CN112634340A - Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data - Google Patents
Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data Download PDFInfo
- Publication number
- CN112634340A CN112634340A CN202011552398.1A CN202011552398A CN112634340A CN 112634340 A CN112634340 A CN 112634340A CN 202011552398 A CN202011552398 A CN 202011552398A CN 112634340 A CN112634340 A CN 112634340A
- Authority
- CN
- China
- Prior art keywords
- target
- point cloud
- cloud data
- determining
- existing building
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000003066 decision tree Methods 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 14
- 239000000178 monomer Substances 0.000 claims description 35
- 238000013461 design Methods 0.000 claims description 30
- 239000013598 vector Substances 0.000 claims description 30
- 238000003860 storage Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 2
- 238000009435 building construction Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 7
- 238000005259 measurement Methods 0.000 description 5
- 238000011960 computer-aided design Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000013499 data model Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 1
- 241000764238 Isis Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 229920001690 polydopamine Polymers 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a method, a device, equipment and a medium for determining a BIM (building information modeling) model based on point cloud data. Wherein, the method comprises the following steps: determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data; classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data; and determining the BIM model of the existing building according to the element label of the target characteristic element. According to the embodiment of the invention, the BIM model of the existing building can be determined by acquiring the point cloud data of the existing building, and the three-dimensional structure of the existing building can be accurately determined by referring to the model, so that the building construction efficiency is effectively improved.
Description
Technical Field
The embodiment of the invention relates to the field of BIM technology application, in particular to a method, a device, equipment and a medium for determining a BIM model based on point cloud data.
Background
The existing building is a built and formed entity building, and needs to be maintained and modified after the existing building is used for a period of time; in the field of building construction, the maintenance and modification of an existing building are mainly performed by a constructor through an initial design drawing of the existing building to analyze and study so as to identify the content characteristic structure of the existing building, and thus the maintenance and modification of the building are realized.
The defects of the scheme are as follows: the construction process is time-consuming and labor-consuming, and constructors need to determine content structures in the existing buildings according to design drawings and mainly depend on human experiences, so that the construction efficiency can be greatly reduced.
Disclosure of Invention
The embodiment of the application provides a method, a device, equipment and a medium for determining a BIM model based on point cloud data, the BIM model of an existing building can be determined by collecting the point cloud data of the existing building, so that the three-dimensional structure of the existing building can be accurately determined through the model, and the building construction efficiency is effectively improved.
In a first aspect, an embodiment of the present invention provides a method for determining a BIM model based on point cloud data, including:
determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data;
classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data;
and determining the BIM model of the existing building according to the element label of the target characteristic element.
Optionally, determining target point cloud data of a target member in an existing building includes:
acquiring first point cloud data of a target component in an existing building in a three-dimensional laser scanning mode, and acquiring second point cloud data of the target component in the existing building in an unmanned aerial vehicle photographing mode; the unmanned aerial vehicle shooting mode is executed through an unmanned aerial vehicle oblique shooting system;
and fusing the first point cloud data and the second point cloud data to obtain target point cloud data of a target component in the existing building.
Optionally, determining surface feature primitives of the target component according to the target point cloud data includes:
and extracting surface feature primitives of the target component under each type from the target point cloud data according to the type of the target component.
Optionally, classifying the surface feature primitives by using a pre-trained multi-class decision tree classifier to obtain target feature primitives, including:
establishing a feature vector of the target component;
and classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier based on the feature vectors of the target member to obtain target feature elements under different feature vectors.
Optionally, determining the BIM model of the existing building according to the primitive tag of the target feature primitive includes:
importing the element labels of the target feature elements into a pre-trained monomer model library to obtain a monomer model of the target member in the existing building;
and integrating the monomer models of the target components to obtain the BIM model of the existing building.
Optionally, integrating the monomer models of the target members to obtain the BIM model of the existing building, including:
determining a pair of interconnected members according to the connection relation of the target members;
and integrating the two associated monomer models by the member to obtain the BIM model of the existing building.
Optionally, the method further includes:
determining a design model of the existing building according to the design information of the existing building;
calling an error detection plug-in to detect coordinate deviation values of target components in the BIM model and the design model;
and if the coordinate deviation value is detected to be larger than a coordinate deviation threshold value, adjusting the position coordinates of the target component in the BIM based on the position coordinates of the target component in the design model to obtain an adjusted BIM.
In a second aspect, an embodiment of the present invention provides an apparatus for determining a BIM model based on point cloud data, including:
the system comprises a first determination module, a second determination module and a third determination module, wherein the first determination module is used for determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data;
the second determination module is used for classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data;
and the third determining module is used for determining the BIM model of the existing building according to the element label of the target characteristic element.
Optionally, the first determining module is specifically configured to:
acquiring first point cloud data of a target component in an existing building in a three-dimensional laser scanning mode, and acquiring second point cloud data of the target component in the existing building in an unmanned aerial vehicle photographing mode; the unmanned aerial vehicle shooting mode is executed through an unmanned aerial vehicle oblique shooting system;
and fusing the first point cloud data and the second point cloud data to obtain target point cloud data of a target component in the existing building.
Optionally, the first determining module is further specifically configured to:
and extracting surface feature primitives of the target component under each type from the target point cloud data according to the type of the target component.
Optionally, the second determining module is specifically configured to:
establishing a feature vector of the target component;
and classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier based on the feature vectors of the target member to obtain target feature elements under different feature vectors.
Optionally, the third determining module includes:
the monomer model determining unit is used for leading the element labels of the target characteristic elements into a monomer model library trained in advance to obtain a monomer model of the target member in the existing building;
and the model determining unit is used for integrating the monomer models of the target components to obtain the BIM model of the existing building.
Optionally, the model determining unit is specifically configured to:
determining a pair of interconnected members according to the connection relation of the target members;
and integrating the two associated monomer models by the member to obtain the BIM model of the existing building.
Optionally, the method further includes:
the design model determining module is used for determining a design model of the existing building according to the design information of the existing building;
the deviation value detection module is used for calling an error detection plug-in to detect the coordinate deviation values of the target components in the BIM model and the design model;
and the model adjusting module is used for adjusting the position coordinates of the target component in the BIM based on the position coordinates of the target component in the design model to obtain an adjusted BIM if the coordinate deviation value is detected to be larger than a coordinate deviation threshold value.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method for determining a BIM model based on point cloud data as described in any of the embodiments of the present invention.
In a fourth aspect, the embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the method for determining a BIM model based on point cloud data according to any one of the embodiments of the present invention.
The method comprises the steps of determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data; classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data; and determining the BIM model of the existing building according to the element label of the target characteristic element. According to the embodiment of the invention, the BIM model of the existing building can be determined by acquiring the point cloud data of the existing building, and the three-dimensional structure of the existing building can be accurately determined by referring to the model, so that the building construction efficiency is effectively improved.
Drawings
FIG. 1 is a schematic flow chart illustrating a method for determining a BIM based on point cloud data according to a first embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating a method for determining a BIM model based on point cloud data according to a second embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an apparatus for determining a BIM model based on point cloud data according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device in the fourth embodiment.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a schematic flowchart of a method for determining a BIM model based on point cloud data according to an embodiment of the present invention. The present embodiment is applicable to the case of determining a BIM model of an existing building. The method of the embodiment may be performed by an apparatus for determining a BIM model based on point cloud data, which may be implemented in hardware and/or software and may be configured in an electronic device. The method for determining the BIM model based on the point cloud data according to any embodiment of the application can be realized. As shown in fig. 1, the method specifically includes the following steps:
s110, determining target point cloud data of a target component in an existing building; and determining surface feature elements of the target component according to the target point cloud data.
In the embodiment, the existing building is a solid building which is formed by means of designing, producing, manufacturing and the like; the target component is an existing component used in the forming process of the existing building; wherein the existing component is either purchased from a component manufacturer or produced by itself.
The target point cloud data is a display mode of attribute data of the target member, can truly reflect the structural condition of the position of each target member in the existing building, can be displayed in a three-dimensional coordinate mode, and possibly comprises color information or reflection intensity information. The surface feature primitives are boundary features of the target member, such as point cloud data on the boundary where the member is located.
Specifically, determining the target point cloud data of the target member in the existing building may include: acquiring initial point cloud data of a target component in an existing building; and denoising the initial point cloud data to obtain target point cloud data. The problem that the BIM model of the existing building is unreal due to the fact that large noise exists in point cloud data when the model is built subsequently can be solved.
S120, classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements, and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data.
In this embodiment, the multi-class decision tree classifier adopts adaboost decision tree, AaBoost is an iterative process, which combines many weak classifiers with different weights to approach the optimal bayes classifier, and the embodiment adopts the maximum entropy decision tree as the weak classifier.
The element label is an identification mark of the target characteristic element, which can support setting, for example, a numerical value or a regular letter, and a combination of a number and a letter, thereby realizing effective distinguishing of each target characteristic element.
And S130, determining a BIM model of the existing building according to the element label of the target feature element.
In this embodiment, a Building Information processing (BIM) model of an existing Building may be automatically constructed by using software Revit, and a monomer model corresponding to a primitive tag may be determined according to the primitive tag of a target feature primitive, so as to determine a complete BIM model of the existing Building.
The method comprises the steps of determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data; classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data; and determining the BIM model of the existing building according to the element label of the target characteristic element. According to the embodiment of the invention, the BIM model of the existing building can be determined by acquiring the point cloud data of the existing building, and the three-dimensional structure of the existing building can be accurately determined by referring to the model, so that the building construction efficiency is effectively improved.
Example two
Fig. 2 is a schematic flowchart of a method for determining a BIM model based on point cloud data according to a second embodiment of the present invention. The embodiment is further expanded and optimized on the basis of the embodiment, and can be combined with any optional alternative in the technical scheme. As shown in fig. 2, the method includes:
s210, determining target point cloud data of a target component in an existing building; and determining surface feature elements of the target component according to the target point cloud data.
In this embodiment, optionally, determining target point cloud data of a target member in an existing building includes:
acquiring first point cloud data of a target component in an existing building in a three-dimensional laser scanning mode, and acquiring second point cloud data of the target component in the existing building in an unmanned aerial vehicle photographing mode; the unmanned aerial vehicle shooting mode is executed through an unmanned aerial vehicle oblique shooting system;
and fusing the first point cloud data and the second point cloud data to obtain target point cloud data of a target member in the existing building.
The three-dimensional laser scanning mode can be that a three-dimensional laser scanner is adopted to scan the existing building; the unmanned aerial vehicle photography mode can be that the unmanned aerial vehicle confirms to gather the angle and scan existing building. Specifically, the first point cloud data and the second point cloud data are point cloud data obtained by scanning the same existing building in different acquisition modes, the first point cloud data is positive acquisition data, and the second point cloud data is oblique acquisition data. The embodiment scans the existing building based on the forward direction through the three-dimensional laser scanning technology, scans the existing building based on the inclined direction through the unmanned aerial vehicle photographing mode, and fuses the data of the existing building and the existing building to avoid the problem of low point cloud data acquisition precision.
The embodiment can scan the existing building by adopting the three-dimensional laser scanner and unmanned aerial vehicle oblique photography, and acquire the point cloud data of the existing building, and the method comprises the two parts of earlier stage preparation and point cloud data acquisition. Before scanning begins, setting three-dimensional laser ground scanning control points and number, unmanned aerial vehicle image control points, number, flight path and the like according to the structure and scale of an existing building; in addition, the precision of the control point cloud can be set according to the purpose of the building BIM model in engineering, when a building with a small scanning scale is scanned, fewer control points can be set, and higher point cloud precision is set, so that a BIM model with higher precision can be constructed reversely according to the point cloud digital model at a later stage, and when a building with a large scanning scale is scanned, more control points are required to be set, the scanning precision is reduced, so that the scanning time is saved, and the construction efficiency is improved.
The embodiment further provides a specific execution process of the three-dimensional laser scanning mode, and the scanning device is a three-dimensional laser scanner as an example for explanation.
The three-dimensional laser scanning system mainly comprises a three-dimensional laser scanner, a computer, a power supply system, a support, related system matched software, an integrated video camera and the like. The three-dimensional laser scanner mainly comprises a laser emitter, a receiver, a time counter, a filter, a control computer board, a microcomputer, relevant software and the like. At present, three-dimensional laser scanners have three control modes, namely a data line connection PC (computer), a wireless remote control (such as a notebook computer and a tablet personal computer) and an integrated control panel; the data line is connected with the PC, and the data line can be divided into two types: ethernet cable and USB data cable, PC can use industrial PC, notebook computer or mobile workstation etc.. The wireless remote control mode comprises the following steps: WiFi, WLAN, Bluetooth can be used for notebook computers, PDAs, mobile phones and the like. And the control panel is integrated, and the control function is integrated on the equipment.
The selection of the three-dimensional laser scanner is a key in the data acquisition process, the scanning mode is determined to be a phase type or a pulse type according to the structural characteristics of a building and the surrounding environment, the scanner is selected according to the measurement precision requirement, and the three-dimensional laser scanners corresponding to different components are different; based on the composition of the building structure, the three-dimensional laser scanner can be selected with reference to table 1 below:
table 1 three-dimensional laser scanner selection standard reference table
A comparison of the property parameters of the three-dimensional laser scanner can be seen in table 2 below.
TABLE 2 comparison table of parameters of three-dimensional laser scanner
The three-dimensional laser scanner is installed as follows:
erecting a scanner; setting related parameters; determining a measurement name; testing the name of the station; measuring the coordinates of the station; the instrument height; calculating a rear view azimuth angle; a measurement coordinate system and a connection device are determined.
Reasonably planning and laying the control points of the survey stations according to the on-site scene of the building, the building structure and the requirement of scanning measurement precision; in order to ensure the data splicing quality, the data overlapping ratio between adjacent stations is more than 30%, and the inter-station distance is set within 50 m. Erecting a three-dimensional laser scanner on a control point of a measuring station; the method comprises the steps of obtaining three-dimensional point cloud data through the cooperation of a medium-long-range three-dimensional laser scanner, a short-range three-dimensional laser scanner and a total station, obtaining external point cloud data through the medium-long-range three-dimensional laser scanner, collecting internal data through the short-range scanner, and registering internal and external point clouds to obtain complete point cloud data. The method comprises the steps of starting up self-checking the three-dimensional laser scanner, establishing an engineering project after the scanner works normally, and setting a substation (namely the moving position of the scanner). Starting a scanner, selecting a proper scanning type (such as a scanning track), then roughly extracting a certain component by using fast scanning, observing the data condition, setting the horizontal and vertical scanning visual angles of the scanner, then adjusting the scanning speed to be slow, and adjusting the resolution to be the highest level; setting other scanning parameters such as grid quality, focusing and the like according to structural design requirements; and starting scanning after all the parameters are completely set, and starting scanning of the next substation after the scanning task of one substation is finished and the measured data is stored.
This embodiment also provides the collection process of unmanned aerial vehicle photography mode, is carried out by unmanned aerial vehicle inclination system.
The unmanned aerial vehicle oblique photography system mainly comprises an unmanned aerial vehicle, a flight platform, a flight control system, a ground station, a remote wireless communication device and a ground data processing system.
The reference index of the course and the side lap of the oblique photography images under different terrain conditions can be referred to the following table 3.
TABLE 3 reference table for reference index of course and lateral overlapping degree of oblique photography
The specific implementation steps comprise: and reasonably planning and distributing the survey station control points according to the on-site scene of the building, the building structure and the requirement of scanning measurement precision. And determining the course and track according to the reference indexes of the course and the lateral overlapping degree of the oblique photography images under different terrain conditions. And (4) starting self-checking the unmanned aerial vehicle, establishing an engineering project after confirming that the unmanned aerial vehicle oblique photography works normally, and setting a substation. Starting the unmanned aerial vehicle, and setting parameters according to design requirements; and starting scanning after all the parameters are completely set, and starting scanning of the next substation after the scanning task of one substation is finished and the measured data is stored. In order to improve the precision of the BIM model, the number of the control points is not less than 5, so that calibration data are provided for later-stage point cloud data model splicing; the arrangement of the control points requires that at least 3 control points are in different x, y and z coordinates, so that the uniqueness of model splicing is ensured. Setting basic control points, and utilizing a GPS (Global Positioning System) Positioning function to enable point cloud data acquired by each part to be in the same coordinate System, so that the integrity of point cloud model splicing is ensured, meanwhile, the one-to-one correspondence between control point coordinates and a real coordinate System is ensured, and the matching with other project three-dimensional models is ensured.
In this embodiment, optionally, determining the surface feature primitives of the target component according to the target point cloud data includes:
and extracting surface feature primitives of the target component under each type from the target point cloud data according to the type of the target component.
The type of the target component is a structural role of the target component in an existing building, for example, the type of the target component can be a beam or a column; the surface feature element is a boundary feature unit of the target component; the embodiment can further look at the three-dimensional composition information of the target component in more detail by determining the surface feature elements of the target component.
The extraction of the surface feature elements of the target member in the present embodiment can be achieved by the following operations.
Given input data for a primitive consisting of multiple surface primitives and corresponding support sets, a series of feature vectors { f }iIs generated from a feature vector corresponding to a surface primitive. Feature vector fiIs that a feature vector is (9+ 2N)p) A dimension vector. From fi＝[Ti,ηi,Vi,Bi,Li]Composition is carried out; wherein N ispIs the number of surface primitive types, the feature vector takes the following information from each input primitive: primitive type, principal direction, normal vector, spatial scale, and neighborhood statistics. The details of each subvector are set forth below:
(1)
is an index that encodes the surface primitive type. For example, in the case of a surface primitive set of a house having planes and cylinders, then the index {1,2} may be encoded as follows:
if the surface element is a plane, Ti1 is ═ 1; if the surface element is a cylinder, Ti＝2
(2)
Is the normal to each surface element and corresponds to the coefficients of the x, y and z terms in the estimated surface original model.
(3)
Is a unit vector which represents the primitive MiOf the main direction of the light beam.
(4)
Is a vector representing a relative scale of the range size (relative to the size of the entire point cloud data set), and sets the entire point cloud data set Pi＝(xi,yi,zi)；
Then P isiThe spatial range of (a) is: [ a, b, c ]]＝[(maxxi-minxi),(maxyi-minyi),(maxzi-minzi)]
(5)
A vector representing neighborhood statistics, representing the number of primitive types directly connected to each primitive. For example, for a building structure with only planes and cylinders, the cells directly connect 4 cylinders and 2 planes, then Li＝(2,4)。
S220, classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data.
In this embodiment, optionally, the classifying the surface feature primitives by using a pre-trained multi-class decision tree classifier to obtain the target feature primitives includes:
establishing a feature vector of a target component;
and classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier based on the feature vectors of the target component to obtain the target feature elements under different feature vectors.
Wherein establishing the feature vector of the target component may include: acquiring geometric characteristics of target components of different types; and establishing a feature vector of the target component according to the geometric features of the target component. The embodiment classifies the surface feature primitives by combining the feature vectors of the target member, further refines the classification dimension, and accordingly increases and optimizes the classification result of the surface feature primitives.
In this embodiment, the feature vector is composed of different variable types (e.g., classification, discrete and continuous variables), so that the classification is non-linear, and the embodiment uses an adaboost decision tree for classification. In the training phase, in each iteration of AdaBoost, a decision tree is first trained using feature vectors. And then weighting the classifier by using the weight of the AdaBoost algorithm according to the classification error, and adopting an exponential loss function.
Specifically, for multiple types of problems under actual conditions, the SAMME algorithm is adopted to utilize the population minimum value of multiple types of index losses. Given an input feature set containing K classes { fiThe algorithm is based on initializing all weak classifiers with weights of
For weak classifier NcIs required to carry out NcSub-iteration to strengthen decision tree
In each iteration, the algorithm first uses a training set and initial weights
To induce a decision tree and then express the empirical error rate as shown in equation (1) below.
Wherein: i is an indicator function: if it is
Is established, this means
At fiClass label and genuine label ofkIf they are not identical, then
Otherwise
As the error rate increases, the weight of the decision tree is adjusted to the following equation (2).
All weights will be normalized before the next iteration; in the passage of NcAfter the second iteration, the final assumption can be seen in equation (3) below.
And S230, importing the element labels of the target feature elements into a pre-trained monomer model library to obtain the monomer model of the target member in the existing building.
In the embodiment, the association relationship between the primitive tag and the monomer model is stored in the monomer model library; that is, a unique monomer model can be obtained according to a primitive label, and the monomer model library is obtained by training a primitive label training sample with the monomer model. Specifically, the monomer model library can be deployed in Revit automatic modeling software to rapidly determine the monomer models corresponding to the primitive labels.
And S240, integrating the monomer models of the target components to obtain a BIM model of the existing building.
In the present embodiment, the existing member is assembled from a plurality of target members, and thus a single model of the plurality of target members may be integrated to form a BIM model of the existing building. The embodiment determines the monomer model of the target component and integrates the monomer model, so that the three-dimensional data model of the existing building can be determined quickly and accurately.
In this embodiment, optionally, integrating the monomer models of the target components to obtain a BIM model of the existing building includes:
determining a pair of interconnected members according to the connection relation of the target members;
and integrating the two associated monomer models by the member to obtain the BIM model of the existing building.
The connection relation of the target components can be obtained from the earlier-stage design drawings of the existing buildings; model integration is carried out on the component pairs with the connection relation, the problem that model connection is wrong in the model building process of the existing building can be solved, and therefore model integration efficiency is greatly improved.
On the basis of the foregoing embodiment, optionally, the method of this embodiment further includes:
determining a design model of the existing building according to the design information of the existing building;
calling an error detection plug-in to detect coordinate deviation values of target components in the BIM model and the design model;
and if the detected coordinate deviation value is larger than the coordinate deviation threshold value, adjusting the position coordinates of the target component in the BIM based on the position coordinates of the target component in the design model to obtain the adjusted BIM.
In the present embodiment, in order to ensure the accuracy of the determined BIM model of the existing building, it is necessary to perform data calibration with reference to the design model. The Design model of the existing building can be a Computer Aided Design (CAD) model of the existing building, and the CAD model can be obtained based on a multi-class decision tree classifier trained in advance; the single model included in the design model can be a cuboid, a cylinder or a thin plate, and correspondingly, the single model in the BIM model can be a bridge deck, a main beam or a pier column.
In the error detection process of the BIM, the detected target component which does not meet the requirements can be marked and automatically reminded, so that subsequent adjusting personnel can determine the target component which needs to be subjected to data adjustment based on the mark of the target component, the workload of the adjusting personnel is reduced, and the consistency of the established BIM, the CAD model and the actual situation of the site is ensured.
EXAMPLE III
Fig. 3 is a schematic structural diagram of an apparatus for determining a BIM model based on point cloud data in the third embodiment of the present invention, which is applicable to the case of determining a BIM model of an existing building. The device is configured in the electronic equipment, and can realize the method for determining the BIM based on the point cloud data in any embodiment of the application. The device specifically comprises the following steps:
a first determining module 310, configured to determine target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data;
a second determining module 320, configured to classify the surface feature primitives by using a pre-trained multi-class decision tree classifier to obtain target feature primitives, and determine primitive labels of the target feature primitives; the multi-class decision tree classifier is obtained by training according to historical target point cloud data;
a third determining module 330, configured to determine a BIM model of the existing building according to the primitive tag of the target feature primitive.
On the basis of the foregoing embodiment, optionally, the first determining module 310 is specifically configured to:
acquiring first point cloud data of a target component in an existing building in a three-dimensional laser scanning mode, and acquiring second point cloud data of the target component in the existing building in an unmanned aerial vehicle photographing mode; the unmanned aerial vehicle shooting mode is executed through an unmanned aerial vehicle oblique shooting system;
and fusing the first point cloud data and the second point cloud data to obtain target point cloud data of a target component in the existing building.
On the basis of the foregoing embodiment, optionally, the first determining module 310 is further specifically configured to:
and extracting surface feature primitives of the target component under each type from the target point cloud data according to the type of the target component.
On the basis of the foregoing embodiment, optionally, the second determining module 320 is specifically configured to:
establishing a feature vector of the target component;
and classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier based on the feature vectors of the target member to obtain target feature elements under different feature vectors.
On the basis of the foregoing embodiment, optionally, the third determining module 330 includes:
the monomer model determining unit is used for leading the element labels of the target characteristic elements into a monomer model library trained in advance to obtain a monomer model of the target member in the existing building;
and the model determining unit is used for integrating the monomer models of the target components to obtain the BIM model of the existing building.
On the basis of the foregoing embodiment, optionally, the model determining unit is specifically configured to:
determining a pair of interconnected members according to the connection relation of the target members;
and integrating the two associated monomer models by the member to obtain the BIM model of the existing building.
On the basis of the foregoing embodiment, optionally, the apparatus of this embodiment further includes:
the design model determining module is used for determining a design model of the existing building according to the design information of the existing building;
the deviation value detection module is used for calling an error detection plug-in to detect the coordinate deviation values of the target components in the BIM model and the design model;
and the model adjusting module is used for adjusting the position coordinates of the target component in the BIM based on the position coordinates of the target component in the design model to obtain an adjusted BIM if the coordinate deviation value is detected to be larger than a coordinate deviation threshold value.
According to the device for determining the BIM model based on the point cloud data, disclosed by the embodiment III of the invention, the BIM model of the existing building can be determined by acquiring the point cloud data of the existing building, so that the three-dimensional structure of the existing building can be accurately determined by referring to the model, and the building construction efficiency is effectively improved.
The device for determining the BIM based on the point cloud data provided by the embodiment of the invention can execute the method for determining the BIM based on the point cloud data provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
Example four
Fig. 4 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention, as shown in fig. 4, the electronic device includes a processor 410, a memory 420, an input device 430, and an output device 440; the number of the processors 410 in the electronic device may be one or more, and one processor 410 is taken as an example in fig. 4; the processor 410, the memory 420, the input device 430 and the output device 440 in the electronic apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 4.
The memory 420, which is a computer-readable storage medium, may be used for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the method for determining a BIM model based on point cloud data in the embodiments of the present invention. The processor 410 executes various functional applications and data processing of the electronic device by executing software programs, instructions and modules stored in the memory 420, namely, implementing the method for determining the BIM model based on the point cloud data provided by the embodiment of the present invention.
The memory 420 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 420 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, memory 420 may further include memory located remotely from processor 410, which may be connected to an electronic device over 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 input device 430 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function control of the electronic apparatus, and may include a keyboard, a mouse, and the like. The output device 440 may include a display device such as a display screen.
EXAMPLE five
The present embodiments provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are used to implement the method for determining a BIM model based on point cloud data provided by the embodiments of the present invention.
Of course, the storage medium provided by the embodiments of the present invention contains computer-executable instructions, and the computer-executable instructions are not limited to the method operations described above, and may also perform related operations in the method for determining a BIM model based on point cloud data provided by any embodiments of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment 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 can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the above search apparatus, each included unit and module are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method of determining a BIM model based on point cloud data, the method comprising:
determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data;
classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data;
and determining the BIM model of the existing building according to the element label of the target characteristic element.
2. The method of claim 1, wherein determining target point cloud data for a target component in an existing building comprises:
acquiring first point cloud data of a target component in an existing building in a three-dimensional laser scanning mode, and acquiring second point cloud data of the target component in the existing building in an unmanned aerial vehicle photographing mode; the unmanned aerial vehicle shooting mode is executed through an unmanned aerial vehicle oblique shooting system;
and fusing the first point cloud data and the second point cloud data to obtain target point cloud data of a target component in the existing building.
3. The method of claim 1, wherein determining surface feature primitives for a target component from the target point cloud data comprises:
and extracting surface feature primitives of the target component under each type from the target point cloud data according to the type of the target component.
4. The method of claim 1, wherein classifying the surface feature primitives by a pre-trained multi-class decision tree classifier to obtain target feature primitives, comprises:
establishing a feature vector of the target component;
and classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier based on the feature vectors of the target member to obtain target feature elements under different feature vectors.
5. The method of claim 1, wherein determining the BIM model of the existing building from the primitive tags of the target feature primitives comprises:
importing the element labels of the target feature elements into a pre-trained monomer model library to obtain a monomer model of the target member in the existing building;
and integrating the monomer models of the target components to obtain the BIM model of the existing building.
6. The method of claim 5, wherein integrating the monomer model of the target component to obtain the BIM model of the existing building comprises:
determining a pair of interconnected members according to the connection relation of the target members;
and integrating the two associated monomer models by the member to obtain the BIM model of the existing building.
7. The method of claim 1, further comprising:
determining a design model of the existing building according to the design information of the existing building;
calling an error detection plug-in to detect coordinate deviation values of target components in the BIM model and the design model;
and if the coordinate deviation value is detected to be larger than a coordinate deviation threshold value, adjusting the position coordinates of the target component in the BIM based on the position coordinates of the target component in the design model to obtain an adjusted BIM.
8. An apparatus for determining a BIM model based on point cloud data, the apparatus comprising:
the system comprises a first determination module, a second determination module and a third determination module, wherein the first determination module is used for determining target point cloud data of a target component in an existing building; determining surface feature elements of the target component according to the target point cloud data;
the second determination module is used for classifying the surface feature elements by adopting a pre-trained multi-class decision tree classifier to obtain target feature elements and determining element labels of the target feature elements; the multi-class decision tree classifier is obtained by training according to historical target point cloud data;
and the third determining module is used for determining the BIM model of the existing building according to the element label of the target characteristic element.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement a method of determining a BIM model based on point cloud data as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method for determining a BIM model based on point cloud data according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011552398.1A CN112634340A (en) | 2020-12-24 | 2020-12-24 | Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011552398.1A CN112634340A (en) | 2020-12-24 | 2020-12-24 | Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112634340A true CN112634340A (en) | 2021-04-09 |
Family
ID=75324418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011552398.1A Pending CN112634340A (en) | 2020-12-24 | 2020-12-24 | Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112634340A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190040A (en) * | 2021-04-29 | 2021-07-30 | 集展通航(北京)科技有限公司 | Method and system for line inspection based on unmanned aerial vehicle video and railway BIM |
| CN113378274A (en) * | 2021-06-16 | 2021-09-10 | 升维科技有限公司 | Building construction monitoring method and system based on BIM |
| CN114387417A (en) * | 2022-03-25 | 2022-04-22 | 北京飞渡科技有限公司 | Three-dimensional building modeling method and device and three-dimensional building group modeling method |
| CN114821013A (en) * | 2022-07-01 | 2022-07-29 | 深圳大学 | Element detection method and device based on point cloud data and computer equipment |
| CN114840900A (en) * | 2022-05-18 | 2022-08-02 | 滁州学院 | Derivative BIM component automatic generation method based on i-GBDT technology |
| CN115311434A (en) * | 2022-10-10 | 2022-11-08 | 深圳大学 | Tree three-dimensional reconstruction method and device based on oblique photography and laser data fusion |
| CN115775116A (en) * | 2023-02-13 | 2023-03-10 | 华设设计集团浙江工程设计有限公司 | BIM-based road and bridge engineering management method and system |
| CN115809557A (en) * | 2022-12-06 | 2023-03-17 | 数云科际(深圳)技术有限公司 | BIM-based existing building digital twin model creating and examining method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107093206A (en) * | 2017-04-20 | 2017-08-25 | 中铁十局集团电务工程有限公司 | The method modeled using 3D laser scanner technique fast Bs IM |
| CN110348621A (en) * | 2019-07-02 | 2019-10-18 | 深圳大学 | A kind of prediction technique and device of building demolition waste yield |
-
2020
- 2020-12-24 CN CN202011552398.1A patent/CN112634340A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107093206A (en) * | 2017-04-20 | 2017-08-25 | 中铁十局集团电务工程有限公司 | The method modeled using 3D laser scanner technique fast Bs IM |
| CN110348621A (en) * | 2019-07-02 | 2019-10-18 | 深圳大学 | A kind of prediction technique and device of building demolition waste yield |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113190040A (en) * | 2021-04-29 | 2021-07-30 | 集展通航(北京)科技有限公司 | Method and system for line inspection based on unmanned aerial vehicle video and railway BIM |
| CN113190040B (en) * | 2021-04-29 | 2021-10-08 | 集展通航(北京)科技有限公司 | Method and system for line inspection based on unmanned aerial vehicle video and railway BIM |
| CN113378274A (en) * | 2021-06-16 | 2021-09-10 | 升维科技有限公司 | Building construction monitoring method and system based on BIM |
| CN113378274B (en) * | 2021-06-16 | 2023-12-01 | 鼎正工程咨询股份有限公司 | Building construction monitoring method and system based on BIM |
| CN114387417A (en) * | 2022-03-25 | 2022-04-22 | 北京飞渡科技有限公司 | Three-dimensional building modeling method and device and three-dimensional building group modeling method |
| CN114387417B (en) * | 2022-03-25 | 2022-05-27 | 北京飞渡科技有限公司 | Three-dimensional building modeling method and device and three-dimensional building group modeling method |
| CN114840900A (en) * | 2022-05-18 | 2022-08-02 | 滁州学院 | Derivative BIM component automatic generation method based on i-GBDT technology |
| CN114840900B (en) * | 2022-05-18 | 2023-06-20 | 滁州学院 | Automatic generation method of derivative BIM (building information modeling) component based on i-GBDT (generic block transformation) technology |
| CN114821013A (en) * | 2022-07-01 | 2022-07-29 | 深圳大学 | Element detection method and device based on point cloud data and computer equipment |
| CN115311434A (en) * | 2022-10-10 | 2022-11-08 | 深圳大学 | Tree three-dimensional reconstruction method and device based on oblique photography and laser data fusion |
| CN115809557A (en) * | 2022-12-06 | 2023-03-17 | 数云科际(深圳)技术有限公司 | BIM-based existing building digital twin model creating and examining method |
| CN115775116A (en) * | 2023-02-13 | 2023-03-10 | 华设设计集团浙江工程设计有限公司 | BIM-based road and bridge engineering management method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112634340A (en) | Method, device, equipment and medium for determining BIM (building information modeling) model based on point cloud data | |
| US10592780B2 (en) | Neural network training system | |
| CN111968229B (en) | High-precision map making method and device | |
| CN109614935B (en) | Vehicle damage assessment method and device, storage medium and electronic equipment | |
| US10043097B2 (en) | Image abstraction system | |
| CN112085840B (en) | Semantic segmentation method, semantic segmentation device, semantic segmentation equipment and computer readable storage medium | |
| CN110400363A (en) | Map constructing method and device based on laser point cloud | |
| CN114758337B (en) | Semantic instance reconstruction method, device, equipment and medium | |
| Axel et al. | Building damage assessment using airborne lidar | |
| Nyaruhuma et al. | Verification of 2D building outlines using oblique airborne images | |
| CN113916130B (en) | Building position measuring method based on least square method | |
| CN112305559A (en) | Power transmission line distance measuring method, device and system based on ground fixed-point laser radar scanning and electronic equipment | |
| CN116051747A (en) | House three-dimensional model reconstruction method, device and medium based on missing point cloud data | |
| CN115563732B (en) | Spraying track simulation optimization method and device based on virtual reality | |
| Song et al. | Development of comprehensive accuracy assessment indexes for building footprint extraction | |
| CN111458691B (en) | Building information extraction method and device and computer equipment | |
| CN114565916A (en) | Target detection model training method, target detection method and electronic equipment | |
| CN112053440A (en) | Method for determining individualized model and communication device | |
| CN117197388A (en) | Live-action three-dimensional virtual reality scene construction method and system based on generation of antagonistic neural network and oblique photography | |
| CN113822892B (en) | Evaluation method, device and equipment of simulated radar and computer storage medium | |
| CN112465977B (en) | Method for repairing three-dimensional model water surface loophole based on dense point cloud | |
| Comert et al. | Object based building extraction and building period estimation from unmanned aerial vehicle data | |
| Zeng | Automated Building Information Extraction and Evaluation from High-resolution Remotely Sensed Data | |
| CN113514053B (en) | Method and device for generating sample image pair and method for updating high-precision map | |
| Croitoru et al. | Monocular right-angle building hypothesis generation in regularized urban areas by pose clustering |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210409 |