CN117928385A - Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor - Google Patents
Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor Download PDFInfo
- Publication number
- CN117928385A CN117928385A CN202410310748.5A CN202410310748A CN117928385A CN 117928385 A CN117928385 A CN 117928385A CN 202410310748 A CN202410310748 A CN 202410310748A CN 117928385 A CN117928385 A CN 117928385A
- Authority
- CN
- China
- Prior art keywords
- building
- point cloud
- cloud data
- construction
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 168
- 238000000691 measurement method Methods 0.000 title claims abstract description 16
- 238000013461 design Methods 0.000 claims abstract description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 60
- 239000011159 matrix material Substances 0.000 claims description 44
- 238000000034 method Methods 0.000 claims description 38
- 239000013598 vector Substances 0.000 claims description 36
- 238000005457 optimization Methods 0.000 claims description 15
- 230000009466 transformation Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 23
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 15
- 238000012545 processing Methods 0.000 description 15
- 230000008569 process Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012067 mathematical method Methods 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor, which relates to the technical field of construction measurement, and comprises the following steps: scanning an engineering construction area by using an unmanned aerial vehicle-mounted laser scanner; building information is extracted based on a construction scheme, building areas in point cloud data of engineering construction areas are divided according to the building information, and building point cloud data are extracted; performing a geometric dimension measurement of a building within the building area based on the building point cloud data; and comparing the measured geometric dimension data with design data in the construction scheme, and evaluating the construction precision. The invention can comprehensively understand the information of the form, the size and the like of the building by measuring the geometric dimension based on the building point cloud data, is beneficial to planning and designing engineering construction schemes, is beneficial to timely adjusting and optimizing the construction schemes, and improves the construction precision and efficiency.
Description
Technical Field
The invention relates to the technical field of construction measurement, in particular to an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor.
Background
Engineering measurements play a critical role in managing engineering projects in construction enterprises. The method is not only a key for grasping the progress of the construction site, but also an important means for ensuring the successful completion of engineering projects and preventing construction accidents. In recent years, as the scale and content of engineering projects are continuously expanded, construction site environments and construction procedures become more complex, which brings new challenges and difficulties to construction measurement.
In the face of these challenges, conventional construction measurement methods require continuous innovation and optimization of their technical approaches. First, it is essential to introduce advanced measurement techniques. For example, engineering measurement is performed by using a remote unmanned aerial vehicle and a sensor, so that high-precision and omnibearing monitoring on a construction site can be realized, and the problems in construction can be found and solved in time. Second, data processing and analysis are also key elements. The application of technologies such as big data, artificial intelligence, machine learning and the like can help construction enterprises to process and analyze massive construction data more effectively, discover potential risks in the construction process, and early warning and adjustment are carried out in advance so as to ensure the smooth progress of construction projects.
In the measuring process of engineering construction areas, a laser scanner is currently commonly used for omnibearing and high-precision measurement. Such measurement techniques can provide detailed and accurate point cloud data, which is critical to understanding and recording complex terrain and building structures at a job site. However, due to the complexity of the construction area, laser scanners tend to produce large amounts of point cloud data. These data contain not only critical information, but also a large number of redundant points, which can lead to significant increases in the difficulty of data processing and analysis, thereby reducing overall data processing efficiency.
For the problems in the related art, no effective solution has been proposed at present.
Disclosure of Invention
In view of the above, the present invention provides an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor to solve the above-mentioned problems that due to the complexity of the construction area, a laser scanner often generates a large amount of point cloud data, which can cause significant increase in difficulty of data processing and analysis and decrease in overall data processing efficiency.
In order to solve the problems, the invention adopts the following specific technical scheme:
the engineering construction intelligent measurement method based on the remote unmanned aerial vehicle and the sensor comprises the following steps:
s1, scanning an engineering construction area by using an unmanned aerial vehicle-mounted laser scanner to obtain point cloud data;
S2, building information is extracted based on a construction scheme, building areas in point cloud data of an engineering construction area are divided according to the building information, and building point cloud data are extracted;
S3, measuring the geometric dimension of the building in the building area based on the building point cloud data;
s4, comparing the measured geometric dimension data with design data in a construction scheme, and evaluating construction accuracy.
Preferably, the extracting building information based on the construction scheme divides building areas in the point cloud data of the engineering construction area according to the building information, and extracting building point cloud data includes the following steps:
S21, acquiring a construction scheme of an engineering construction area, wherein the construction scheme comprises a construction drawing and a construction layout;
s22, carrying out coordinate alignment on the point cloud data and the construction drawing by utilizing a registration algorithm, and dividing building areas in the point cloud data according to the construction layout;
S23, classifying and dividing the point cloud data of the building areas by a building identification algorithm for each building area, and extracting building point cloud data.
Preferably, the coordinate alignment of the point cloud data and the construction drawing by using the registration algorithm, and the partitioning of the building area in the point cloud data according to the construction layout includes the following steps:
S221, respectively extracting ground characteristic points from the point cloud data and the construction drawing;
s222, matching the point cloud data with ground characteristic points in a construction drawing through a characteristic matching algorithm;
s223, adjusting the position and the direction of the point cloud data based on the matching result so that the point cloud data are aligned with the coordinates of the construction drawing;
and S224, identifying and dividing the building area based on the construction layout according to the aligned point cloud data.
Preferably, the matching of the point cloud data with the ground feature points in the construction drawing by the feature matching algorithm comprises the following steps:
S2221, calculating normal vectors of each feature point for the point cloud data and the ground feature points in the construction drawing, searching feature points in a preset range from the ground feature points by using a KD tree, and storing the normal vectors;
S2222, dividing the ground characteristic points in the ground characteristic point preset range into a plurality of three-dimensional subspaces, and sorting according to the distances between the three-dimensional subspaces and the ground characteristic points;
s2223, analyzing the points in each three-dimensional subspace, calculating normal angles between the points and the ground characteristic points, and mapping the angles into a preset coordinate system;
s2224, counting the number of points falling into each section of the coordinate system, generating a dimension vector for each subspace, and combining the vectors of all subspaces into a feature descriptor of a high dimension vector;
S2225, matching the ground characteristic points in the point cloud data with the ground characteristic points in the construction drawing by utilizing Euclidean distances among the characteristic descriptors.
Preferably, adjusting the position and direction of the point cloud data based on the matching result so that the point cloud data is aligned with the coordinates of the construction drawing includes the following steps:
S2231, selecting a matching point pair with Euclidean distance meeting a preset threshold value from a matching result;
s2232, estimating a transformation matrix based on the selected matching point pairs, wherein the transformation matrix comprises rotation, translation and scaling information;
S2233, optimizing a transformation matrix through an iterative nearest point algorithm to minimize errors between the point cloud data and the construction drawing;
and S2234, when the preset maximum iteration times are reached, obtaining an optimal change matrix, and adjusting the position and the direction of the point cloud data based on the optimal change matrix so as to align the coordinates of the point cloud data and the construction drawing.
Preferably, for each building area, the classifying and dividing the point cloud data of the building area by a building identification algorithm, and extracting the building point cloud data includes the following steps:
s231, calculating the similarity between points in the point cloud data of each building area according to a point cloud similarity function, and constructing a similarity matrix based on a similarity result;
S232, adding each row of elements in the similarity matrix to obtain a secondary matrix, and subtracting the similarity matrix from the secondary matrix to obtain a tertiary matrix;
S233, calculating the first K minimum eigenvalues and the corresponding eigenvectors in the three-level matrix;
S234, clustering the obtained feature vectors through a clustering algorithm, and identifying building point cloud data from the point cloud data of the building area.
Preferably, the calculation formula for calculating the similarity between points in the point cloud data of the building area through the point cloud similarity function is as follows:
;
Wherein G ij represents the similarity between the i-th point x i and the j-th point x j in the point cloud data of the building area;
II x i-xj‖2 represents the Manhattan distance between the i-th point x i and the j-th point x j;
Beta represents a gaussian kernel parameter;
exp represents an exponential function;
n represents the number of points in the point cloud data.
Preferably, the geometric dimension measurement of the building in the building area based on the building point cloud data comprises the following steps:
s31, optimizing the building point cloud data through an optimization algorithm based on the building point cloud data;
s32, extracting the outline of the building through an outline extraction algorithm based on the optimized building point cloud data;
s33, calculating the geometric feature size of the building by using the extracted building outline.
Preferably, the optimizing the building point cloud data through the optimization algorithm based on the building point cloud data comprises the following steps:
s311, calculating the curvature value of each point on the surface of the building based on the building point cloud data;
S312, identifying uniform areas and non-uniform areas of the building point cloud according to the calculated curvature value;
S313, optimizing point cloud data of a uniform area of the building point cloud by using a random uniform sampling method;
S314, optimizing the point cloud data of the non-uniform area of the building point cloud by utilizing a feature detection algorithm.
Preferably, the calculating the geometric feature size of the building using the extracted contour of the building includes the steps of:
S331, performing smoothing on the extracted outline of the building, and identifying geometric features of the building;
s332, building a geometric model of the building based on each identified geometric feature;
s333, calculating the geometric feature size of the building by utilizing a trigonometric function and a geometric formula based on the geometric model of the building.
Compared with the prior art, the invention provides an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor, which has the following beneficial effects:
(1) According to the invention, the unmanned aerial vehicle is provided with the laser scanner, so that the point cloud data of an engineering construction area can be quickly and efficiently acquired, the time and labor cost are greatly saved, the laser scanner can provide high-precision point cloud data, therefore, the measurement result is more accurate and reliable, the construction quality is ensured, the geometric dimension measurement based on the building point cloud data can comprehensively know the information such as the form and the dimension of a building, the planning and the design of an engineering construction scheme are facilitated, the measured geometric dimension data is compared with the design data in the construction scheme, the deviation and the problem in the construction process can be found in time, the timely adjustment and optimization of the construction scheme are facilitated, and the construction precision and the construction efficiency are improved.
(2) According to the invention, the point cloud data can be accurately aligned with the construction drawing through the feature matching algorithm and the registration algorithm, the accuracy of subsequent analysis is ensured, the point cloud data of the building area can be rapidly and accurately extracted through the building identification algorithm, the identification efficiency is improved, the feature information of the building can be effectively extracted from the point cloud data through the point cloud similarity function and the clustering algorithm, the utilization of the data is more comprehensive and deeper, the accurate segmentation and identification of the building area can be realized through the similarity matrix and the feature vector processing, the subsequent building measurement and evaluation work is facilitated, and the efficiency and the accuracy in the construction process are improved.
(3) According to the invention, the quality and accuracy of the building point cloud data can be improved by processing the building point cloud data through an optimization algorithm, a more reliable basis is provided for the subsequent geometric dimension measurement, the curvature value of each point on the surface of the building is calculated, the uniform area and the non-uniform area are identified, different optimization methods can be adopted for areas with different characteristics, the pertinence and the efficiency of the data processing are improved, the contour of the building can be accurately extracted by utilizing a contour extraction algorithm, reliable basis data is provided for the subsequent geometric feature dimension calculation, the geometric model of the building can be established by carrying out smoothing treatment and geometric feature identification on the extracted contour of the building, the geometric feature dimension of the building is accurately calculated by utilizing a mathematical method, the geometric dimension measurement of the building is more accurate and reliable, the construction precision and quality are evaluated, and the engineering measurement reliability is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
Fig. 1 is a flow chart of an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor according to an embodiment of the present invention.
Detailed Description
In order to make the technical solution of the present application better understood by those skilled in the art, the technical solution of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments of the present application. All other embodiments, based on the embodiments of the application, which would be apparent to one of ordinary skill in the art without undue burden are intended to be within the scope of the application.
According to the embodiment of the invention, an engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor is provided.
The invention will be further described with reference to the accompanying drawings and specific embodiments, as shown in fig. 1, according to an embodiment of the invention, the engineering construction intelligent measurement method based on a remote unmanned aerial vehicle and a sensor includes the following steps:
s1, scanning an engineering construction area by using an unmanned aerial vehicle-mounted laser scanner to obtain point cloud data;
Specifically, before scanning, the flight path and the height of the unmanned aerial vehicle need to be planned according to the size and complexity of the construction area, so that the flight path can cover the whole construction area, and shielding and overlapping areas are avoided. The unmanned aerial vehicle flies along a preset route, and the laser scanner starts to work to scan the ground. During scanning, the laser scanner may generate a large amount of point cloud data.
S2, building information is extracted based on a construction scheme, building areas in point cloud data of an engineering construction area are divided according to the building information, and building point cloud data are extracted;
As a preferred embodiment, the extracting building information based on the construction scheme, dividing building areas in the point cloud data of the engineering construction area according to the building information, and extracting building point cloud data includes the steps of:
S21, acquiring a construction scheme of an engineering construction area, wherein the construction scheme comprises a construction drawing and a construction layout;
The construction drawing includes the design intention, the structural layout, the dimension specification and the building material requirement of the building, and the construction layout includes the building position, the construction road, the material storage area, the equipment placement and the like.
S22, carrying out coordinate alignment on the point cloud data and the construction drawing by utilizing a registration algorithm, and dividing building areas in the point cloud data according to the construction layout;
as a preferred embodiment, the coordinate alignment of the point cloud data and the construction drawing by using the registration algorithm, and the partitioning of the building area in the point cloud data according to the construction layout includes the following steps:
S221, respectively extracting ground characteristic points from the point cloud data and the construction drawing;
It should be noted that, with the point cloud processing software, a ground extraction algorithm, such as a method based on a normal vector or a method based on RANSAC, may be used to extract ground points from the point cloud data, and for the extracted point cloud data, a normal vector of each point may be calculated, and screening of ground feature points may be performed according to the direction of the normal vector and the density of the points.
The construction drawing is a planar two-dimensional drawing, typically in the form of a CAD file, which is converted into a format that can be processed by the analysis software to identify ground standard symbols and marks in the drawing, which typically represent specific ground features.
S222, matching the point cloud data with ground characteristic points in a construction drawing through a characteristic matching algorithm;
as a preferred embodiment, the matching of the point cloud data with the ground feature points in the construction drawing by the feature matching algorithm includes the following steps:
S2221, calculating normal vectors of each feature point for the point cloud data and the ground feature points in the construction drawing, searching feature points in a preset range from the ground feature points by using a KD tree, and storing the normal vectors;
it should be noted that, a KD-tree (k-dimensional tree) is a data structure for organizing points in k-dimensional space, and is very suitable for fast search operations.
Specifically, for each feature point, points within its neighborhood are found (e.g., using a radius search or nearest neighbor search) and a local surface is fitted, and the normal vector of the surface is calculated based on the fitted local surface. Typically, this can be obtained by computing a covariance matrix of the local area of the point cloud and extracting principal components therefrom;
constructing a KD tree for the point cloud data, and for each ground characteristic point, quickly finding all adjacent characteristic points within a preset range by using the KD tree;
And extracting and storing normal vector information of each searched adjacent feature point.
S2222, dividing the ground characteristic points in the ground characteristic point preset range into a plurality of three-dimensional subspaces, and sorting according to the distances between the three-dimensional subspaces and the ground characteristic points;
it should be noted that, in the preset range of each feature point, the space is divided into a plurality of small three-dimensional subspaces by constructing a three-dimensional grid, the ground feature points in the range are distributed into the corresponding three-dimensional subspaces according to the coordinates thereof, and the distances between the ground feature points and the central ground feature point are calculated for the points in each three-dimensional subspace. The points in each subspace are then ordered according to the calculated distance.
S2223, analyzing the points in each three-dimensional subspace, calculating normal angles between the points and the ground characteristic points, and mapping the angles into a preset coordinate system;
It should be noted that, by defining a preset coordinate system, which may be a two-dimensional or three-dimensional coordinate system, the coordinate position of each point and the corresponding included angle value are calculated and mapped into the defined coordinate system.
S2224, counting the number of points falling into each section of the coordinate system, generating a dimension vector for each subspace, and combining the vectors of all subspaces into a feature descriptor of a high dimension vector;
It should be noted that, a dimension vector is defined for each subspace, each element of the vector represents the point number in the corresponding section, each dimension vector is filled according to the counted point number, if no point exists in one section, the vector element corresponding to the section is zero, the dimension vectors of all subspaces are sequentially combined into a single high-dimension vector, and the high-dimension vector is the feature descriptor of the whole point cloud data.
S2225, matching the ground characteristic points in the point cloud data with the ground characteristic points in the construction drawing by utilizing Euclidean distances among the characteristic descriptors.
It should be noted that, matching the ground feature points in the point cloud data and the ground feature points in the construction drawing by using the euclidean distance between the feature descriptors includes the following steps:
Extracting ground characteristic points from the point cloud data and the construction drawing, and generating a characteristic descriptor for each ground characteristic point, wherein the characteristic descriptor is an abstract representation of a local structure around each characteristic point, and is usually a vector;
For each ground characteristic point in the point cloud data, calculating Euclidean distances between the characteristic descriptors of the point cloud data and the characteristic descriptors of all the ground characteristic points in the construction drawing;
and setting a threshold value, wherein the threshold value is used for determining whether the ground characteristic points in the point cloud data are matched with the ground characteristic points in the construction drawing. If the Euclidean distance between two feature descriptors is less than the set threshold, then they are considered to be successfully matched;
And (3) carrying out matching screening on the ground characteristic points in the point cloud data according to the calculated Euclidean distance and a set threshold value, and considering that the matching is successful only when the distance between the ground characteristic points and a certain ground characteristic point in a construction drawing is smaller than the threshold value.
S223, adjusting the position and the direction of the point cloud data based on the matching result so that the point cloud data are aligned with the coordinates of the construction drawing;
as a preferred embodiment, the adjusting the position and the direction of the point cloud data based on the matching result, so that the point cloud data is aligned with the coordinates of the construction drawing includes the following steps:
S2231, selecting a matching point pair with Euclidean distance meeting a preset threshold value from a matching result;
s2232, estimating a transformation matrix based on the selected matching point pairs, wherein the transformation matrix comprises rotation, translation and scaling information;
it should be noted that, a certain number of feature points are selected from the successfully matched point pairs, and the points correspond to the point cloud data and the ground feature points in the construction drawing; the selected matching point pairs are constructed into corresponding relations, and the corresponding relations can be used for deriving a transformation matrix.
S2233, optimizing a transformation matrix through an iterative nearest point algorithm to minimize errors between the point cloud data and the construction drawing;
It should be noted that, in the initial stage, a least square method or other optimization method may be used to estimate an initial transformation matrix, where the initial matrix may not be optimal, but is the starting point of the subsequent optimization process; the transformation matrix is continuously adjusted by an iterative algorithm (such as an iterative nearest point algorithm) or other optimization methods, so that errors between the point cloud data and the construction drawing are minimized.
And S2234, when the preset maximum iteration times are reached, obtaining an optimal change matrix, and adjusting the position and the direction of the point cloud data based on the optimal change matrix so as to align the coordinates of the point cloud data and the construction drawing.
And S224, identifying and dividing the building area based on the construction layout according to the aligned point cloud data.
The construction layout is analyzed firstly, including information such as building position, construction road, material storage area, equipment placement, etc., and according to the characteristics of the construction layout, a rule and algorithm for identifying building areas are established, and the building areas in the point cloud data are identified by utilizing the analysis of the construction layout and the established identification rule.
S23, classifying and dividing the point cloud data of the building areas by a building identification algorithm for each building area, and extracting building point cloud data.
As a preferred embodiment, for each building area, classifying and dividing the point cloud data of the building area by a building identification algorithm, and extracting the building point cloud data includes the following steps:
s231, calculating the similarity between points in the point cloud data of each building area according to a point cloud similarity function, and constructing a similarity matrix based on a similarity result;
It should be noted that, the main idea of the point cloud similarity function is to measure the similarity between points in the point cloud data by using a gaussian function. The gaussian function is widely applied in the fields of pattern recognition and data analysis, and can effectively describe the relationship between data due to the characteristics of smoothness and continuity, and has the characteristics that the similarity is higher when the distance between two points is smaller, and the similarity is rapidly reduced when the distance is increased, so that a smooth attenuation trend is presented. This feature enables the gaussian function to describe well the local similarity between points in the point cloud data, without being affected by local noise.
Specifically, the calculation formula for calculating the similarity between points in the point cloud data of the building area through the point cloud similarity function is as follows:
;
Wherein G ij represents the similarity between the i-th point x i and the j-th point x j in the point cloud data of the building area;
II x i-xj‖2 represents the Manhattan distance between the i-th point x i and the j-th point x j;
Beta represents a gaussian kernel parameter;
exp represents an exponential function;
n represents the number of points in the point cloud data.
S232, adding each row of elements in the similarity matrix to obtain a secondary matrix, and subtracting the similarity matrix from the secondary matrix to obtain a tertiary matrix;
S233, calculating the first K minimum eigenvalues and the corresponding eigenvectors in the three-level matrix;
It should be noted that, for a given three-level matrix, a singular value decomposition algorithm is used to calculate the eigenvalues and the corresponding eigenvectors, the eigenvalues obtained by calculation are arranged in ascending order so as to find the first K minimum eigenvalues, the first K minimum eigenvalues are selected from the ordered eigenvalue list, and the eigenvectors corresponding to the first K minimum eigenvalues are obtained.
S234, clustering the obtained feature vectors through a clustering algorithm, and identifying building point cloud data from the point cloud data of the building area.
Specifically, the clustering algorithm is used for clustering the obtained feature vectors, and the identification of building point cloud data from the point cloud data of the building area comprises the following steps:
Determining the number of clusters, and carrying out normalization processing on the feature vectors to ensure that the scale of each feature is similar;
Clustering the feature vectors by using a K-means clustering algorithm, and once clustering is completed, identifying building point cloud data in the point cloud data of the building area according to a clustering result; each cluster represents a category or a building area.
S3, measuring the geometric dimension of the building in the building area based on the building point cloud data;
As a preferred embodiment, the geometric dimension measurement of the building in the building area based on the building point cloud data comprises the steps of:
s31, optimizing the building point cloud data through an optimization algorithm based on the building point cloud data;
As a preferred embodiment, the optimizing the building point cloud data by an optimization algorithm based on the building point cloud data includes the following steps:
s311, calculating the curvature value of each point on the surface of the building based on the building point cloud data;
specifically, calculating a curvature value of each point of a building surface based on building point cloud data includes the steps of:
for each point, defining a neighborhood around it, which may be a sphere or cube of fixed size centered on the point;
For each neighborhood, using least squares fitting and method to approximate description point cloud data;
Calculating the main curvature and the main curvature direction of the curved surface obtained by fitting, wherein the main curvature is a characteristic value of the curved surface along different direction change rates at a certain point;
The product of the principal curvatures is the curvature value of the point; in general, a product of a gaussian curvature reflecting the degree of curvature of a curved surface and an average curvature representing the degree of curvature of a curved surface in a certain direction may be used as a curvature value.
S312, identifying uniform areas and non-uniform areas of the building point cloud according to the calculated curvature value;
Specifically, for each point in the building point cloud data, according to the calculated curvature value, the change condition of the curvature value in the neighborhood around the point cloud data is analyzed, and a curvature threshold value is set for judging the change degree of the curvature value. In general, if a curvature value of a certain region changes little, the region may belong to a uniform region; conversely, if the curvature value changes greatly, the region may belong to a non-uniform region.
S313, optimizing point cloud data of a uniform area of the building point cloud by using a random uniform sampling method;
Specifically, for point cloud data of a uniform area, a random uniform sampling method is adopted, a certain proportion of points are randomly selected as sampling points in the area, then the sampling points are produced, and the rest of points are used as optimized point cloud data.
S314, optimizing the point cloud data of the non-uniform area of the building point cloud by utilizing a feature detection algorithm.
Specifically, for a non-uniform area of a building point cloud, optimizing point cloud data using a feature detection algorithm includes the steps of:
Extracting characteristic points or characteristic areas from point cloud data of the non-uniform area by using a Harris angular point detection algorithm, wherein the characteristics possibly comprise points with higher curvature, edge points, corner points and the like;
For each extracted characteristic point, processing nearby point cloud data, which may involve operations such as point cloud smoothing, reconstruction, surface fitting and the like of a local area;
screening and filtering the extracted characteristic points to eliminate noise and unnecessary points;
And integrating the point cloud data subjected to feature detection and processing into an optimized point cloud data set.
S32, extracting the outline of the building through an outline extraction algorithm based on the optimized building point cloud data;
Specifically, based on the optimized building point cloud data, extracting the outline of the building through an outline extraction algorithm comprises the following steps:
In the building point cloud data, firstly, performing plane segmentation operation to separate a plane part from a non-plane part of a building;
Once the planar portion of the building is segmented, identifying the boundary of the building by a contour extraction algorithm; common contour extraction algorithms include convex hull-based methods, edge detection-based methods, segmentation-based methods, and the like;
After the contour extraction is completed, the contour of the building may be represented as a set of points, a set of line segments, or a closed polygon.
S33, calculating the geometric feature size of the building by using the extracted building outline.
As a preferred embodiment, the calculating the geometric feature size of the building using the extracted building contour includes the steps of:
S331, performing smoothing on the extracted outline of the building, and identifying geometric features of the building;
it should be noted that, by using gaussian filtering to perform the contour of the building, the processing helps to eliminate noise in the contour and make the contour more continuous and smooth, and for the smoothed contour, curve fitting techniques, such as bezier curves, spline curves, etc., may be applied to approximately describe the shape of the contour, and geometric feature extraction may be performed based on the smoothed contour. This involves calculating geometric features of the building, such as area, perimeter, form factor, convexity, etc.
S332, building a geometric model of the building based on each identified geometric feature;
It should be noted that, according to the geometric features extracted from the outline of the building, the features with the most representation and correlation are selected, and these features may include the area, perimeter, shape factor, convexity, etc. of the building, and according to the characteristics and application requirements of the building, a geometric model is built, for example, for a two-dimensional model, a simple polygon or curve may be used to describe the outline of the building, and for a three-dimensional model, a curved surface modeling technique or voxel-based method may be used to generate the geometric model of the building; the building geometry is converted into parameters of the model. The geometric features are converted into parameters of the model according to the modeling method selected so that the shape and size of the model can be dynamically adjusted.
S333, calculating the geometric feature size of the building by utilizing a trigonometric function and a geometric formula based on the geometric model of the building.
Specifically, for an area, the area of the building can be obtained by adding individual area elements of the geometric model. For a two-dimensional geometric model, the areas of the area elements can be directly calculated and added; for three-dimensional geometric models, it can be calculated by summing the surface areas.
For perimeter, if the building is closed, the perimeter can be calculated simply by the length of the boundary curve of the geometric model, for a two-dimensional geometric model, the length of the boundary curve can be measured directly; for three-dimensional geometric models, path integration can be performed over a curved surface to calculate.
For simple geometries such as cuboids or cubes, the height, width and length can be read directly from the geometric model.
Other geometric feature dimensions may also be calculated, such as radius of curvature of the building, average inclination angle of the surface, etc., depending on the particular building feature.
S4, comparing the measured geometric dimension data with design data in a construction scheme, and evaluating construction accuracy.
Specifically, the geometric dimension data obtained by measurement is compared with design data in a construction scheme, and the construction precision is evaluated, which comprises the following steps:
the measured geometry data and design data in the construction plan are collected and consolidated. Ensuring the accuracy and the integrity of data, including the measurement values of various geometric characteristics such as area, perimeter, length, width, height and the like;
And establishing a comparison standard according to design data in the construction scheme. These criteria may be the size, angle, position, etc. requirements specified in the design drawing;
Comparing the measured geometric dimension data with design data, comparing the measured value and the design value of each geometric feature item by item, and determining the difference and deviation between the measured value and the design value; differences and deviations between the measured data and the design data are analyzed. Determining if there are differences outside of the tolerance range and the effects these differences may have on engineering quality and function;
According to the comparison result, evaluating the construction precision and quality; if the difference between the measured data and the design data is within the allowable range and does not affect the safety, stability and functionality of the project, then the construction may be considered to meet the expected accuracy requirements; if a large difference or deviation is found between the measured data and the design data, the construction process needs to be adjusted and improved in time.
In summary, by means of the technical scheme, the unmanned aerial vehicle is provided with the laser scanner, so that the point cloud data of an engineering construction area can be quickly and efficiently obtained, time and labor cost are greatly saved, the laser scanner can provide high-precision point cloud data, therefore, a measurement result is more accurate and reliable, construction quality is guaranteed, geometric dimension measurement based on building point cloud data can be used for comprehensively knowing information such as the form and the dimension of a building, planning and designing engineering construction schemes are facilitated, deviation and problems in the construction process can be found timely, and timely adjustment and optimization of the construction scheme are facilitated, and construction precision and efficiency are improved. According to the invention, the point cloud data can be accurately aligned with the construction drawing through the feature matching algorithm and the registration algorithm, the accuracy of subsequent analysis is ensured, the point cloud data of the building area can be rapidly and accurately extracted through the building identification algorithm, the identification efficiency is improved, the feature information of the building can be effectively extracted from the point cloud data through the point cloud similarity function and the clustering algorithm, the utilization of the data is more comprehensive and deeper, the accurate segmentation and identification of the building area can be realized through the similarity matrix and the feature vector processing, the subsequent building measurement and evaluation work is facilitated, and the efficiency and the accuracy in the construction process are improved. According to the invention, the quality and accuracy of the building point cloud data can be improved by processing the building point cloud data through an optimization algorithm, a more reliable basis is provided for the subsequent geometric dimension measurement, the curvature value of each point on the surface of the building is calculated, the uniform area and the non-uniform area are identified, different optimization methods can be adopted for areas with different characteristics, the pertinence and the efficiency of the data processing are improved, the contour of the building can be accurately extracted by utilizing a contour extraction algorithm, reliable basis data is provided for the subsequent geometric feature dimension calculation, the geometric model of the building can be established by carrying out smoothing treatment and geometric feature identification on the extracted contour of the building, the geometric feature dimension of the building is accurately calculated by utilizing a mathematical method, the geometric dimension measurement of the building is more accurate and reliable, the construction precision and quality are evaluated, and the engineering measurement reliability is improved.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the invention, and is not meant to limit the scope of the invention, but to limit the invention to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (10)
1. The engineering construction intelligent measurement method based on the remote unmanned aerial vehicle and the sensor is characterized by comprising the following steps of:
s1, scanning an engineering construction area by using an unmanned aerial vehicle-mounted laser scanner to obtain point cloud data;
S2, building information is extracted based on a construction scheme, building areas in point cloud data of an engineering construction area are divided according to the building information, and building point cloud data are extracted;
S3, measuring the geometric dimension of the building in the building area based on the building point cloud data;
s4, comparing the measured geometric dimension data with design data in a construction scheme, and evaluating construction accuracy.
2. The intelligent measuring method for engineering construction based on a remote unmanned aerial vehicle and a sensor according to claim 1, wherein the steps of extracting building information based on a construction scheme, dividing building areas in point cloud data of an engineering construction area according to the building information, and extracting building point cloud data comprise the following steps:
S21, acquiring a construction scheme of an engineering construction area, wherein the construction scheme comprises a construction drawing and a construction layout;
s22, carrying out coordinate alignment on the point cloud data and the construction drawing by utilizing a registration algorithm, and dividing building areas in the point cloud data according to the construction layout;
S23, classifying and dividing the point cloud data of the building areas by a building identification algorithm for each building area, and extracting building point cloud data.
3. The intelligent measuring method for engineering construction based on the remote unmanned aerial vehicle and the sensor according to claim 2, wherein the coordinate alignment of the point cloud data and the construction drawing by using the registration algorithm and the division of the building area in the point cloud data according to the construction layout comprise the following steps:
S221, respectively extracting ground characteristic points from the point cloud data and the construction drawing;
s222, matching the point cloud data with ground characteristic points in a construction drawing through a characteristic matching algorithm;
s223, adjusting the position and the direction of the point cloud data based on the matching result so that the point cloud data are aligned with the coordinates of the construction drawing;
and S224, identifying and dividing the building area based on the construction layout according to the aligned point cloud data.
4. The intelligent engineering construction measurement method based on the remote unmanned aerial vehicle and the sensor according to claim 3, wherein the matching of the point cloud data with the ground characteristic points in the construction drawing through the characteristic matching algorithm comprises the following steps:
S2221, calculating normal vectors of each feature point for the point cloud data and the ground feature points in the construction drawing, searching feature points in a preset range from the ground feature points by using a KD tree, and storing the normal vectors;
S2222, dividing the ground characteristic points in the ground characteristic point preset range into a plurality of three-dimensional subspaces, and sorting according to the distances between the three-dimensional subspaces and the ground characteristic points;
s2223, analyzing the points in each three-dimensional subspace, calculating normal angles between the points and the ground characteristic points, and mapping the angles into a preset coordinate system;
s2224, counting the number of points falling into each section of the coordinate system, generating a dimension vector for each subspace, and combining the vectors of all subspaces into a feature descriptor of a high dimension vector;
S2225, matching the ground characteristic points in the point cloud data with the ground characteristic points in the construction drawing by utilizing Euclidean distances among the characteristic descriptors.
5. The intelligent measuring method for engineering construction based on the remote unmanned aerial vehicle and the sensor according to claim 3, wherein the adjusting the position and the direction of the point cloud data based on the matching result so that the point cloud data are aligned with the coordinates of the construction drawing comprises the following steps:
S2231, selecting a matching point pair with Euclidean distance meeting a preset threshold value from a matching result;
s2232, estimating a transformation matrix based on the selected matching point pairs, wherein the transformation matrix comprises rotation, translation and scaling information;
S2233, optimizing a transformation matrix through an iterative nearest point algorithm to minimize errors between the point cloud data and the construction drawing;
and S2234, when the preset maximum iteration times are reached, obtaining an optimal change matrix, and adjusting the position and the direction of the point cloud data based on the optimal change matrix so as to align the coordinates of the point cloud data and the construction drawing.
6. The intelligent measuring method for engineering construction based on a remote unmanned aerial vehicle and a sensor according to claim 2, wherein for each building area, classifying and dividing the point cloud data of the building area by a building identification algorithm, and extracting the building point cloud data comprises the following steps:
s231, calculating the similarity between points in the point cloud data of each building area according to a point cloud similarity function, and constructing a similarity matrix based on a similarity result;
S232, adding each row of elements in the similarity matrix to obtain a secondary matrix, and subtracting the similarity matrix from the secondary matrix to obtain a tertiary matrix;
S233, calculating the first K minimum eigenvalues and the corresponding eigenvectors in the three-level matrix;
S234, clustering the obtained feature vectors through a clustering algorithm, and identifying building point cloud data from the point cloud data of the building area.
7. The intelligent measuring method for engineering construction based on the remote unmanned aerial vehicle and the sensor according to claim 6, wherein the calculation formula for calculating the similarity between points in the point cloud data of the building area through the point cloud similarity function is as follows:
;
Wherein G ij represents the similarity between the i-th point x i and the j-th point x j in the point cloud data of the building area;
II x i-xj‖2 represents the Manhattan distance between the i-th point x i and the j-th point x j;
Beta represents a gaussian kernel parameter;
exp represents an exponential function;
n represents the number of points in the point cloud data.
8. The intelligent measurement method for engineering construction based on remote unmanned aerial vehicle and sensor according to claim 1, wherein the measuring of the geometric dimensions of the building in the building area based on the building point cloud data comprises the following steps:
s31, optimizing the building point cloud data through an optimization algorithm based on the building point cloud data;
s32, extracting the outline of the building through an outline extraction algorithm based on the optimized building point cloud data;
s33, calculating the geometric feature size of the building by using the extracted building outline.
9. The intelligent measuring method for engineering construction based on the remote unmanned aerial vehicle and the sensor according to claim 8, wherein the optimizing the building point cloud data through the optimizing algorithm based on the building point cloud data comprises the following steps:
s311, calculating the curvature value of each point on the surface of the building based on the building point cloud data;
S312, identifying uniform areas and non-uniform areas of the building point cloud according to the calculated curvature value;
S313, optimizing point cloud data of a uniform area of the building point cloud by using a random uniform sampling method;
S314, optimizing the point cloud data of the non-uniform area of the building point cloud by utilizing a feature detection algorithm.
10. The intelligent measuring method for engineering construction based on remote unmanned aerial vehicle and sensor according to claim 8, wherein calculating the geometric feature size of the building by using the extracted outline of the building comprises the following steps:
S331, performing smoothing on the extracted outline of the building, and identifying geometric features of the building;
s332, building a geometric model of the building based on each identified geometric feature;
s333, calculating the geometric feature size of the building by utilizing a trigonometric function and a geometric formula based on the geometric model of the building.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410310748.5A CN117928385A (en) | 2024-03-19 | 2024-03-19 | Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410310748.5A CN117928385A (en) | 2024-03-19 | 2024-03-19 | Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117928385A true CN117928385A (en) | 2024-04-26 |
Family
ID=90766933
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410310748.5A Pending CN117928385A (en) | 2024-03-19 | 2024-03-19 | Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117928385A (en) |
-
2024
- 2024-03-19 CN CN202410310748.5A patent/CN117928385A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yang et al. | An individual tree segmentation method based on watershed algorithm and three-dimensional spatial distribution analysis from airborne LiDAR point clouds | |
Biosca et al. | Unsupervised robust planar segmentation of terrestrial laser scanner point clouds based on fuzzy clustering methods | |
Xu et al. | Classification of LiDAR point clouds using supervoxel-based detrended feature and perception-weighted graphical model | |
Zhang et al. | KDD: A kernel density based descriptor for 3D point clouds | |
CN110945568A (en) | System and method for evaluating 3D models | |
WO2021082380A1 (en) | Laser radar-based pallet recognition method and system, and electronic device | |
CN109871818B (en) | Face recognition method based on normal vector distribution histogram and covariance descriptor | |
Liu et al. | A novel rock-mass point cloud registration method based on feature line extraction and feature point matching | |
CN114648445B (en) | Multi-view high-resolution point cloud splicing method based on feature point extraction and fine registration optimization | |
CN110348478B (en) | Method for extracting trees in outdoor point cloud scene based on shape classification and combination | |
CN115830587A (en) | Structural surface rapid automatic identification method based on high-precision point cloud data | |
CN114898118A (en) | Automatic statistical method and system for power transmission line house removal amount based on multi-source point cloud | |
CN109359680B (en) | Explosion sillar automatic identification and lumpiness feature extracting method and device | |
CN113255677B (en) | Method, equipment and medium for rapidly extracting rock mass structural plane and occurrence information | |
CN110310322A (en) | Method for detecting assembly surface of 10-micron-level high-precision device | |
CN106980878B (en) | Method and device for determining geometric style of three-dimensional model | |
CN112270746A (en) | Aluminum alloy 3D printing point cloud simplification algorithm based on neighborhood covariance characteristic parameter threshold | |
CN117292181A (en) | Sheet metal part hole group classification and full-size measurement method based on 3D point cloud processing | |
CN117928385A (en) | Engineering construction intelligent measurement method based on remote unmanned aerial vehicle and sensor | |
Santos et al. | Classification of LiDAR data over building roofs using k-means and principal component analysis | |
CN115147471A (en) | Laser point cloud automatic registration method based on curvature density characteristics | |
CN110533781B (en) | Automatic labeling method for multi-class three-dimensional model components | |
Wu et al. | A Systematic Point Cloud Edge Detection Framework for Automatic Aircraft Skin Milling | |
CN117571341B (en) | System and method for detecting omnibearing wear of tire | |
Douadi et al. | Stable keypoints selection for 2D LiDAR based place recognition with map data reduction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication |