CN117952322A - Engineering management system based on BIM technology - Google Patents
Engineering management system based on BIM technology Download PDFInfo
- Publication number
- CN117952322A CN117952322A CN202410345874.4A CN202410345874A CN117952322A CN 117952322 A CN117952322 A CN 117952322A CN 202410345874 A CN202410345874 A CN 202410345874A CN 117952322 A CN117952322 A CN 117952322A
- Authority
- CN
- China
- Prior art keywords
- point
- clustering
- ith
- area
- center
- 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.)
- Granted
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 54
- 239000013598 vector Substances 0.000 claims abstract description 54
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 29
- 230000011218 segmentation Effects 0.000 claims abstract description 22
- 238000012937 correction Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000005259 measurement Methods 0.000 claims description 38
- 230000001154 acute effect Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 5
- 239000000758 substrate Substances 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an engineering management system based on BIM technology. The system acquires point cloud data, performs voxel division, acquires an initial clustering center of a super-pixel segmentation algorithm according to the positions of points in a non-empty voxel block, acquires outliers according to the positions of each point in a clustering area and other points, the direction of a unit normal vector and the intensity value difference under the same angle, and corrects the initial clustering center to acquire a corrected clustering center; and acquiring a distance value according to the distance between the corrected cluster center and the initial cluster center, determining whether to carry out correction iteration on the cluster center, acquiring a final cluster center, and determining a voxel downsampling result. According to the method, the final clustering center in the super-pixel segmentation algorithm is obtained, so that the reference point of the non-empty voxel block is determined, the voxel downsampling result of the point cloud data is accurately obtained, and the condition that the voxel downsampling result is unreasonable due to the size of the voxel block is avoided.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an engineering management system based on BIM technology.
Background
The BIM technology can provide a comprehensive information platform, integrates various data such as design, construction, equipment, materials, cost, progress and the like, and is beneficial to realizing collaborative work and information sharing of each engineering stage. Under different progress of engineering, the construction of the BIM model needs to acquire point cloud data of a target object such as a building by using a laser radar, and because the point cloud data volume is huge, the efficiency of analyzing the point cloud data is reduced, therefore, in the existing method, the acquired point cloud data is processed by a voxel downsampling algorithm, the scale of the point cloud data is reduced, and meanwhile, the detail information of the target object is reserved. However, the voxel downsampling algorithm only considers the point cloud data information in each voxel block, when the size division of the voxel blocks is unreasonable, some detail information of the target object is lost, and further the details of the original surface of the target object cannot be completely restored, so that the BIM technology cannot accurately manage each stage of engineering.
Disclosure of Invention
In order to solve the technical problem that the original surface details of a target object cannot be completely restored due to the fact that only point cloud data information in each voxel block is considered in a voxel downsampling algorithm, when the size division of the voxel blocks is unreasonable, the technical scheme adopted by the engineering management system based on the BIM technology is as follows:
the invention provides an engineering management system based on BIM technology, which comprises the following contents:
The data acquisition module is used for acquiring point cloud data of the target object under different angles;
The outlier acquisition module is used for carrying out voxel division on the point cloud data to acquire at least two non-empty voxel blocks; acquiring an initial clustering center of a super-pixel segmentation algorithm according to the position of each point in each non-empty voxel block, and constructing a clustering area of each initial clustering center; acquiring an outlier of each point in each clustering area according to the position of each point in each clustering area and other points, the direction of a unit normal vector and the intensity value difference under the same angle;
The corrected cluster center acquisition module is used for correcting the initial cluster center according to the outlier and the coordinates of each point in each cluster area to obtain a corrected cluster center;
The final cluster center acquisition module is used for acquiring a distance value between the initial cluster center and the corrected cluster center according to the distance between the corrected cluster center and the initial cluster center in each cluster area; when the distance value is smaller than a preset distance value threshold, taking the corrected cluster center as a final cluster center; when the distance value is greater than or equal to a preset distance value threshold, carrying out correction iteration on the correction clustering center until the distance value between two adjacent correction clustering centers is smaller than the preset distance value threshold, stopping iteration, and taking the correction clustering center of the last time as a final clustering center;
And the data processing module is used for screening out target clustering centers according to the number of the points contained in the clustering area where each final clustering center is located, and acquiring a voxel downsampling result of the point cloud data.
Further, the method for obtaining the outlier of each point in each clustering area according to the positions of each point in each clustering area and other points, the direction of the unit normal vector and the intensity value difference under the same angle comprises the following steps:
Obtaining a distance measurement value between the ith point and other various points according to the difference between the included acute angle of the unit normal vector direction of the ith point and the other various points in the a clustering area and the space position;
acquiring the confidence coefficient between the ith point and other points according to the intensity value difference of the ith point and other points under the same angle;
correcting the distance measurement value according to the confidence coefficient between the ith point and other points to obtain an actual distance measurement value between the ith point and other points;
And the result of accumulating the actual distance measurement values between the ith point and other various points is used as an outlier of the ith point in the a clustering area.
Further, the method for obtaining the distance measurement value comprises the following steps:
Obtaining a unit vector of an angular bisector of an acute angle clamped between an ith point and other points in a unit normal vector direction of the ith point in the a clustering area as a target vector between the ith point and other points;
Taking a space coordinate difference vector between the ith point and other various points in the a clustering area as a position vector between the ith point and other various points;
And obtaining the distance metric value of the ith point and other various points in the ith clustering area according to the target vector and the position vector of the ith point and other various points in the ith clustering area and the included acute angle of the unit normal vector direction of the ith point and other various points in the ith clustering area.
Further, the calculation formula of the distance metric value is as follows:
in the method, in the process of the invention, A distance measurement value between an ith point and a jth point in the (a) th clustering area; /(I)An acute angle between the unit normal vector direction of the ith point and the jth point in the a-th clustering region; /(I)The target vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)The position vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)Taking the modulo symbol; sin is a sine function; /(I)Is a first preset constant, greater than 0.
Further, the confidence coefficient is calculated according to the following formula:
in the method, in the process of the invention, Confidence between the ith point and the jth point in the (a) th clustering area; m is the number of different angles; /(I)The intensity value of the ith point in the a-th clustering area under the m-th angle is obtained; /(I)The intensity value of the jth point in the a-th clustering area under the mth angle is obtained; /(I)As a function of absolute value.
Further, the method for obtaining the actual distance measurement value comprises the following steps:
in the method, in the process of the invention, The actual distance measurement value between the ith point and the jth point in the (a) th clustering area is obtained; /(I)Confidence between the ith point and the jth point in the (a) th clustering area; /(I)A distance measurement value between an ith point and a jth point in the (a) th clustering area; norm is a normalization function.
Further, the calculation formula of the modified clustering center is as follows:
in the method, in the process of the invention, A corrected cluster center in the ith cluster area; /(I)The total number of points contained in the ith cluster area; /(I)An outlier of an nth point in the ith cluster region; /(I)Is a second preset constant, greater than 0; /(I)The spatial coordinates of an nth point in the ith clustering area; norm is a normalization function.
Further, the distance value obtaining method comprises the following steps:
And taking the result of accumulating the distances between the corrected cluster center and the initial cluster center in each cluster area as the distance value between the initial cluster center and the corrected cluster center.
Further, the method for screening out the target clustering center according to the number of the points contained in the clustering area where each final clustering center is located and obtaining the voxel downsampling result of the point cloud data comprises the following steps:
acquiring the total number of the contained points in the clustering area where each final clustering center is located as a first number;
When the first quantity is larger than a preset first quantity threshold value, taking the corresponding final clustering center as a target clustering center;
And taking the target cluster center as a reference point of the corresponding non-empty voxel block, and acquiring a voxel downsampling result of the point cloud data.
Further, the method for acquiring the initial cluster center comprises the following steps:
the centroid of each non-empty voxel block is used as the initial cluster center of the super-pixel segmentation.
The invention has the following beneficial effects:
performing voxel division on point cloud data to obtain at least two non-empty voxel blocks, primarily dividing points corresponding to the point cloud data, improving the clustering efficiency of the points corresponding to the point cloud data, further obtaining initial clustering centers of a super-pixel segmentation algorithm according to the positions of the points in each non-empty voxel block, constructing clustering areas of each initial clustering center, further obtaining outliers of the points in each clustering area according to the positions of the points and other points in each clustering area and the direction of unit normal vectors and the intensity value difference under the same angle, accurately determining the positions of the points in each clustering area, further obtaining final clustering centers according to the outliers of the points, accurately obtaining the result of the super-pixel segmentation algorithm, further screening out target clustering centers according to the number of the points contained in the clustering area where each final clustering center is located, and reducing noise point interference; and taking the target clustering center as a reference point of a corresponding non-empty voxel block, accurately obtaining a voxel downsampling result, so that the obtained voxel downsampling result is not easy to lose the detail information of the original part of the target object due to unreasonable size of the voxel block, and accurately preserving the original detail information of the target object, and accurately managing each engineering stage by using a BIM technology.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of an engineering management system based on BIM technology according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purposes, the following detailed description is given below of a project management system based on BIM technology according to the present invention with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the engineering management system based on the BIM technology provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of an engineering management system based on BIM technology according to an embodiment of the present invention is shown, where the system includes: the system comprises a data acquisition module 10, an outlier acquisition module 20, a modified clustering center acquisition module 30, a final clustering center acquisition module 40 and a data processing module 50.
The data acquisition module 10 is configured to acquire point cloud data of a target object under different angles.
Specifically, in the embodiment of the present invention, a certain building is taken as an example, and the building is the target object. The laser radar is used for collecting point cloud data of the target object in different periods, so that cooperative work and information sharing of each stage of the target object can be realized. Meanwhile, the angle of the laser radar is adjusted, point cloud data of the target object under different angles are obtained, loss of detail information of the target object is avoided, and the target object is analyzed more accurately. The point cloud data acquired from the same position of the target object under different angles of the laser radar only have different intensity values. In order to accurately and efficiently analyze information of each stage of the target object, the embodiment of the invention processes the point cloud data acquired each time through a voxel downsampling algorithm, but the voxel downsampling algorithm only considers the point cloud data in each voxel block, so that when the size of the voxel block is not reasonably selected, partial detail information of the target object can be lost, and the original surface detail of the target object can not be completely restored. In order to solve the problem that partial detail information of a target object is possibly lost due to unreasonable size selection of a voxel block in a voxel downsampling algorithm, the embodiment of the invention further performs cluster analysis on points in each voxel block by utilizing a super-pixel segmentation algorithm on the basis of point cloud data voxel division, so that points corresponding to similar point cloud data are gathered into one type, a final cluster center is accurately determined, further a datum point of each voxel block is accurately acquired, and the voxel downsampling algorithm is not influenced by the size of the voxel block.
The super-pixel segmentation algorithm is a method for dividing points with similar colors, textures or brightness in an image into the same area, so that an image block with more semantic information is generated, but the super-pixel segmentation algorithm is generally carried out based on two-dimensional data, when three-dimensional data is analyzed, the iteration of a clustering center in each clustering process cannot simply adopt the mean value or the mass center of each clustering area again, because the mean value or the mass center of each clustering area is directly and simply used as the clustering center, spatial structure information of the three-dimensional data can be lost, the clustering result cannot accurately determine the datum point of each voxel block, further, detailed information of the original surface of a target object cannot be well represented, further, the characteristics of points corresponding to the point cloud data in the spatial structure are needed to be considered, the points with similar spatial structure characteristics and intensity values are gathered into the same type, the clustering center of each clustering area in the super-pixel segmentation algorithm is determined, the accuracy of the super-pixel segmentation algorithm is improved, further, the voxel downsampling result of the target object is accurately obtained, and the voxel downsampling result is not easy to be represented by unreasonable parts of the size of the block, and the original detailed information of the target object can be represented accurately.
It should be noted that, the voxel downsampling algorithm and the superpixel segmentation algorithm are both in the prior art, and will not be described in detail.
An outlier obtaining module 20, configured to perform voxel division on point cloud data to obtain at least two non-empty voxel blocks; acquiring an initial clustering center of a super-pixel segmentation algorithm according to the position of each point in each non-empty voxel block, and constructing a clustering area of each initial clustering center; and obtaining an outlier of each point in each clustering area according to the positions of each point in each clustering area and other points, the direction of the unit normal vector and the intensity value difference under the same angle.
Specifically, 3D point cloud data acquired at the same moment are subjected to voxel division through a voxel downsampling algorithm, and the embodiment of the invention sets the size of a voxel block as followsThe size of the voxel block may be set by the practitioner according to the actual situation, and is not limited here. So far, a preset number of non-empty voxel blocks are acquired. According to the position of each point cloud data corresponding point in each non-empty voxel block, acquiring the mass center of each non-empty voxel block, taking the mass center as an initial clustering center of a super-pixel segmentation algorithm, and constructing/> by taking each initial clustering center as the centerThe size of the cluster area can be set by the practitioner according to the actual situation, and is not limited herein. The method comprises the steps of obtaining the Euclidean distance between a point and each initial cluster center corresponding to the cluster region where the point is located, taking the initial cluster center corresponding to the smallest first distance as a first distance, taking the initial cluster center corresponding to the smallest first distance as a reference cluster center, and dividing the point into the cluster region where the reference cluster center is located. The method for acquiring the euclidean distance is in the prior art, and will not be described in detail. If at least two reference clustering centers exist, dividing the points into clustering areas where any one of the reference clustering centers is located. The size of the cluster area may be set by the practitioner according to the actual situation, and is not limited here. The embodiment of the invention amplifies the pixel block by 2 times to obtain the clustering area, and ensures that the points corresponding to similar point cloud data are divided into the same category as far as possible. Knowing that the cluster center of each cluster most represents the characteristic of the corresponding cluster, in order to better determine the reference point of each non-empty voxel block, the final cluster center in the super-pixel segmentation algorithm needs to be accurately determined, so that the spatial characteristic of each point in each cluster region is analyzed, the outlier of each point is obtained, the initial cluster center of each cluster region is further corrected and iterated until the final cluster center of each cluster region is obtained, the reference point of each non-empty voxel block is determined, and the influence of the size of the voxel block on the reference point of the acquired non-empty voxel block is eliminated.
Preferably, the method for acquiring the outlier is as follows: obtaining a distance measurement value between the ith point and other various points according to the difference between the included acute angle of the unit normal vector direction of the ith point and the other various points in the a clustering area and the space position; acquiring the confidence coefficient between the ith point and other points according to the intensity value difference of the ith point and other points under the same angle; correcting the distance measurement value according to the confidence coefficient between the ith point and other various points to obtain an actual distance measurement value between the ith point and other various points; and the result of accumulating the actual distance measurement values between the ith point and other various points is used as an outlier of the ith point in the a clustering area.
As an example, taking the ith point in the ith cluster area as an example, a specific method for obtaining the outlier of the ith point in the ith cluster area is as follows:
(1) And obtaining a distance measurement value.
In order to determine whether each initial cluster center in the super-pixel segmentation algorithm needs to carry out correction iteration, the embodiment of the invention analyzes the spatial positions of each point in each cluster area and other points, obtains the distance metric value of each point in each cluster area and other points, judges whether each point in each cluster area and other points are in the same plane, and when the points in the cluster areas are in the same plane, the more likely that the points in the cluster areas are in the same category is indicated.
In order to accurately and efficiently obtain distance measurement values of an ith point and other various points in an ith clustering area, the method embodiment obtains a unit vector of an angular bisector of an acute angle clamped between the ith point and other various points in the unit normal vector direction of the ith point in the ith clustering area as a target vector of the ith point and other corresponding points; the plane corresponding to the target vector is a plane projected by the unit normal vector of the ith point and other corresponding points; taking a space coordinate difference vector between the ith point and other points in the a clustering area as a position vector between the ith point and other corresponding points; when the direction difference between the target vector and the position vector of the i-th point and the other respective points is larger, it is explained that the i-th point and the other respective points are less likely to be in the same plane. Considering the situation that corners exist in the non-empty voxel block, namely the points in the non-empty voxel block are not in the same plane, and further the points in the clustering area corresponding to the non-empty voxel block are not in the same plane, at this time, the target vector and the position vector of the ith point and other points cannot better represent the plane positions of the ith point and other points, and therefore the sine value of the acute angle clamped between the ith point and the other points in the unit normal vector direction of the ith point is utilized for correction, so that the distance measurement value between the ith point and the other points is obtained. Therefore, according to the target vector and the position vector of the ith point and other various points in the ith clustering area and the included acute angle of the unit normal vector direction of the ith point and other various points in the ith clustering area, the distance measurement value of the ith point and other various points in the ith clustering area is obtained.
As an example, a calculation formula for obtaining a distance metric value between an ith point and a jth point in an a-th cluster region is:
in the method, in the process of the invention, A distance measurement value between an ith point and a jth point in the (a) th clustering area; /(I)An acute angle between the unit normal vector direction of the ith point and the jth point in the a-th clustering region; /(I)The target vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)The position vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)Taking the modulo symbol; sin is a sine function; /(I)Is a first preset constant, greater than 0.
Embodiments of the invention willSet to 1, avoid/>0, The practitioner can set/>, according to the actual situationIs not limited herein.
It should be noted that the number of the substrates,The smaller the i-th and j-th points in the a-th cluster region are, the more likely the i-th and j-th points are in the same surface of the target object,/>The smaller; /(I)The smaller the direction of the unit normal vector of the ith point and the jth point in the a-th clustering area is, the more consistent the direction of the unit normal vector of the ith point and the jth point in the a-th clustering area is, the more the ith point and the jth point in the a-th clustering area belong to the same plane,Smaller,/>The smaller; thus,/>The smaller the description is, the more likely the ith and jth points in the ith cluster region belong to the same plane.
According to the method for obtaining the distance measurement value between the ith point and the jth point in the ith clustering area, the distance measurement value between the ith point and other various points in the ith clustering area is obtained.
(2) And obtaining the confidence.
When any two points in each clustering area are positioned on the same plane or similar curved surfaces, the intensity values of the two points in the same direction of the laser radar are the same or similar, otherwise, the intensity values of the two points in the same direction of the laser radar are greatly different. Therefore, according to the intensity value difference of the ith point and other points in the a-th clustering area under the same angle, the confidence coefficient between the ith point and other points is obtained.
As an example, a calculation formula for obtaining the confidence between the ith point and the jth point in the ith cluster area is:
in the method, in the process of the invention, Confidence between the ith point and the jth point in the (a) th clustering area; m is the number of different angles; /(I)The intensity value of the ith point in the a-th clustering area under the m-th angle is obtained; /(I)The intensity value of the jth point in the a-th clustering area under the mth angle is obtained; /(I)As a function of absolute value.
It should be noted that the number of the substrates,The smaller the intensity values of the ith point and the jth point in the (a) th clustering area under the (m) th angle of the laser radar are, the more similar the intensity values of the ith point and the jth point in the (a) th clustering area are, the more likely the ith point and the jth point are in the same plane,/>The smaller the more accurately the i-th and j-th points in the a-th cluster region are, the more likely they are to be in the same plane.
And obtaining the confidence coefficient between the ith point and other various points in the ith clustering area according to the method for obtaining the confidence coefficient between the ith point and the jth point in the ith clustering area.
(3) And obtaining an actual distance measurement value.
The distance measurement value only considers the space position coordinates of two points and the relation of normal vectors corresponding to the two points, and when two different planes possibly exist in a certain clustering area at the same time, the distance measurement value gathers the points in the two different planes into one type, so that the representativeness of a clustering result is poor; therefore, the distance measurement value is corrected according to the confidence coefficient between the ith point and other various points in the ith clustering area, the actual distance measurement value between the ith point and other various points in the ith clustering area is obtained, and the position relation between the ith point and other various points in the ith clustering area is accurately reflected.
As an example, a calculation formula for obtaining an actual distance metric value between the ith point and the jth point in the ith cluster area is:
in the method, in the process of the invention, The actual distance measurement value between the ith point and the jth point in the (a) th clustering area is obtained; /(I)Confidence between the ith point and the jth point in the (a) th clustering area; /(I)A distance measurement value between an ith point and a jth point in the (a) th clustering area; norm is a normalization function.
It should be noted that the number of the substrates,The smaller the description is that the more likely the ith and jth points are in the same plane within the ith cluster region,/>Smaller,/>The smaller; /(I)The smaller the description is that the more likely the ith and jth points themselves are in the same plane within the ith cluster region,/>The smaller; thus,/>The smaller the description is, the more likely the ith and jth points are in the same plane within the ith cluster region.
According to the method for obtaining the actual distance measurement value between the ith point and the jth point in the ith clustering area, the actual distance measurement value between the ith point and other various points in the ith clustering area is obtained.
(4) An outlier is obtained.
When the actual distance measurement value between the ith point and other various points in the ith clustering area is smaller, the fact that the ith point and other various points in the ith clustering area are more likely to be in the same plane is indicated, and therefore the result of accumulating the actual distance measurement value between the ith point and other various points is taken as an outlier of the ith point in the ith clustering area, the smaller the outlier is, and the fact that the ith point and other points in the ith clustering area are more likely to be in the same plane is indicated.
And acquiring the outlier of each point in each clustering area according to the method for acquiring the outlier of the ith point in the a clustering area.
And the corrected cluster center acquisition module 30 is used for correcting the initial cluster center according to the outlier and the coordinates of each point in each cluster area to obtain a corrected cluster center.
Specifically, after the outliers corresponding to the points in each clustering area are obtained, the initial clustering center in each clustering area needs to be corrected according to the magnitude of the outliers of the points, so that the influence of abnormal points or noise points on clusters is reduced, and meanwhile, the points in the same plane are clustered into the same class, namely, the weight coefficient of the abnormal points is reduced when the initial clustering center is corrected. Therefore, the initial clustering center is corrected according to the outlier and the coordinates of each point in each clustering area, and a corrected clustering center is obtained.
Taking the ith clustering area as an example, a calculation formula for obtaining a corrected clustering center in the ith clustering area is as follows:
in the method, in the process of the invention, A corrected cluster center in the ith cluster area; /(I)The total number of points contained in the ith cluster area; /(I)An outlier of an nth point in the ith cluster region; /(I)Is a second preset constant, greater than 0; /(I)The spatial coordinates of an nth point in the ith clustering area; norm is a normalization function.
Embodiments of the invention willSet to 1, avoid denominator to 0, and the practitioner can set/>, according to the actual situationIs not limited herein.
It should be noted that the number of the substrates,The smaller the n-th point in the i-th cluster region is, the more likely it is that the n-th point in the i-th cluster region is in the same plane as the other points in the i-th cluster region,/>The larger the/>The larger the/>The larger the duty cycle involved in revising the initial cluster center. So far, the corrected cluster center in the ith cluster area is obtained.
And acquiring the corrected cluster center in each cluster area according to the method for acquiring the corrected cluster center in the ith cluster area.
A final cluster center obtaining module 40, configured to obtain a distance value between the initial cluster center and the revised cluster center according to a distance between the revised cluster center and the initial cluster center in each cluster region; when the distance value is smaller than a preset distance value threshold, taking the corrected cluster center as a final cluster center; and when the distance value is greater than or equal to a preset distance value threshold, carrying out correction iteration on the correction clustering center until the distance value between two adjacent correction clustering centers is smaller than the preset distance value threshold, stopping iteration, and taking the correction clustering center of the last time as a final clustering center.
Specifically, in order to determine whether the modified cluster center is a final cluster center of the super-pixel segmentation algorithm, the embodiment of the invention obtains the Euclidean distance between the modified cluster center and the initial cluster center in each cluster region, and the result of accumulating the Euclidean distance between the modified cluster center and the initial cluster center in each cluster region is used as the distance value between the initial cluster center and the modified cluster center. The embodiment of the invention sets the preset distance value threshold value asHere, sum is the total number of clustered regions, and the operator may set the preset distance value threshold according to the actual situation, which is not limited herein. When the distance value is smaller than a preset distance value threshold, taking the corrected cluster center as a final cluster center; when the distance value is greater than or equal to a preset distance value threshold, carrying out correction iteration on the correction clustering center, namely, taking the correction clustering center as the center to construct/>And (3) analyzing, and iteratively obtaining a second modified cluster center until the distance value between two adjacent modified cluster centers is smaller than a preset distance value threshold value, stopping iteration, and taking the last modified cluster center in each cluster area as a final cluster center. So far, the final clustering center of the super-pixel segmentation algorithm is obtained.
The data processing module 50 is configured to screen out a target cluster center according to the number of points included in the cluster area where each final cluster center is located, and obtain a voxel downsampling result of the point cloud data.
Specifically, the total number of the points contained in the clustering area where each final clustering center is located is obtained and used as a first number; in the embodiment of the invention, the preset first quantity threshold value is set to be 5, and an implementer can set the size of the preset first quantity threshold value according to actual conditions, without limitation, when the first quantity is smaller than or equal to the preset first quantity threshold value, the cluster area corresponding to the final cluster center contains too few points, the points in the corresponding cluster area have a larger probability of noise points, and the cluster area should be removed; when the first quantity is larger than a preset first quantity threshold value, taking the corresponding final clustering center as a target clustering center; and taking the target cluster center as a reference point of the corresponding non-empty voxel block, and further acquiring a voxel downsampling result of the point cloud data.
The reference points of the non-empty voxel blocks obtained by the embodiment of the invention are not greatly influenced by the size of the voxel blocks, so that the condition that the reference points of the non-empty voxel blocks are unreasonable due to unreasonable size setting of the voxel blocks is eliminated, and the original surface detail information of a target object can be accurately reflected by a voxel downsampling result.
The present invention has been completed.
In summary, the embodiment of the invention obtains the point cloud data, performs voxel division, obtains the initial clustering center of the super-pixel segmentation algorithm according to the positions of points in the non-empty voxel block, obtains outliers according to the positions of each point in the clustering area and other points, the direction of the unit normal vector and the intensity value difference under the same angle, and corrects the initial clustering center to obtain a corrected clustering center; and acquiring a distance value according to the distance between the corrected cluster center and the initial cluster center, determining whether to carry out correction iteration on the cluster center, acquiring a final cluster center, and determining a voxel downsampling result. According to the method, the final clustering center in the super-pixel segmentation algorithm is obtained, so that the reference point of the non-empty voxel block is determined, the voxel downsampling result of the point cloud data is accurately obtained, and the condition that the voxel downsampling result is unreasonable due to the size of the voxel block is avoided.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. A system for engineering management based on BIM technology, the system comprising:
The data acquisition module is used for acquiring point cloud data of the target object under different angles;
The outlier acquisition module is used for carrying out voxel division on the point cloud data to acquire at least two non-empty voxel blocks; acquiring an initial clustering center of a super-pixel segmentation algorithm according to the position of each point in each non-empty voxel block, and constructing a clustering area of each initial clustering center; acquiring an outlier of each point in each clustering area according to the position of each point in each clustering area and other points, the direction of a unit normal vector and the intensity value difference under the same angle;
The corrected cluster center acquisition module is used for correcting the initial cluster center according to the outlier and the coordinates of each point in each cluster area to obtain a corrected cluster center;
The final cluster center acquisition module is used for acquiring a distance value between the initial cluster center and the corrected cluster center according to the distance between the corrected cluster center and the initial cluster center in each cluster area; when the distance value is smaller than a preset distance value threshold, taking the corrected cluster center as a final cluster center; when the distance value is greater than or equal to a preset distance value threshold, carrying out correction iteration on the correction clustering center until the distance value between two adjacent correction clustering centers is smaller than the preset distance value threshold, stopping iteration, and taking the correction clustering center of the last time as a final clustering center;
And the data processing module is used for screening out target clustering centers according to the number of the points contained in the clustering area where each final clustering center is located, and acquiring a voxel downsampling result of the point cloud data.
2. The engineering management system based on the BIM technology as claimed in claim 1, wherein the method for obtaining the outlier of each point in each cluster area according to the position of each point in each cluster area and other points, the direction of the unit normal vector, and the intensity value difference under the same angle is as follows:
Obtaining a distance measurement value between the ith point and other various points according to the difference between the included acute angle of the unit normal vector direction of the ith point and the other various points in the a clustering area and the space position;
acquiring the confidence coefficient between the ith point and other points according to the intensity value difference of the ith point and other points under the same angle;
correcting the distance measurement value according to the confidence coefficient between the ith point and other points to obtain an actual distance measurement value between the ith point and other points;
And the result of accumulating the actual distance measurement values between the ith point and other various points is used as an outlier of the ith point in the a clustering area.
3. The engineering management system based on the BIM technology as claimed in claim 2, wherein the distance metric obtaining method is as follows:
Obtaining a unit vector of an angular bisector of an acute angle clamped between an ith point and other points in a unit normal vector direction of the ith point in the a clustering area as a target vector between the ith point and other points;
Taking a space coordinate difference vector between the ith point and other various points in the a clustering area as a position vector between the ith point and other various points;
And obtaining the distance metric value of the ith point and other various points in the ith clustering area according to the target vector and the position vector of the ith point and other various points in the ith clustering area and the included acute angle of the unit normal vector direction of the ith point and other various points in the ith clustering area.
4. The project management system according to claim 3, wherein the calculation formula of the distance metric value is:
in the method, in the process of the invention, A distance measurement value between an ith point and a jth point in the (a) th clustering area; /(I)An acute angle between the unit normal vector direction of the ith point and the jth point in the a-th clustering region; /(I)The target vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)The position vector of the ith point and the jth point in the (a) th clustering area is obtained; /(I)Taking the modulo symbol; sin is a sine function; /(I)Is a first preset constant, greater than 0.
5. The engineering management system based on the BIM technique according to claim 2, wherein the confidence coefficient is calculated by the formula:
in the method, in the process of the invention, Confidence between the ith point and the jth point in the (a) th clustering area; m is the number of different angles; The intensity value of the ith point in the a-th clustering area under the m-th angle is obtained; /(I) The intensity value of the jth point in the a-th clustering area under the mth angle is obtained; /(I)As a function of absolute value.
6. The engineering management system based on the BIM technology as claimed in claim 2, wherein the method for obtaining the actual distance metric is as follows:
in the method, in the process of the invention, The actual distance measurement value between the ith point and the jth point in the (a) th clustering area is obtained; /(I)Confidence between the ith point and the jth point in the (a) th clustering area; /(I)A distance measurement value between an ith point and a jth point in the (a) th clustering area; norm is a normalization function.
7. The engineering management system based on the BIM technology as claimed in claim 1, wherein the calculation formula of the modified clustering center is:
in the method, in the process of the invention, A corrected cluster center in the ith cluster area; /(I)The total number of points contained in the ith cluster area; an outlier of an nth point in the ith cluster region; /(I) Is a second preset constant, greater than 0; /(I)The spatial coordinates of an nth point in the ith clustering area; norm is a normalization function.
8. The engineering management system based on the BIM technology as claimed in claim 1, wherein the distance value obtaining method is as follows:
And taking the result of accumulating the distances between the corrected cluster center and the initial cluster center in each cluster area as the distance value between the initial cluster center and the corrected cluster center.
9. The engineering management system based on the BIM technology as claimed in claim 1, wherein the method for screening out the target cluster center according to the number of points contained in the cluster area where each final cluster center is located, and obtaining the voxel downsampling result of the point cloud data is as follows:
acquiring the total number of the contained points in the clustering area where each final clustering center is located as a first number;
When the first quantity is larger than a preset first quantity threshold value, taking the corresponding final clustering center as a target clustering center;
And taking the target cluster center as a reference point of the corresponding non-empty voxel block, and acquiring a voxel downsampling result of the point cloud data.
10. The engineering management system based on the BIM technology as claimed in claim 1, wherein the obtaining method of the initial cluster center is as follows:
the centroid of each non-empty voxel block is used as the initial cluster center of the super-pixel segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410345874.4A CN117952322B (en) | 2024-03-26 | 2024-03-26 | Engineering management system based on BIM technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410345874.4A CN117952322B (en) | 2024-03-26 | 2024-03-26 | Engineering management system based on BIM technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117952322A true CN117952322A (en) | 2024-04-30 |
| CN117952322B CN117952322B (en) | 2024-06-11 |
Family
ID=90803244
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410345874.4A Active CN117952322B (en) | 2024-03-26 | 2024-03-26 | Engineering management system based on BIM technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117952322B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600622A (en) * | 2016-12-06 | 2017-04-26 | 西安电子科技大学 | Point cloud data partitioning method based on hyper voxels |
| CN114638953A (en) * | 2022-02-22 | 2022-06-17 | 深圳元戎启行科技有限公司 | Point cloud data segmentation method and device and computer readable storage medium |
| US20230072731A1 (en) * | 2021-08-30 | 2023-03-09 | Thomas Enxu LI | System and method for panoptic segmentation of point clouds |
| KR102516008B1 (en) * | 2022-09-21 | 2023-03-30 | (주)이노시뮬레이션 | Method of down sampling voxel using point cloud data and apparatus performing thereof |
| CN116167668A (en) * | 2023-04-26 | 2023-05-26 | 山东金至尊装饰工程有限公司 | BIM-based green energy-saving building construction quality evaluation method and system |
| CN116778085A (en) * | 2023-06-21 | 2023-09-19 | 郑州大学 | Point cloud simplifying method and system based on clustering and saliency |
| WO2023174555A1 (en) * | 2022-03-18 | 2023-09-21 | Synthetic Dimension Gmbh | Computer-implemented reconstruction of interior rooms |
| CN117523084A (en) * | 2023-10-27 | 2024-02-06 | 中铁二院工程集团有限责任公司 | Automatic three-dimensional reconstruction method of mobile scanning point cloud and electronic equipment |
-
2024
- 2024-03-26 CN CN202410345874.4A patent/CN117952322B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106600622A (en) * | 2016-12-06 | 2017-04-26 | 西安电子科技大学 | Point cloud data partitioning method based on hyper voxels |
| US20230072731A1 (en) * | 2021-08-30 | 2023-03-09 | Thomas Enxu LI | System and method for panoptic segmentation of point clouds |
| CN114638953A (en) * | 2022-02-22 | 2022-06-17 | 深圳元戎启行科技有限公司 | Point cloud data segmentation method and device and computer readable storage medium |
| WO2023174555A1 (en) * | 2022-03-18 | 2023-09-21 | Synthetic Dimension Gmbh | Computer-implemented reconstruction of interior rooms |
| KR102516008B1 (en) * | 2022-09-21 | 2023-03-30 | (주)이노시뮬레이션 | Method of down sampling voxel using point cloud data and apparatus performing thereof |
| CN116167668A (en) * | 2023-04-26 | 2023-05-26 | 山东金至尊装饰工程有限公司 | BIM-based green energy-saving building construction quality evaluation method and system |
| CN116778085A (en) * | 2023-06-21 | 2023-09-19 | 郑州大学 | Point cloud simplifying method and system based on clustering and saliency |
| CN117523084A (en) * | 2023-10-27 | 2024-02-06 | 中铁二院工程集团有限责任公司 | Automatic three-dimensional reconstruction method of mobile scanning point cloud and electronic equipment |
Non-Patent Citations (4)
| Title |
|---|
| ABDELAAZIZ MAHDAOUI ET AL: "3D Point Cloud Simplification Based on k-Nearest Neighbor and Clustering", 《ADVANCES IN MULTIMEDIA》, 15 July 2020 (2020-07-15) * |
| 刘金典等: "基于建筑信息模型和激光扫描的装配式 建造管理与质量控制", 《同 济 大 学 学 报（自然科学版）》, 31 January 2020 (2020-01-31) * |
| 江宇: "基于三维扫描和 BIM 的装配式结构 精度检测和预拼装", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》, 15 February 2023 (2023-02-15) * |
| 郑书富;李渭;刘莉;: "基于超体素聚类的三维点云轮廓特征提取", 兰州文理学院学报(自然科学版), no. 03, 10 May 2018 (2018-05-10) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117952322B (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108537876B (en) | Three-dimensional reconstruction method, device, equipment and storage medium | |
| CN111780763B (en) | Visual positioning method and device based on visual map | |
| CN107179768B (en) | Obstacle identification method and device | |
| CN109165680B (en) | Single-target object dictionary model improvement method in indoor scene based on visual SLAM | |
| EP3317852B1 (en) | Method in constructing a model of a scenery and device therefor | |
| CN112347550A (en) | Coupling type indoor three-dimensional semantic graph building and modeling method | |
| CN108154104A (en) | A kind of estimation method of human posture based on depth image super-pixel union feature | |
| CN109685841B (en) | Registration method and system of three-dimensional model and point cloud | |
| US11568598B2 (en) | Method and device for determining an environment map by a server using motion and orientation data | |
| CN112328715A (en) | Visual positioning method, training method of related model, related device and equipment | |
| CN114217665A (en) | Camera and laser radar time synchronization method, device and storage medium | |
| CN115222884A (en) | Space object analysis and modeling optimization method based on artificial intelligence | |
| CN114972882A (en) | Wear surface damage depth estimation method and system based on multi-attention machine system | |
| CN116844124A (en) | Three-dimensional object detection frame labeling method, three-dimensional object detection frame labeling device, electronic equipment and storage medium | |
| CN117952322B (en) | Engineering management system based on BIM technology | |
| CN116894876A (en) | 6-DOF positioning method based on real-time image | |
| CN108319717B (en) | The calculation method of optimum lattice size range when a kind of vector point is converted to grid point | |
| CN116704024A (en) | Vehicle-mounted laser point cloud pose graph optimization method and system combining multiclass constraints | |
| JP6853150B2 (en) | Sorting equipment, methods and programs | |
| CN114399547B (en) | Monocular SLAM robust initialization method based on multiframe | |
| JP6761388B2 (en) | Estimator and program | |
| CN113762310B (en) | Point cloud data classification method, device, computer storage medium and system | |
| CN114782640A (en) | Strength information matching rapid laser SLAM algorithm based on quadruped robot | |
| CN117197410B (en) | Virtual splicing method, device and equipment for steel structure and storage medium | |
| CN118097265A (en) | Visual SLAM optimization method based on deep learning and GPU acceleration under dynamic scene |
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 |