Method for marking building object in three-dimensional point cloud data

Abstract

The invention discloses a method for marking a building object in three-dimensional point cloud data, and particularly relates to the technical field of three-dimensional data processing, and the method comprises the steps: S1, marking region determination: determining data in a target city region as a target marking region, and dividing the target marking region into sub-regions according to an equal-time division mode, sequentially numbering the sub-regions in the target labeling region as 1, 2,..., i,..., n; s2, data acquisition: scanning a target city area by using a three-dimensional scanner to obtain three-dimensional point cloud data including a building object, transmitting the three-dimensional point cloud data to a three-dimensional point cloud data processing step, and sequentially numbering the three-dimensional point cloud data as 1, 2,..., j,..., m according to the number of point clouds; according to the invention, through technologies such as laser scanning, high-precision point cloud data containing multi-dimensional information such as three-dimensional coordinates, laser reflection intensity and color of the object can be obtained, the actual condition of the building object can be reflected more truly, and the precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of three-dimensional data processing, and more specifically, to a method for marking building objects in three-dimensional point cloud data. Background Art

[0002] With the development of technologies such as laser scanning and photogrammetry, a large amount of three-dimensional point cloud data is generated and widely used in many scenarios such as urban modeling, building monitoring, and autonomous driving map construction; therefore, developing a system that can automatically or semi-automatically annotate building objects has important practical value.

[0003] The traditional method of annotating building objects includes data collection, data processing, annotation steps, database management, and display steps; the data collection step is responsible for obtaining the attribute data of the building object, providing basic materials for subsequent annotation; the data processing step cleans, unifies, and classifies the collected data; the annotation step provides a variety of annotation methods to assist in annotation; the database management step is responsible for storing the annotated data, and stores the annotation information of the building object and the original collected data in the database according to certain rules, which is convenient for subsequent query, update, and management; the display step displays the annotated building object information in an intuitive way; the annotated building floor plan, section view, three-dimensional model, etc. can be displayed to the user.

[0004] However, it still has some shortcomings in actual use, such as low data accuracy. Traditional annotation systems mostly rely on manual measurement and two-dimensional images to collect data, which makes it difficult to obtain comprehensive and accurate three-dimensional spatial information of building objects, and the accurate size and spatial position relationship of complex building structures are unclear; data processing efficiency is low, and the data format obtained by traditional data collection methods is relatively single, and a lot of time is required for manual sorting and conversion to make it suitable for annotation work; data collection is incomplete, and the annotation results of traditional annotation systems are separated from the original data. Once the building object changes, the data needs to be re-collected and annotated, resulting in untimely data updates, which makes data collection incomplete.

[0005] Therefore, it is urgent to provide a method for marking building objects in 3D point cloud data to solve the problems of low data accuracy, low data processing efficiency and incomplete data collection in the existing annotation system. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method for marking building objects in three-dimensional point cloud data, and solves the problems raised in the above-mentioned background technology through the following scheme.

[0007] To achieve the above object, the present invention provides the following technical solution: a method for marking building objects in three-dimensional point cloud data, comprising:

[0008] S1, determining the marked area, for determining the data in the target city area as the target marked area, and dividing it into sub-areas according to the equal time division method, and numbering each sub-area in the target marked area as 1, 2, ..., i, ..., n in sequence;

[0009] S2, data acquisition: use a 3D scanner to scan the target urban area, obtain 3D point cloud data containing building objects, and transmit it to the 3D point cloud data processing step, and number them in sequence as 1, 2, ..., j, ..., m according to the number of point clouds;

[0010] S3, 3D point cloud data processing, for analyzing and processing the 3D point cloud data obtained in S2 to obtain the influence coefficient of the 3D building object size data, the influence coefficient of the 3D building object space shape data and the influence coefficient of the 3D building object material color data, and transmit them to the 3D point cloud data analysis step;

[0011] S4, 3D point cloud data analysis, importing the influence coefficient value of the 3D building object size data, the influence coefficient value of the 3D building object space shape data and the coefficient value of the 3D building object material color data obtained in S3 into the building object labeling method rationality index mathematical model, obtaining the building object labeling method rationality index value, and transmitting it to the building judgment step;

[0012] S5, building judgment, comparing the building object labeling method rationality index value obtained in S4 with the preset building object labeling method rationality index values ​​under multiple standards, calculating the minimum value of the difference between the building object labeling method rationality index value and the preset building object labeling method rationality index under multiple standards, when the minimum value of the difference value is less than the preset difference value, transmitting the judgment result and the data in the target labeling area to the labeling execution step and the data interactive transmission step, when the minimum value of the difference value is greater than the preset difference value, transmitting the data in the target labeling area as historical data to the data collection step;

[0013] S6, labeling execution, automatically assigning labeling labels to the building objects according to the judgment result obtained in S5 and the data in the target labeling area, presenting the labeling information, and then transmitting it to the data interaction transmission step;

[0014] S7 data interaction transmission is used to transmit the obtained judgment results, data in the target computing server and annotation information to the user information terminal, providing reference data for users to make adjustment measures.

[0015] Preferably, the three-dimensional point cloud data includes parameters influencing the three-dimensional building object size data, parameters influencing the three-dimensional building object space shape data, and parameters influencing the three-dimensional building object material color data.

[0016] Preferably, the influencing parameters of the three-dimensional building object size data include the number of building object point cloud points, denoted as P; the building object volume, denoted as V; the building object surface area, denoted as S; the building object length, denoted as L; the building object width, denoted as W; the building object height, denoted as H; the influencing parameters of the three-dimensional building object spatial shape data include the curvature of the building component, denoted as μ; the flatness of the building object wall, denoted as F; the building object point density, denoted as ρ; the building object point spacing, denoted as d; the influencing parameters of the three-dimensional building object material color data include the building object reflectivity, denoted as r; the building object color RGB values, denoted as R, G, and B respectively; and the building object texture density, denoted as t.

[0017] Preferably, the three-dimensional point cloud data processing step includes a three-dimensional building object size data influence coefficient calculation step, a three-dimensional building object space shape data influence coefficient calculation step and a three-dimensional building object material color data coefficient calculation step.

[0018] Preferably, the three-dimensional building object size data influence coefficient calculation step is used to import the three-dimensional building object size data influence parameters into the three-dimensional building object size data influence coefficient mathematical model to obtain the three-dimensional building object size data influence coefficient value; the three-dimensional building object space shape data influence coefficient calculation step is used to import the three-dimensional building object space shape data influence parameters into the three-dimensional building object space shape data influence coefficient mathematical model to obtain the three-dimensional building object space shape data influence coefficient value; the three-dimensional building object material color data coefficient calculation step is used to import the three-dimensional building object material color data parameters into the three-dimensional building object material color data coefficient mathematical model to obtain the three-dimensional building object material color data coefficient value.

[0019] Preferably, the mathematical model of the influence coefficient of the three-dimensional building object size data is specifically:

[0020]

[0021] Preferably, the mathematical model of the influence coefficient of the spatial shape data of the three-dimensional building object is specifically:

[0022]

[0023] where ρ j represents the building object point density of the j-th point cloud, represents the average building object point density; Represents the average building object point spacing, μ max Represents the maximum curvature of the building component, μ min Indicates the minimum curvature of a building component.

[0024] Preferably, the three-dimensional building object material color data coefficient mathematical model is specifically:

[0025]

[0026] Where R j represents the R value of the building object in the jth point cloud, R max Indicates the maximum R value of the building object color, R min Indicates the minimum R value of the building object color, G j represents the G value of the building object in the jth point cloud, G max Indicates the maximum G value of the building object color, G min Indicates the minimum G value of the building object color, B j represents the B value of the building object in the jth point cloud, B max Indicates the maximum B value of the building object color, B min Indicates the minimum B value of the building object color.

[0027] Preferably, the mathematical model of the rationality index of the building object labeling method is specifically:

[0028]

[0029] where α i represents the influence coefficient value of the 3D building object size data in the i-th time period, α i-1 represents the influence coefficient value of the 3D building object size data in the i-1th time period, β i represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-th time period, β i-1 represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-1th time period, γ i Represents the color data coefficient value of the three-dimensional building object material in the i-th time period, γ i-1 Represents the color data coefficient value of the three-dimensional building object material in the i-1th time period.

[0030] Technical effects and advantages of the present invention:

[0031] 1. The present invention can obtain high-precision point cloud data containing multi-dimensional information such as the three-dimensional coordinates of the object, laser reflection intensity, color, etc. through laser scanning and other technologies, which can more realistically reflect the actual situation of the building object and improve accuracy;

[0032] 2. The present invention can directly mark building objects in a visualized three-dimensional space, and observe and mark them from different perspectives through operations such as rotation and scaling. Some systems have automatic marking or auxiliary marking functions, which can greatly improve marking efficiency;

[0033] 3. The present invention collects building object data in real time and can directly modify and update the original point cloud data, so as to conveniently and quickly reflect the changes of the building object and ensure the timeliness and accuracy of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] As attached Figure 1 A method for marking building objects in three-dimensional point cloud data is shown, including S1, marking area determination, S2, data collection, S3, three-dimensional point cloud data processing, S4, three-dimensional point cloud data analysis, S5, building judgment, S6, marking execution, S7, data interactive transmission.

[0037] S1. Determine the marked area, which is used to determine the data in the target city area as the target marked area, and divide it into sub-areas according to the equal time division method, and number the sub-areas in the target marked area as 1, 2, ..., i, ..., n in sequence.

[0038] S2. Data collection: Use a 3D scanner to scan the target urban area, obtain 3D point cloud data containing building objects, and transmit it to the 3D point cloud data processing step, and number them as 1, 2, ..., j, ..., m in sequence according to the number of point clouds.

[0039] In this embodiment, it should be specifically explained that the three-dimensional point cloud data includes three-dimensional building object size data influencing parameters, three-dimensional building object space shape data influencing parameters and three-dimensional building object material color data influencing parameters.

[0040] In this embodiment, it should be specifically explained that the influencing parameters of the three-dimensional building object size data include the number of building object point cloud points, denoted as P; the building object volume, denoted as V; the building object surface area, denoted as S; the building object length, denoted as L; the building object width, denoted as W; the building object height, denoted as H; the influencing parameters of the three-dimensional building object spatial shape data include the curvature of the building component, denoted as μ; the flatness of the building object wall, denoted as F; the building object point density, denoted as ρ; the building object point spacing, denoted as d; the influencing parameters of the three-dimensional building object material color data include the building object reflectivity, denoted as r; the building object color RGB values, denoted as R, G, and B respectively; and the building object texture density, denoted as t.

[0041] In this embodiment, it is specifically necessary to explain that the data acquisition step obtains the number of building object point cloud points by counting the number of points in the three-dimensional point cloud data; determines the height of the building object by calculating the maximum difference in the point cloud in the vertical direction; obtains the length and width of the building object by finding the maximum distance between the boundary points of the building point cloud on the horizontal plane; obtains the flatness of the wall of the building object by calculating the average distance between the wall point cloud and the fitting plane; spatially divides the point cloud data by using a voxelization method, calculates the total volume of the voxels occupied by the building point cloud, and obtains the volume of the building object; and obtains the height of the building object by calculating the maximum distance between the building point cloud and the fitting plane. The curvature value of the building object is obtained by the second-order derivative; the surface area of ​​the building is obtained by triangulating the point cloud of the building's outer surface and then calculating the sum of the areas of all triangles; the point density of the building object is obtained by calculating the density of each point in the selected area; the point spacing of the building object is obtained by measuring the distance between adjacent points in the point cloud; the point cloud is obtained by scanning and recording the reflection intensity of each point, and then the reflectivity of the building object is calculated using the reflection intensity information in the point cloud data; the RGB value of the building object color is obtained through the color information collected from the point cloud data; the texture density of the building object is obtained by calculating the number of texture feature points per unit area.

[0042] S3, three-dimensional point cloud data processing, is used to analyze and process the three-dimensional point cloud data obtained in S2 to obtain the influence coefficient of the three-dimensional building object size data, the influence coefficient of the three-dimensional building object space shape data and the influence coefficient of the three-dimensional building object material color data, and transmit it to the three-dimensional point cloud data analysis step.

[0043] In this embodiment, it should be specifically explained that the three-dimensional point cloud data processing step includes a three-dimensional building object size data influence coefficient calculation step, a three-dimensional building object space shape data influence coefficient calculation step and a three-dimensional building object material color data coefficient calculation step.

[0044] In this embodiment, it is specifically necessary to explain that the three-dimensional building object size data influence coefficient calculation step is used to import the three-dimensional building object size data influence parameters into the three-dimensional building object size data influence coefficient mathematical model to obtain the three-dimensional building object size data influence coefficient value; the three-dimensional building object space shape data influence coefficient calculation step is used to import the three-dimensional building object space shape data influence parameters into the three-dimensional building object space shape data influence coefficient mathematical model to obtain the three-dimensional building object space shape data influence coefficient value; the three-dimensional building object material color data coefficient calculation step is used to import the three-dimensional building object material color data parameters into the three-dimensional building object material color data coefficient mathematical model to obtain the three-dimensional building object material color data coefficient value.

[0045] In this embodiment, it should be specifically explained that the mathematical model of the influence coefficient of the three-dimensional building object size data is specifically:

[0046]

[0047] The mathematical model of the influence coefficient of the spatial shape data of three-dimensional building objects is as follows:

[0048]

[0049] The mathematical model of the color data coefficient of three-dimensional building objects is as follows:

[0050]

[0051] where ρ j represents the building object point density of the j-th point cloud, represents the average building object point density; Represents the average building object point spacing, μ max Represents the maximum curvature of the building component, μ min Represents the minimum curvature of the building component, R j represents the R value of the building object in the jth point cloud, R max Indicates the maximum R value of the building object color, R min Indicates the minimum R value of the building object color, G j represents the G value of the building object in the jth point cloud, G max Indicates the maximum G value of the building object color, G min Indicates the minimum G value of the building object color, B j represents the B value of the building object in the jth point cloud, B max Indicates the maximum B value of the building object color, B min Indicates the minimum B value of the building object color.

[0052] S4, 3D point cloud data analysis, importing the influence coefficient value of the 3D building object size data, the influence coefficient value of the 3D building object space shape data and the influence coefficient value of the 3D building object material color data obtained in S3 into the mathematical model of the rationality index of the building object marking method, obtaining the rationality index value of the building object marking method, and transmitting it to the building judgment step.

[0053] In this embodiment, it should be specifically explained that the mathematical model of the rationality index of the building object labeling method is specifically:

[0054]

[0055] where α i represents the influence coefficient value of the 3D building object size data in the i-th time period, α i-1 represents the influence coefficient value of the 3D building object size data in the i-1th time period, β i represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-th time period, β i-1 represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-1th time period, γ i Represents the color data coefficient value of the three-dimensional building object material in the i-th time period, γ i-1 Represents the color data coefficient value of the three-dimensional building object material in the i-1th time period.

[0056] S5. Building judgment: compare the building object labeling method rationality index value obtained in S4 with the preset building object labeling method rationality index values ​​under multiple standards, calculate the minimum value of the difference between the building object labeling method rationality index value and the preset building object labeling method rationality index under multiple standards, and when the minimum value of the difference value is less than the preset difference value, transmit the judgment result and the data in the target labeling area to the labeling execution step and the data interactive transmission step; when the minimum value of the difference value is greater than the preset difference value, transmit the data in the target labeling area as historical data to the data collection step.

[0057] S6, labeling execution, automatically assigning labeling labels to the building objects according to the judgment results obtained in S5 and the data in the target labeling area, presenting the labeling information, and then transmitting it to the data interaction transmission step.

[0058] S7 data interaction transmission is used to transmit the obtained judgment results, data in the target computing server and annotation information to the user information terminal, providing reference data for users to make adjustment measures.

[0059] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0060] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for marking building objects in three-dimensional point cloud data, characterized in that: include: S1, determining the marked area, for determining the data in the target city area as the target marked area, and dividing it into sub-areas according to the equal time division method, and numbering each sub-area in the target marked area as 1, 2, ..., i, ..., n in sequence; S2, data acquisition: use a 3D scanner to scan the target urban area, obtain 3D point cloud data containing building objects, and transmit it to the 3D point cloud data processing step, and number them in sequence as 1, 2, ..., j, ..., m according to the number of point clouds; S3, 3D point cloud data processing, for analyzing and processing the 3D point cloud data obtained in S2 to obtain the influence coefficient of the 3D building object size data, the influence coefficient of the 3D building object space shape data and the influence coefficient of the 3D building object material color data, and transmit them to the 3D point cloud data analysis step; S4, 3D point cloud data analysis, importing the influence coefficient value of the 3D building object size data, the influence coefficient value of the 3D building object space shape data and the coefficient value of the 3D building object material color data obtained in S3 into the building object labeling method rationality index mathematical model, obtaining the building object labeling method rationality index value, and transmitting it to the building judgment step; S5, building judgment, comparing the building object labeling method rationality index value obtained in S4 with the preset building object labeling method rationality index values ​​under multiple standards, calculating the minimum value of the difference between the building object labeling method rationality index value and the preset building object labeling method rationality index under multiple standards, when the minimum value of the difference value is less than the preset difference value, transmitting the judgment result and the data in the target labeling area to the labeling execution step and the data interactive transmission step, when the minimum value of the difference value is greater than the preset difference value, transmitting the data in the target labeling area as historical data to the data collection step; S6, labeling execution, automatically assigning labeling labels to the building objects according to the judgment result obtained in S5 and the data in the target labeling area, presenting the labeling information, and then transmitting it to the data interaction transmission step; S7 data interaction transmission is used to transmit the obtained judgment results, data in the target computing server and annotation information to the user information terminal, providing reference data for users to make adjustment measures.

2. The method for marking building objects in three-dimensional point cloud data according to claim 1, characterized in that: The three-dimensional point cloud data includes three-dimensional building object size data influencing parameters, three-dimensional building object space shape data influencing parameters and three-dimensional building object material color data influencing parameters.

3. The method for marking building objects in three-dimensional point cloud data according to claim 2, characterized in that: The three-dimensional building object size data influencing parameters include the number of building object point cloud points, denoted as P; the building object volume, denoted as V; the building surface area, denoted as S; The length of the building object is recorded as L; the width of the building object is recorded as W; The height of the building object is denoted as H; the parameters affecting the spatial shape data of the three-dimensional building object include the curvature of the building components, denoted as μ; The flatness of the wall of the building object is recorded as F; The density of building object points, denoted as ρ; The distance between building object points is denoted as d; The parameters affecting the color data of three-dimensional building objects include the reflectivity of the building objects, denoted as r; the RGB values ​​of the building object colors, denoted as R, G, and B respectively; The texture density of building objects is denoted by t.

4. The method for marking building objects in three-dimensional point cloud data according to claim 1, characterized in that: The three-dimensional point cloud data processing step includes a three-dimensional building object size data influence coefficient calculation step, a three-dimensional building object space shape data influence coefficient calculation step and a three-dimensional building object material color data coefficient calculation step.

5. The method for marking building objects in three-dimensional point cloud data according to claim 4, characterized in that: The three-dimensional building object size data influence coefficient calculation step is used to import the three-dimensional building object size data influence parameter into the three-dimensional building object size data influence coefficient mathematical model to obtain the three-dimensional building object size data influence coefficient value; The step of calculating the influence coefficient of the spatial shape data of the three-dimensional building object is used to import the influence parameters of the spatial shape data of the three-dimensional building object into the mathematical model of the influence coefficient of the spatial shape data of the three-dimensional building object to obtain the influence coefficient value of the spatial shape data of the three-dimensional building object; The three-dimensional building object material color data coefficient calculation step is used to import the three-dimensional building object material color data parameters into the three-dimensional building object material color data coefficient mathematical model to obtain the three-dimensional building object material color data coefficient value.

6. The method for marking building objects in three-dimensional point cloud data according to claim 5, characterized in that: The mathematical model of the influence coefficient of the three-dimensional building object size data is specifically:

7. The method for marking building objects in three-dimensional point cloud data according to claim 5, characterized in that: The mathematical model of the influence coefficient of the spatial shape data of the three-dimensional building object is specifically: where ρ j represents the building object point density of the j-th point cloud, represents the average building object point density; Represents the average building object point spacing, μ max Represents the maximum curvature of the building component, μ min Indicates the minimum curvature of building components.

8. The method for marking building objects in three-dimensional point cloud data according to claim 5, characterized in that: The mathematical model of the color data coefficient of the three-dimensional building object material is specifically: Where R j represents the R value of the building object in the jth point cloud, R max Indicates the maximum R value of the building object color, R min Indicates the minimum R value of the building object color, G j represents the G value of the building object in the jth point cloud, G max Indicates the maximum G value of the building object color, G min Indicates the minimum G value of the building object color, B j represents the B value of the building object in the jth point cloud, B max Indicates the maximum B value of the building object color, B min Indicates the minimum B value of the building object color.

9. The method for marking building objects in three-dimensional point cloud data according to claim 1, characterized in that: The mathematical model of the rationality index of the building object marking method is specifically: where α i represents the influence coefficient value of the 3D building object size data in the i-th time period, α i-1 represents the influence coefficient value of the 3D building object size data in the i-1th time period, β i represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-th time period, β i-1 represents the influence coefficient value of the spatial shape data of the three-dimensional building object in the i-1th time period, γ i Represents the color data coefficient value of the three-dimensional building object material in the i-th time period, γ i-1 Represents the color data coefficient value of the three-dimensional building object material in the i-1th time period.

Images