_PATH_IMAGE004
and
Figure 459345DEST_PATH_IMAGE005
is a global coordinate direction vector parameter, scaleX and scaleY are scaling parameters.
In this embodiment, the road geometry includes the spatially distributed locations of road shoulder lines, green belts, trees, and the like. The position, the spatial range and the elevation of the green belt, the trees and the like can be extracted when point cloud filtering is carried out. The extraction of the road surface is realized by a region growing algorithm and a sideline repairing algorithm based on unmanned aerial vehicle point cloud data and road center line vector data, fig. 4 is a schematic diagram of the extraction of geometric features of the road surface region in the embodiment of the invention, the road center line is arranged in the middle of the road surface, the road shoulder lines are arranged on two sides of the road surface, and the extraction process of the geometric features comprises the following steps:
(1) and partitioning the road area point cloud data to obtain a rectangular block on the road center line as a seed point. Let the seed point height be diCalculating the average height of the seed points with the point cloud number of num
Figure 994231DEST_PATH_IMAGE014
And judging whether an isolation zone exists in the middle of the road, and marking a relevant area if the isolation zone exists.
(2) And respectively carrying out growth search to two sides perpendicular to the road center line along the seed points until the average height of the seed points is obviously higher than the road surface height, and obtaining the blocks where the road shoulder lines are located.
(3) And acquiring the point cloud data in the blocks determined in the previous step and the adjacent blocks thereof, setting the size of the blocks to be half of that of the blocks in the previous step, and performing block region growing algorithm search again to determine the road shoulder line point cloud. Setting a road surface height threshold value as D, an angle threshold value of a normal line and a direction vertical to the road surface as a, and a road shoulder point cloud height as DiAngle of thetaiThen the road shoulder line point cloud needs to satisfy the following conditions:
Figure 451888DEST_PATH_IMAGE015
；
(4) and filtering, smoothing and connecting the extracted road shoulder line point cloud to obtain the road shoulder line.
For a straight road section, let three points on the road shoulder line be (adjacent characteristic points) pi、pi+1、pi+2And t is a sampling parameter, and the shoulder line smoothing treatment is realized based on a secondary B spline curve:
Figure 845961DEST_PATH_IMAGE016
；
for a turning road section, sampling and completing based on a corner arc curve, and if the arc radius is R and the sampling angle is theta, then the local coordinate P (x) of the ith sampling pointi，yi) Can be calculated as:
Figure 398165DEST_PATH_IMAGE017
；
based on a seven-parameter conversion method, local coordinates of corner sampling points are converted into global Cartesian coordinates, the global coordinates are set to be P',
Figure 725241DEST_PATH_IMAGE003
is a global coordinate, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 478433DEST_PATH_IMAGE004
and
Figure 969589DEST_PATH_IMAGE005
is the global coordinate direction vector parameter, scaleX and scaleY are the scaling parameters, then the global coordinate P' can be calculated as:
Figure 263167DEST_PATH_IMAGE018
。
in another aspect of this embodiment, the road texture features of the road area are extracted from the point cloud of the road pavement, and the ortho-image texture data of the corresponding area is obtained from the real three-dimensional model according to the road area determined in the geometric feature extraction process. The texture features comprise road surface textures, green belt textures and the like, and preparation is made for a road three-dimensional reconstruction map behind by acquiring texture data.
In an implementation manner of this embodiment, triangulating each segment by using the geometric features of the road to reconstruct a three-dimensional road includes:
for a single straight road or curved road in the road vector line, selecting a first vertex of each segment from road shoulder line characteristic points of road geometric characteristics as an origin to construct a local Cartesian coordinate system;
and loading the segmented road model onto the three-dimensional map by adopting a scaling matrix, a rotation matrix and a translation matrix.
The road shoulder line of the road surface area is obtained based on geometric feature extraction, the road area is segmented for facilitating model organization, and each segment is triangulated. The road geometric reconstruction comprises two types of single road network construction and intersection network construction:
for a single straight road or a single curved road, a Dironey Delaunay triangulation network is constructed based on information such as a road shoulder line characteristic point, a road shoulder height, a pedestrian road width and the like, and FIG. 5 is a schematic diagram of the single road partition geometric modeling in the embodiment of the invention. In order to reduce the modeling data amount, for a certain section of road, the first vertex of the road is selected as an origin to construct a local cartesian coordinate system (O-X ' Y ' Z '), and other vertex coordinates can be calculated as:
Figure 913591DEST_PATH_IMAGE019
；
and constructing a road triangulation network based on a local Cartesian coordinate system, and then judging the azimuth deflection angle of the road. Let two points in its deflection orientation be (x)1，y1，z1) And (x)2，y2，z2) Then the rotation angle is calculated as:
Figure 696739DEST_PATH_IMAGE020
；
the rotation matrix to load the road model onto the three-dimensional map may be calculated as:
Figure 144032DEST_PATH_IMAGE021
；
scaling matrix M in combination with a road modelscaling(according to the original ratio of 1: 1) and a rotation matrix MrotationTranslation matrix MtranslationMatrix transformation is carried out (obtained by conversion according to actual longitude and latitude), and the geometric reconstruction of a large-range road scene can be realized:
Figure 178984DEST_PATH_IMAGE022
；
in some examples of this embodiment, triangulating each segment with road geometry comprises: judging the type of an intersection road in the road vector line based on the road central point vector data; according to the type of the intersection road, a road triangular net is constructed by adopting the feature points of the road shoulder lines, the road shoulder height and the width of the walking way, wherein the geometrical features of the road comprise the road shoulder lines and the width of the walking way, and the road shoulder height is obtained from a digital earth surface model of oblique photography data.
For an intersection road, judging intersection types such as a three-way intersection and a cross intersection based on road center point vector data, and then constructing a Dirofeni Delaunay triangulation network based on information such as a road shoulder line, a road shoulder height, a pedestrian path width and the like, wherein FIG. 6 is a schematic diagram of the road intersection partition geometric modeling of the embodiment of the invention, AB and CD are central lines of two roads of the cross intersection, O is an intersection, MN and PQ are segments, EF is a road shoulder line of the road intersection, and two points on EF form the triangulation network to the O point. The intersection geometric model realizes the connection of different straight roads or curved roads to form a complete urban road model.
In some examples of this embodiment, sequentially performing texture mapping on a three-dimensional road by using texture features, and performing affiliated facility modeling and semantic information modeling, includes:
S21，determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 339707DEST_PATH_IMAGE023
；
in one example, the road texture map mainly implements road area plane simulation, such as a road surface, a green belt, and the like, and fig. 7 is a schematic diagram of a road partition texture map according to an embodiment of the present invention, including areas 1 to 5, which are distributed in a long and narrow shape, so that the representation is performed based on a repeated texture map, where the length of a current segmented road is set to be L, the real road length corresponding to a texture is set to be D, and a sampling point C is set to be Ci+1And CiD is the distance of Ci+1The UV texture coordinates of (a) may be calculated as:
Figure 169123DEST_PATH_IMAGE024
；
s22, determining the number of the instantiated auxiliary facilities, calculating the area of the target area of the corresponding example according to the number of the examples, and searching the next Level matched with the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 634870DEST_PATH_IMAGE007
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
in one example, the distribution of the satellite models can be divided into three categories: point-like distribution, linear distribution and planar distribution. Different distribution characteristics are based on an instantiation modeling method, and for point-shaped distribution facilities such as signs, instantiation is carried out according to the placement position, scaling and translation coefficients of each accessory. And for a linear distribution facility such as a row tree, extracting a path for setting the distribution of the attachments and a distribution interval, and performing instantiation modeling. For planar distribution facilities such as shrubs, a spatial range for setting accessory distribution is extracted, and a certain distribution coefficient is set for instantiation modeling.
Since the types of the affiliated facilities are various and the number of the affiliated facilities is large, in order to express the road and the affiliated facility distribution finely, the block organization based on the quadtree is required. A reasonable pyramid level is particularly important because too high a pyramid level may result in too large a number of partitions, while too low a pyramid level may result in too high a single block of data. Firstly, the number of instances of a specific instantiation technology is determined according to experience, then a general area containing the corresponding number of instances is determined according to the number of the instances, and further the side length of a square (or parameters such as the radius of a circle) is determined to match the nearest next layer, and the Level can be calculated as follows:
Figure 473513DEST_PATH_IMAGE025
wherein S is the size of the pyramid top-level tile, T is the size of the target area, and Floor is a Floor rounding function.
And S23, acquiring the semantic attribute information and overloading the semantic attribute information to the model component of the three-dimensional road.
After a complete three-dimensional model is obtained through geometric reconstruction, block organization and indexing can be continuously performed on the three-dimensional model, road accessory facility distribution is performed on the basis of a pyramid Level quadtree index, fig. 8 is a block organization schematic diagram of a large-range road model in the embodiment of the present invention, a pyramid includes n +1 levels (from 0 to n), k levels of quadtrees include 1,2,3 and 4, and a longitude and latitude range of a road area is set as (MinLon, MinLat, MaxLon and MaxLat), so that a row and column range of a corresponding quadtree is:
Figure 629688DEST_PATH_IMAGE026
；
the road three-dimensional construction is realized based on geometric modeling and texture mapping, but detailed semantic information needs to be added for each part on the road to realize single query. Semantic attribute information is obtained through live-action three-dimensional extraction and a road vector center line, and is overloaded onto a correct road three-dimensional model component, so that semantic information modeling is realized.
The embodiment provides a real three-dimensional road reconstruction scheme, the advantages of laser point cloud road modeling and vector road automatic modeling are combined, parameter information such as road shoulder lines, green belts and trees is extracted through a real three-dimensional model and a point cloud model generated by oblique photography image data, road three-dimensional reconstruction is realized based on a partition modeling and fusion method, data parameter extraction, geometric modeling, texture mapping, semantic modeling and block organization are carried out through oblique photography images, real three-dimensional road reconstruction is realized, and fig. 9 is an effect diagram after the real road three-dimensional reconstruction is carried out according to the embodiment of the invention. Compared with the vehicle-mounted laser point cloud road modeling and vector road automatic modeling method in the related technology, the method has the following advantages: the processing speed is high. The unmanned aerial vehicle oblique photography three-dimensional model is automatically processed to obtain modeling information, road three-dimensional reconstruction is achieved based on a partition modeling and fusion method, and automatic rapid processing can be achieved based on a program. The modeling cost is low. The method does not need laser point cloud measurement hardware and software equipment, the unmanned aerial vehicle oblique photography cost is low, and large-range urban three-dimensional road reconstruction can be realized by extracting oblique photography model information and carrying out program automatic modeling. The modeling effect is good. Because the road parameter information is obtained by extracting the inclined real scene data, the road distribution is the same as the actual road scene, and the single modeling and semantic information hanging of different areas of the road, such as the road surface, green belts, isolation fences and the like, are realized, thereby being convenient for query and analysis.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
Example 2
In this embodiment, a real-scene three-dimensional road reconstruction apparatus is further provided, which is used to implement the foregoing embodiments and preferred embodiments, and the description of which has been already given is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 10 is a block diagram of a real three-dimensional road reconstruction apparatus according to an embodiment of the present invention, as shown in fig. 10, the apparatus includes: an acquisition module 100, a generation module 102, a filtering module 104, an extraction module 106, and a reconstruction module 108, wherein,
an obtaining module 100 for obtaining oblique photography data of a road area and an original three-dimensional model corresponding to the oblique photography data;
a generating module 102 for generating a dense point cloud from the oblique photography data;
the filtering module 104 is used for filtering the ground object point cloud in the dense point cloud to obtain a road pavement point cloud;
an extraction module 106, configured to extract road geometric features and road texture features of the road region from the road surface point cloud;
and the reconstruction module 108 is used for performing segmentation processing on the road area in the original three-dimensional model, performing triangulation network on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially performing texture mapping on the three-dimensional road by adopting the road texture characteristics, performing accessory facility modeling and semantic information modeling, and generating a live-action three-dimensional road.
Optionally, the filtering module includes: the first filtering unit is used for performing buffer area cutting according to the preset buffer area radius based on the road centerline vector data in the original three-dimensional model, and filtering the dense point cloud to obtain a first road area point cloud; the second filtering unit is used for acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and the third filtering unit is used for acquiring the road pavement point cloud height of the second road area point cloud and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road pavement point cloud height.
Optionally, the geometric feature of the road includes a spatial distribution position of a road shoulder line, and the extraction module includes: the processing unit is used for partitioning the road surface point cloud, taking a rectangular block on a road center line in the original three-dimensional model as a seed point, judging whether an isolation zone exists in the middle of the road according to the seed point, and marking an isolation zone area in the original three-dimensional model if the isolation zone exists; the searching unit is used for respectively carrying out growth searching on two sides perpendicular to the road center line along the seed points until the average height of the seed points is higher than the road surface height to obtain the road shoulder blocks where the road shoulder lines are located; the extraction unit is used for acquiring point cloud data in the road shoulder blocks and the adjacent blocks thereof, and continuing to perform growth search on the road shoulder blocks after the sizes of the blocks are halved to extract road shoulder line point clouds; and the generating unit is used for generating a road shoulder line in the original three-dimensional model according to the road shoulder line point cloud.
Optionally, the generating unit includes: a processing subunit, configured to, for a straight road segment point cloud, pair the feature points p in the road shoulder line point cloud based on the following formulaiPerforming shoulder line smoothing to generate a straight shoulder line:
Figure 754639DEST_PATH_IMAGE016
；
wherein, Pi(t) is, t is a sampling parameter, pi+1Is a characteristic point piAdjacent point of (a), pi+2Is a characteristic point pi+1Adjacent points of (a);
a conversion subunit, configured to, for a turning road segment point cloud, convert a local coordinate point P (xi, yi) of an ith sampling point P in the road shoulder line point cloud into a global cartesian coordinate point P' based on the following conversion algorithm, and generate a curved road shoulder line:
Figure 629054DEST_PATH_IMAGE027
；
wherein the content of the first and second substances,
Figure 271388DEST_PATH_IMAGE003
is a global coordinate, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 360698DEST_PATH_IMAGE004
and
Figure 594233DEST_PATH_IMAGE005
is a global coordinate direction vector parameter, scaleX and scaleY are scaling parameters.
Optionally, the reconstruction module includes: the first construction unit is used for selecting the first vertex of each segment from the road shoulder line characteristic points of the road geometric characteristics as an origin to construct a local Cartesian coordinate system for a single straight road or curved road in a road vector line; and the loading unit is used for loading the segmented road model onto the three-dimensional map by adopting the scaling matrix, the rotation matrix and the translation matrix.
Optionally, the reconstruction module includes: the judging unit is used for judging the type of the intersection on the basis of the road central point vector data for the intersection road in the road vector line; and the second construction unit is used for constructing a road triangulation network by adopting the feature points of the road shoulder lines, the road shoulder height and the width of the pedestrian path according to the type of the intersection road, wherein the geometrical features of the road comprise the road shoulder lines and the width of the pedestrian path, and the road shoulder height is obtained from the digital earth surface model of the oblique photography data.
Optionally, the reconstruction module includes: a calculating unit for determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 893627DEST_PATH_IMAGE028
；
the searching unit is used for determining the number of the instantiated auxiliary facilities, calculating the area of the target area of the corresponding example according to the number of the examples, and searching the next Level matched with the examples of the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 870811DEST_PATH_IMAGE007
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
and the overload unit is used for acquiring semantic attribute information from the original three-dimensional model and overloading the semantic attribute information to the model component of the three-dimensional road.
It should be noted that, the above modules may be implemented by software or hardware, and for the latter, the following may be implemented, but not limited to: the modules are all positioned in the same processor; alternatively, the modules are respectively located in different processors in any combination.
Example 3
Embodiments of the present invention also provide a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the above method embodiments when executed.
Alternatively, in the present embodiment, the storage medium may be configured to store a computer program for executing the steps of:
s1, acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model;
s2, extracting point cloud data of a road area from the point cloud model along a road vector line;
s3, extracting road geometric features of a road specific area from the point cloud data, and extracting texture features corresponding to the road geometric area from the orthophoto map;
s4, carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture characteristics, carrying out accessory facility modeling and semantic information modeling, and generating a live-action three-dimensional road;
s5, based on the earth tile pyramid structure and the quadtree subdivision, indexing and organizing the three-dimensional road so as to realize the road three-dimensional visualization of the target area.
Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing computer programs, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Embodiments of the present invention also provide an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.
Optionally, the electronic device may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
s1, acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model;
s2, extracting point cloud data of a road area from the point cloud model along a road vector line;
s3, extracting road geometric features of a road specific area from the point cloud data, and extracting texture features corresponding to the road geometric area from the orthophoto map;
s4, carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture characteristics, carrying out accessory facility modeling and semantic information modeling, and generating a live-action three-dimensional road;
s5, based on the earth tile pyramid structure and the quadtree subdivision, indexing and organizing the three-dimensional road so as to realize the road three-dimensional visualization of the target area.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation manners, and this embodiment is not described herein again.
The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present application, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present application and it should be noted that those skilled in the art can make several improvements and modifications without departing from the principle of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims (8)

1. A real three-dimensional road reconstruction method is characterized by comprising the following steps:
acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model;
extracting point cloud data of a road area from the point cloud model along a road vector line;
extracting road geometric features of a road specific area from the point cloud data, and extracting texture features corresponding to the road geometric area from the orthophoto map;
carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture characteristics, and carrying out accessory facility modeling and semantic information modeling to generate a real three-dimensional road;
based on the earth tile pyramid structure and the quadtree subdivision, performing index organization on the three-dimensional road to realize the road three-dimensional visualization of the target area;
the texture features are adopted to sequentially carry out texture mapping on the three-dimensional road, and accessory facility modeling and semantic information modeling are carried out, and the method comprises the following steps:
determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 808773DEST_PATH_IMAGE001
；
determining the number of the instantiated auxiliary facilities, calculating the area of a target area of a corresponding example according to the number of the examples, and searching the next Level matched with the examples of the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 725914DEST_PATH_IMAGE002
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
obtaining semantic attribute information, and overloading the semantic attribute information to a model component of the three-dimensional road;
extracting point cloud data of a road area from the point cloud model along a road vector line comprises the following steps: performing buffer area cutting according to the preset buffer area radius based on the road center line vector in the preset two-dimensional data, and filtering the point cloud model to obtain a first road area point cloud; acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and acquiring the road surface point cloud height of the second road area point cloud, and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road surface point cloud height.
2. The method of claim 1, wherein the road geometry comprises a road shoulder line, and wherein extracting road geometry for a specific area of a road from the point cloud data comprises:
partitioning the point cloud data, taking a rectangular block on a road central line in the road vector line as a seed point, judging whether an isolation zone exists in the middle of the road according to the seed point, and marking an isolation zone area;
respectively carrying out growth search to two sides perpendicular to the road center line along the seed points until the average height of the seed points is higher than the road surface height to obtain road shoulder blocks where the road shoulder lines are located;
acquiring point cloud data in the road shoulder blocks and the adjacent blocks thereof, reducing the size of the blocks by half, continuing to perform growth search on the road shoulder blocks, and extracting to obtain road shoulder line point cloud;
and generating a road shoulder line according to the road shoulder line point cloud.
3. The method of claim 2, wherein generating a road shoulder line from the road shoulder line point cloud comprises:
aiming at the straight road section point cloud, the characteristic point p in the road shoulder line point cloud is processed based on the following formulaiPerforming shoulder line smoothing to generate a straight shoulder line:
Figure 567968DEST_PATH_IMAGE003
；
where t is a sampling parameter, pi+1Is a characteristic point piAdjacent point of (a), pi+2Is a characteristic point pi+1Adjacent points of (a);
aiming at a turning road section point cloud, a local coordinate point P (x) of the ith sampling point P in the road shoulder line point cloud is converted based on the following conversion algorithmi，yi) Converting into a global Cartesian coordinate point P', and generating a curved road shoulder line:
Figure 963177DEST_PATH_IMAGE004
；
wherein the content of the first and second substances,
Figure 777549DEST_PATH_IMAGE005
is a global coordinate, x, corresponding to the origin of the local coordinate systemiAnd yiIs the coordinate of P in the local coordinate system,
Figure 413061DEST_PATH_IMAGE006
and
Figure 680094DEST_PATH_IMAGE007
is a global coordinate direction vector parameter, scaleX and scaleY are scaling parameters.
4. The method of claim 1, wherein triangulating each segment using the road geometry to reconstruct a three-dimensional road comprises:
for a single straight road or curved road in the road vector line, selecting a first vertex of each segment from the road shoulder line characteristic points of the road geometric characteristics as an origin to construct a local Cartesian coordinate system;
and loading the road model of each segment onto the three-dimensional map by adopting a scaling matrix, a rotation matrix and a translation matrix.
5. The method of claim 1, wherein triangulating each segment using the road geometry comprises:
judging the type of an intersection road in the road vector line based on the road central point vector data;
and constructing a road triangulation network by adopting shoulder line characteristic points, shoulder heights and pedestrian path widths according to the intersection types of the intersection roads, wherein the road geometric characteristics comprise the shoulder lines and the pedestrian path widths, and the shoulder heights are obtained from the digital earth surface model of the oblique photography data.
6. A live-action three-dimensional road reconstruction device is characterized by comprising:
the system comprises an acquisition module, a storage module and a display module, wherein the acquisition module is used for acquiring an oblique photography image of a target area, generating an oblique three-dimensional model based on the oblique photography image, and generating a point cloud model and an orthophoto map according to the oblique three-dimensional model;
the filtering module is used for extracting point cloud data of a road area from the point cloud model along a road vector line;
the extraction module is used for extracting the road geometric characteristics of a road specific area from the point cloud data and extracting the texture characteristics corresponding to the road geometric area from the orthophoto map;
the reconstruction module is used for carrying out segmentation processing on the road vector lines, carrying out triangulation network construction on each segment by adopting the road geometric characteristics, reconstructing a three-dimensional road, sequentially carrying out texture mapping on the three-dimensional road by adopting the texture characteristics, carrying out accessory facility modeling and semantic information modeling, and generating a live-action three-dimensional road;
the organization module is used for carrying out index organization on the three-dimensional roads based on the earth tile pyramid structure and the quadtree subdivision so as to realize the road three-dimensional visualization of the target area;
wherein the reconstruction module comprises: a calculating unit for determining the length L of the current subsection road of the three-dimensional road, the real road length D corresponding to the texture, and the adjacent mapping sampling point Ci+1And CiD, calculating the map sampling point C by using the following formulai+1UV texture coordinates of (a):
Figure 878994DEST_PATH_IMAGE008
；
the searching unit is used for determining the number of the instantiated auxiliary facilities, calculating the area of the target area of the corresponding example according to the number of the examples, and searching the next Level matched with the examples of the auxiliary facilities according to the graphic parameters of the quad-tree block organization by adopting the following formula:
Figure 547873DEST_PATH_IMAGE002
；
wherein S is the size of the pyramid top-layer tile, T is the area of the target region, and Floor is a down-rounding function;
the overload unit is used for acquiring semantic attribute information and overloading the semantic attribute information to the model component of the three-dimensional road;
wherein the filter module comprises: the first filtering unit is used for performing buffer area cutting according to the preset buffer area radius based on the road center line vector in the preset two-dimensional data, and filtering the point cloud model to obtain a first road area point cloud; the second filtering unit is used for acquiring a highest point H of the first road area point cloud, gradually reducing the filtering height by taking a fixed step length d as a unit from the highest point H, and filtering the first road area point cloud to obtain a second road area point cloud; and the third filtering unit is used for acquiring the road pavement point cloud height of the second road area point cloud and filtering the ground object point cloud of the road ground in the second road area point cloud based on the road pavement point cloud height.
7. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program is operative to perform the method steps of any of the preceding claims 1 to 5.
8. An electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus; wherein:
a memory for storing a computer program;
a processor for performing the method steps of any of claims 1 to 5 by executing a program stored on a memory.
CN202111242316.8A 2021-10-25 2021-10-25 Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment Active CN113706698B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111242316.8A CN113706698B (en) 2021-10-25 2021-10-25 Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111242316.8A CN113706698B (en) 2021-10-25 2021-10-25 Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment

Publications (2)

Publication Number Publication Date
CN113706698A CN113706698A (en) 2021-11-26
CN113706698B true CN113706698B (en) 2022-01-25

Family

ID=78646801

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111242316.8A Active CN113706698B (en) 2021-10-25 2021-10-25 Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN113706698B (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114494849B (en) * 2021-12-21 2024-04-09 重庆特斯联智慧科技股份有限公司 Road surface state identification method and system for wheeled robot
CN114818091B (en) * 2022-06-23 2022-09-27 武汉幻城经纬科技有限公司 BIM (building information modeling) refined modeling method and device for pipeline joint, electronic equipment and storage medium
CN114863195B (en) * 2022-07-07 2022-09-13 山东矩阵软件工程股份有限公司 Point cloud data processing method and system, storage medium and electronic equipment
CN115205472B (en) * 2022-09-16 2022-12-02 成都国星宇航科技股份有限公司 Grouping method, device and equipment for live-action reconstruction pictures and storage medium
CN115908706B (en) * 2022-11-15 2023-08-08 中国铁路设计集团有限公司 High-speed railway completion acceptance method with fusion of live three-dimensional model and image
CN115937456B (en) * 2023-02-15 2023-05-05 天津市测绘院有限公司 Top layer reconstruction method and reconstruction system for live-action three-dimensional model
CN116310185B (en) * 2023-05-10 2023-09-05 江西丹巴赫机器人股份有限公司 Three-dimensional reconstruction method for farmland field and intelligent agricultural robot thereof
CN116342783B (en) * 2023-05-25 2023-08-08 吉奥时空信息技术股份有限公司 Live-action three-dimensional model data rendering optimization method and system
CN116630598B (en) * 2023-07-19 2023-09-29 齐鲁空天信息研究院 Visual positioning method and device under large scene, electronic equipment and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107356230A (en) * 2017-07-12 2017-11-17 深圳市武测空间信息有限公司 A kind of digital mapping method and system based on outdoor scene threedimensional model
CN110260876A (en) * 2019-06-27 2019-09-20 厦门建研建筑产业研究有限公司 A kind of road model generation method and system based on oblique photograph and GIS technology
CN110889900A (en) * 2019-11-22 2020-03-17 北航（四川）西部国际创新港科技有限公司 Low-airspace-oriented three-dimensional modeling and visualization method
CN111428306A (en) * 2020-04-17 2020-07-17 中交第二航务工程局有限公司 Road fine design method
CN112461205A (en) * 2020-11-12 2021-03-09 中国铁路设计集团有限公司 Method for manufacturing cross section of existing railway line based on unmanned aerial vehicle oblique photogrammetry

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10109106B2 (en) * 2016-09-14 2018-10-23 Toyota Jidosha Kabushiki Kaisha Scalable curve visualization for conformance testing in vehicle simulation
US10768304B2 (en) * 2017-12-13 2020-09-08 Luminar Technologies, Inc. Processing point clouds of vehicle sensors having variable scan line distributions using interpolation functions
CN109934914B (en) * 2019-03-28 2023-05-16 东南大学 Embedded city design scene simulation method and system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107356230A (en) * 2017-07-12 2017-11-17 深圳市武测空间信息有限公司 A kind of digital mapping method and system based on outdoor scene threedimensional model
CN110260876A (en) * 2019-06-27 2019-09-20 厦门建研建筑产业研究有限公司 A kind of road model generation method and system based on oblique photograph and GIS technology
CN110889900A (en) * 2019-11-22 2020-03-17 北航（四川）西部国际创新港科技有限公司 Low-airspace-oriented three-dimensional modeling and visualization method
CN111428306A (en) * 2020-04-17 2020-07-17 中交第二航务工程局有限公司 Road fine design method
CN112461205A (en) * 2020-11-12 2021-03-09 中国铁路设计集团有限公司 Method for manufacturing cross section of existing railway line based on unmanned aerial vehicle oblique photogrammetry

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
Template-Based 3D Road Modeling for Generating Large-Scale Virtual Road Network Environment;Xuequan Zhang 等;《International Journal of Geo-Information》;20190822;第1-17页 *
Traffic Simulation Method for Dynamic Road Network Model Construction Based on Oblique Aerial Photogrammetry Technology;Wang H 等;《Design Engineering》;20201231;第105-116页 *
一种基于无人机倾斜摄影的三维路网提取方法;李磊;《中国公路学报》;20191130;第219-226页 *
基于LIDAR的高速公路测设技术应用研究;富志鹏;《中国优秀博硕士学位论文全文数据库(博士)工程科技Ⅱ辑》;20130515;C034-41 *
基于虚拟地球的流场三维动态可视化方法;吴红燕 等;《气象科技》;20190630;第17-19页 *
李磊.一种基于无人机倾斜摄影的三维路网提取方法.《中国公路学报》.2019, *
车载移动测量地物分类方法与重建技术研究;胡晓斌;《中国优秀博硕士学位论文全文数据库(博士)基础科学辑》;20200115;A008-38 *

Also Published As

Publication number Publication date
CN113706698A (en) 2021-11-26

Similar Documents

Publication Publication Date Title
CN113706698B (en) Live-action three-dimensional road reconstruction method and device, storage medium and electronic equipment
CN106780736B (en) Map data processing method and device and three-dimensional map generation method and device
CN109242862B (en) Real-time digital surface model generation method
CN111008422B (en) Building live-action map making method and system
CN116310192A (en) Urban building three-dimensional model monomer reconstruction method based on point cloud
CN109934911B (en) OpenGL-based three-dimensional modeling method for high-precision oblique photography of mobile terminal
CN104809759A (en) Large-area unstructured three-dimensional scene modeling method based on small unmanned helicopter
CN105279793A (en) Modeling method and system based on DEM real three-dimensional map and greenway
CN109900338B (en) Method and device for measuring volume of pavement pit
CN105354882A (en) Method for constructing big data architecture based three-dimensional panoramic display platform for large-spatial-range electricity transmission
US7778808B2 (en) Geospatial modeling system providing data thinning of geospatial data points and related methods
CN115471634B (en) Modeling method and device for urban green plant twins
CN110705381A (en) Remote sensing image road extraction method and device
CN112241661A (en) Urban ground feature fine classification method combining airborne LiDAR point cloud data and aerial image
CN112925865A (en) 3D WebGIS prediction analysis method and system for mining area ground surface movement deformation
CN109872390B (en) Organization method of massive three-dimensional models
CN115752432A (en) Method and system for automatically extracting dotted lane lines in road traffic map acquired by unmanned aerial vehicle
CN109612445A (en) High-precision landform method for building up under a kind of WebGIS platform based on unmanned plane
CN113378844A (en) Engineering garbage stacking segmentation method, device, equipment and computer storage medium
CN110472327B (en) Urban multi-layer temperature and humidity sensing space-time system design method
CN116468869A (en) Live-action three-dimensional modeling method, equipment and medium based on remote sensing satellite image
CN111895907A (en) Electricity tower point cloud extraction method, system and equipment
CN113704945B (en) Pipe network road correlation analysis method and device, storage medium and electronic equipment
CN115205484B (en) Three-dimensional space display method, device, equipment and medium for historical culture block
KR101079531B1 (en) A system for generating road layer using point cloud data

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant