CN117392632B

CN117392632B - Road element change monitoring method and device

Info

Publication number: CN117392632B
Application number: CN202311687926.8A
Authority: CN
Inventors: 刘德强; 余顺新; 庄稼丰; 吴游宇; 余绍淮; 余飞; 罗博仁; 徐乔; 姚金玺
Original assignee: CCCC Second Highway Consultants Co Ltd
Current assignee: CCCC Second Highway Consultants Co Ltd
Priority date: 2023-12-11
Filing date: 2023-12-11
Publication date: 2024-03-15
Anticipated expiration: 2043-12-11
Also published as: CN117392632A

Abstract

The invention provides a road element change monitoring method and device, comprising the following steps: acquiring initial point cloud data of a vehicle; filtering the initial point cloud data to obtain first point cloud data; grid segmentation is carried out on the first point cloud data to obtain a rectangular grid, and coordinate values of each first point cloud data are calculated to obtain grid row and column numbers of each first point cloud data in the rectangular grid; according to the first point cloud data, all grid rows and all column numbers, obtaining triangles corresponding to each first point cloud data, and detecting all triangles to obtain second point cloud data; separating the second point cloud data to obtain nodes, so as to obtain an R tree structure composed of the nodes, and thinning the second point cloud data according to the R tree structure to obtain third point cloud data in the R tree structure; and analyzing the third point cloud data to obtain the change of the elements on the vehicle running road. The invention realizes the purposes of data analysis and change detection of the vehicle-mounted laser point cloud.

Description

Road element change monitoring method and device

Technical Field

The invention relates to the technical field of road element detection, in particular to a road element change monitoring method and device.

Background

Roads are the most popular transportation hub in cities and are composed of many different elements. These elements include roadways, street trees, street lights, guardrails, traffic lights, traffic signs, etc., which interact and operate in complex environments. The problem of any single element may lead to disturbances in the road system due to frequent traffic and pedestrian activity. Therefore, real-time monitoring of the road is required to ensure its proper operation. The vehicle-mounted mobile measurement system can rapidly and efficiently acquire three-dimensional space data with high precision, high density and high resolution by utilizing a laser scanning technology, and provides powerful data support for urban road information acquisition and updating.

In the prior art, in the process of detecting road elements, the scene of the vehicle-mounted laser point cloud is complex, targets are various, the data volume is large, and the point density distribution is uneven, so that the difficulty of data analysis and change monitoring of the laser point cloud data is high.

Therefore, it is urgently needed to provide a road element change monitoring method and device, which solve the technical problems of complex scene, multiple targets, large data volume and uneven point density distribution of the vehicle-mounted laser point cloud in the prior art, and the difficulty of data analysis and change monitoring of the laser point cloud data is high.

Disclosure of Invention

In view of the foregoing, it is necessary to provide a road element change monitoring method and device, so as to solve the technical problems of complex scene, multiple targets, large data volume and uneven distribution of point density of the vehicle-mounted laser point cloud in the prior art, which results in greater difficulty in data analysis and change monitoring of the laser point cloud data.

In one aspect, the present invention provides a road element change monitoring method, including:

acquiring initial point cloud data of a vehicle; filtering the initial point cloud data to obtain first point cloud data with preset number;

dividing the preset number of first point cloud data into grids to obtain a preset number of rectangular grids, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids;

obtaining triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid lines and all column numbers, and detecting all triangles to obtain the preset number of second point cloud data;

separating the second point cloud data with the preset number to obtain a preset number of nodes, so as to obtain an R tree structure composed of the preset number of nodes, and thinning the second point cloud data with the preset number according to the R tree structure to obtain third point cloud data with the preset number in the R tree structure;

And analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road.

In some possible implementations, the filtering the initial point cloud data to obtain a preset number of first point cloud data includes:

randomly sampling the preset number of initial point cloud data to obtain preset number of first random point cloud data; performing plane fitting on the first random point cloud data with the preset number to obtain a fitting plane;

respectively calculating the distances between the fitting planes and other initial point cloud data except the first random point cloud data of the preset number in the initial point cloud data of the preset number to obtain initial distances corresponding to the other initial point cloud data;

judging each initial distance respectively, and judging whether the first distance is smaller than a preset distance or not;

if not, removing initial point cloud data corresponding to the initial distance;

if so, reserving the initial point cloud data corresponding to the initial distance, and accordingly obtaining the first point cloud data with the preset number according to all the reserved initial point cloud data.

In some possible implementations, the performing grid segmentation on the preset number of first point cloud data to obtain a preset number of rectangular grids includes:

According to the coordinate values of the first point cloud data of the preset number, a first maximum value and a first minimum value of an x coordinate and a second maximum value and a second minimum value of a y coordinate are determined;

and performing grid segmentation on the preset number of first point cloud data according to the first maximum value, the first minimum value, the second maximum value, the second minimum value and the data quantity of the preset number of first point cloud data to obtain a preset number of rectangular grids.

In some possible implementations, the calculating the coordinate value of each first point cloud data to obtain the grid row and column number of each first point cloud data in the preset number of rectangular grids includes:

judging whether all the grid row and column numbers are determined by the point cloud data in the first point cloud data of the preset number;

if not, randomly determining second random point cloud data without determining grid row and column numbers from the first point cloud data with the preset number, and determining third random point cloud data according to the second random point cloud data; the third random point cloud data is first point cloud data closest to the second random point cloud data;

calculating the second random point cloud data and the third random point cloud data according to the grid line number, the first maximum value, the first minimum value, the second maximum value, the second minimum value and the grid column number of the preset number of rectangular grids to obtain grid line numbers and column numbers of the second random point cloud data in the preset number of rectangular grids;

If yes, grid row and column numbers of each first point cloud data in the preset number of rectangular grids are obtained.

In some possible implementations, the rectangular grid includes a corresponding triangle linked list;

obtaining triangles corresponding to the first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, detecting all triangles, and obtaining the preset number of second point cloud data, wherein the method comprises the following steps:

determining a minimum row number, a maximum row number, a minimum column number and a maximum column number from all grid rows and all column numbers, and obtaining a target rectangular grid consisting of a preset number of rectangular grids according to the minimum row number, the maximum row number, the minimum column number and the maximum column number;

determining first point cloud data included in the target rectangular grid as target point cloud data, and obtaining all target point cloud data;

determining triangles corresponding to each target point cloud data according to the second random point cloud data and the third random point cloud data respectively to obtain all triangles of all the target point cloud data;

judging whether the circumscribed circle of the triangle corresponding to the target point cloud data comprises the point cloud data or not, and judging whether a triangle linked list of the rectangular grid corresponding to the target point cloud data comprises a corresponding triangle or not;

If not, storing the triangle into the triangle linked list of the rectangular grid corresponding to the target point cloud data;

obtaining all stored triangles according to all triangle linked lists of all rectangular grids included in the target rectangular grid, and obtaining all second point cloud data according to all target point cloud data of all triangles.

In some possible implementations, the separating the second point cloud data of the preset number to obtain a preset number of nodes, so as to obtain an R tree structure formed by the preset number of nodes, includes:

judging the elevation value of each second point cloud data according to a preset elevation difference value, and determining preset number of characteristic point cloud data belonging to characteristic points in the preset number of second point cloud data;

determining the approximate trend of the point cloud data according to the outer surrounding rectangle of the preset number of characteristic point cloud data;

determining all rectangular grids correspondingly included in each row or each column in the coordinate axis as a node according to the approximate trend, so as to obtain all nodes of all rows or all columns;

dividing all characteristic point cloud data included in all rectangular grids corresponding to each node according to preset child node values to obtain a preset number of new nodes; and obtaining an R tree structure according to the new nodes with the preset number.

In some possible implementations, the thinning the second preset number of point cloud data according to the R tree structure to obtain the third preset number of point cloud data in the R tree structure includes:

determining the number of points stored by a root node, the layer number of each node, the total number of points of all the characteristic point cloud data, the number of grids of all the nodes and the maximum number of points of the characteristic point cloud data contained in a single grid according to the R tree structure;

obtaining grid layer sampling total points corresponding to each rectangular grid according to the number of points stored by the root node, the total number of points, the total number of grids, the maximum number of points and the layer number of each node;

determining the corresponding area of each rectangular grid according to the minimum outer surrounding rectangular vertex coordinates of each rectangular grid, so as to obtain the sum of the areas of all the rectangular grids;

obtaining a sampling interval corresponding to each rectangular grid according to the number of the included points, the area, the total number of sampling points of the grid layer and the sum of the areas, which correspond to each rectangular grid;

thinning each rectangular grid according to the sampling interval of each rectangular grid to obtain third point cloud data with preset number in the R tree structure; the amount of the third point cloud data of each rectangular grid is proportional to the area thereof.

In some possible implementations, after the thinning the preset number of second point cloud data according to the R tree structure to obtain the preset number of third point cloud data in the R tree structure, the method further includes:

judging each third point cloud data in the R tree structure respectively, and judging whether the layer number of the third point cloud data is smaller than or equal to the layer number of a rectangular grid where the third point cloud data is located;

if yes, the third point cloud data is saved in the sequence of breadth-first traversal;

if not, the third point cloud data is saved in the depth-first traversal order.

In some possible implementations, the analyzing the third point cloud data of the preset number to obtain the change of the element on the vehicle driving road includes:

when the third point cloud data with the preset number are rendered, if the viewpoint is far and the coverage area of the viewing port is large, rendering the third point cloud data with the breadth-first traversal and storage to obtain a point cloud result in the visual range;

if the viewpoint is closer and the coverage area of the view port is smaller, rendering the third point cloud data stored by the depth-first traversal to obtain a point cloud result in the visual field;

According to the point cloud result, obtaining an artificial ground object, detecting the artificial ground object, and determining the change of elements on the vehicle running road; the change includes a complete absence or addition of the artificial feature and a partial absence of the artificial feature component.

On the other hand, the invention also provides a road element change monitoring device, which comprises:

the data acquisition module is used for acquiring initial point cloud data of the vehicle; filtering the initial point cloud data to obtain first point cloud data with preset number;

the grid dividing module is used for dividing the preset number of first point cloud data into grids to obtain a preset number of rectangular grids, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids;

the triangle detection module is used for obtaining triangles corresponding to the first point cloud data according to the preset number of first point cloud data, all grid lines and all column numbers, and detecting all the triangles to obtain the preset number of second point cloud data;

the structure determining module is used for separating the preset number of second point cloud data to obtain preset number of nodes, so that an R tree structure formed by the preset number of nodes is obtained, and the preset number of second point cloud data is thinned according to the R tree structure to obtain preset number of third point cloud data in the R tree structure;

And the change determining module is used for analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road.

The beneficial effects of adopting the embodiment are as follows: according to the road element change monitoring method provided by the invention, initial point cloud data of a vehicle are obtained; filtering the initial point cloud data to obtain first point cloud data with preset number; dividing the grid of the preset number of first point cloud data to obtain a preset number of rectangular grids, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids; obtaining triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, and detecting all triangles to obtain the preset number of second point cloud data; separating the second point cloud data with the preset number to obtain a preset number of nodes, so as to obtain an R tree structure composed of the preset number of nodes, and thinning the second point cloud data with the preset number according to the R tree structure to obtain third point cloud data with the preset number in the R tree structure; and analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road. According to the invention, the third point cloud data is obtained by performing a series of processes on the initial point cloud data, such as filtering process, grid division, R tree structure formed, thinning and the like, and then the change of elements on a vehicle driving road is obtained by analyzing the third point cloud data, so that the purposes of data analysis and change detection on the vehicle-mounted laser point cloud are realized.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly explain the drawings needed in the description of the embodiments, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of an embodiment of a road element variation monitoring method according to the present invention;

FIG. 2 is a schematic diagram of an embodiment of a target rectangular grid Rx according to the present invention;

FIG. 3 is a schematic flow chart of an embodiment of a circumscribed circle of a triangle in a target rectangular grid Rx according to the present invention;

FIG. 4 is a flow chart of one embodiment of a mesh area covered by a triangle circumscribing circle provided by the present invention;

FIG. 5 is a schematic structural diagram of an embodiment of a road element change monitoring device according to the present invention;

fig. 6 is a schematic structural diagram of an embodiment of an electronic device according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.

The embodiment of the invention provides a road element change monitoring method and device, which are respectively described below.

Fig. 1 is a flow chart of an embodiment of a road element change monitoring method according to the present invention, where, as shown in fig. 1, the road element change monitoring method includes:

s101, acquiring initial point cloud data of a vehicle; filtering the initial point cloud data to obtain first point cloud data with preset number;

S102, carrying out grid segmentation on the first point cloud data with the preset number to obtain a rectangular grid with the preset number, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the rectangular grid with the preset number;

s103, obtaining triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, and detecting all triangles to obtain the preset number of second point cloud data;

s104, separating the second point cloud data with the preset number to obtain nodes with the preset number, so as to obtain an R tree structure composed of the nodes with the preset number, and thinning the second point cloud data with the preset number according to the R tree structure to obtain third point cloud data with the preset number in the R tree structure;

s105, analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road.

Compared with the prior art, the road element change monitoring method provided by the embodiment of the invention acquires the initial point cloud data of the vehicle; filtering the initial point cloud data to obtain first point cloud data with preset number; dividing the grid of the preset number of first point cloud data to obtain a preset number of rectangular grids, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids; obtaining triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, and detecting all triangles to obtain the preset number of second point cloud data; separating the second point cloud data with the preset number to obtain a preset number of nodes, so as to obtain an R tree structure composed of the preset number of nodes, and thinning the second point cloud data with the preset number according to the R tree structure to obtain third point cloud data with the preset number in the R tree structure; and analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road. According to the invention, the third point cloud data is obtained by performing a series of processes on the initial point cloud data, such as filtering process, grid division, R tree structure formed, thinning and the like, and then the change of elements on a vehicle driving road is obtained by analyzing the third point cloud data, so that the purposes of data analysis and change detection on the vehicle-mounted laser point cloud are realized.

In the embodiment of the invention, a multi-platform laser radar measurement system is used as a base vehicle-mounted laser scanner for collecting point cloud data, the cloud data is calculated and processed to obtain the point cloud data meeting the road monitoring precision requirement, and the collected point cloud data is used as a data source for urban road element change monitoring.

It should be noted that: to randomly sample the initial point cloud data, in some embodiments of the present invention, step S101 includes:

respectively calculating the distances between other initial point cloud data except the first random point cloud data of the preset number in the initial point cloud data of the preset number and the fitting plane to obtain the initial distance corresponding to each other initial point cloud data;

if so, the initial point cloud data corresponding to the initial distance is reserved, so that the first point cloud data with the preset number are obtained according to all the reserved initial point cloud data.

In a specific embodiment of the present invention, a preset distance TolError, tolError may be set to represent a threshold amount of a point cloud denoising error, a point greater than the threshold is determined to be a noise point, a nearest neighboring point of each initial point cloud data may be randomly sampled in a preset number of initial point cloud data to obtain a preset number of first random point cloud data, for example, K nearest neighboring points of an a initial point cloud data in the preset number of initial point cloud data are sampled to obtain K first random point cloud data, and then plane fitting may be performed on the K first random point cloud data to obtain a fitting plane.

Further, the distances between the fitting planes and other initial point cloud data except the first random point cloud data in the preset number of initial point cloud data can be calculated to obtain the initial distances corresponding to the other initial point cloud data, and one initial point cloud data in the other initial point cloud data can be determined Calculating the distance between the initial point cloud data and the fitting plane, wherein the plane equation of the fitting plane is +.>Initial point cloud data +.>Distance to the fitting planedThe calculation of (2) is shown in the formula (1):

（1）

in the method, in the process of the invention,A、B、C、Dis a constant describing the nature of the fit plane,x、y、zis the coordinates of the initial point cloud data on the plane.

Also can judge the distancedWhether the preset distance TolError is exceeded, if so, the initial point cloud data is obtainedRecording and removing, and then one of the other initial point cloud data can be redeterminedRepeating the above steps, and determining other initial point cloud data which does not exceed the preset distance TolError as first point cloud data until the distance of all other initial point cloud datadAnd (5) finishing calculation, and obtaining all the first point cloud data. The preset distance TolError may be set according to practical situations, which is not limited herein.

In some embodiments of the present invention, step S102 includes:

according to the coordinate values of the first point cloud data of the preset number, determining a first maximum value and a first minimum value of an x coordinate and a second maximum value and a second minimum value of a y coordinate;

It should be noted that, in order to better manage the point cloud data, a topological relation between the point cloud data can be established, and the embodiment of the invention adopts a point cloud filtering method facing road element monitoring, namely a Delaunay triangulation method based on grid indexes, and simultaneously adopts a stack to replace a Delaunay triangulation recursive algorithm, so that the information storage amount during stacking is reduced, and the memory consumption is reduced.

In a specific embodiment of the present invention, a first maximum value and a first minimum value of an x coordinate and a second maximum value and a second minimum value of a y coordinate in the preset number of first point cloud data may be determined according to coordinate values of the preset number of first point cloud data in a coordinate system, that isX _min 、X _max 、Y _min 、Y _max Then, grid division can be performed on the preset number of first point cloud data according to the first maximum value, the first minimum value, the second maximum value, the second minimum value and the preset number of data volume of the first point cloud data to obtain a preset number of rectangular grids, wherein the number of grids is about 1/8 of the discrete points, and the calculation of the grid number and the grid number after grid division is shown in formulas (2), (3) and (4):

（2）

（3）

（4）

in the method, in the process of the invention,the grid line number is; />The number of the grid columns; / >The number of the points is the number of the points; />Is an X coordinate range; />Is the Y coordinate range. According to the grid line number and the grid column number of each grid, a preset number of rectangular grids can be obtained.

In some embodiments of the present invention, step S102 includes:

judging whether all the grid line and column numbers are determined by the point cloud data in the first point cloud data of the preset number;

if not, randomly determining second random point cloud data without determining grid row and column numbers from the first point cloud data with the preset number, and determining third random point cloud data according to the second random point cloud data; the third random point cloud data is the first point cloud data closest to the second random point cloud data;

In a specific embodiment of the present invention, grid row and column numbers of all first point cloud data in the preset number of first point cloud data need to be determined, so that whether all the grid row and column numbers are determined by the point cloud data in the preset number of first point cloud data needs to be judged first, if not, a second random point cloud data without determining the grid row and column numbers needs to be determined randomly from the preset number of first point cloud data, the second random point cloud data can be determined randomly, and then the point cloud data closest to the second random point cloud data can be determined as third random point cloud data; then, the grid row number where the second random point cloud data and the third random point cloud data are located can be calculated, and the calculation is shown as a formula (5):

（5）

Wherein x and y are plane coordinates of the point cloud data.

Then judging whether the grid row and column numbers are all determined by the point cloud data in the first point cloud data of the preset number, if not, returning the second random point cloud data which is randomly determined to be not determined by the grid row and column numbers from the first point cloud data of the preset number, and determining third random point cloud data according to the second random point cloud data; and (3) recalculating grid row and column numbers when the third random point cloud data is the first point cloud data closest to the second random point cloud data, and if so, indicating that the grid row and column numbers are all determined by the point cloud data in the first point cloud data with the preset number, so as to obtain the grid row and column numbers of each first point cloud data in the rectangular grid with the preset number.

In some embodiments of the present invention, the rectangular grid includes a corresponding triangular linked list; step S103 includes:

Determining first point cloud data included in a target rectangular grid as target point cloud data, and obtaining all target point cloud data;

determining triangles corresponding to each target point cloud data according to the second random point cloud data and the third random point cloud data respectively to obtain all triangles of all target point cloud data;

judging whether the circumscribed circle of the triangle corresponding to the target point cloud data comprises the point cloud data or not, and judging whether the triangle linked list of the rectangular grid corresponding to the target point cloud data comprises the corresponding triangle or not;

if not, storing the triangle into a triangle linked list of a rectangular grid corresponding to the target point cloud data;

and obtaining all the stored triangles according to all triangle linked lists of all the rectangular grids included in the target rectangular grid, and obtaining all second point cloud data according to all target point cloud data of all the triangles.

In a specific embodiment of the present invention, each rectangular grid includes a corresponding triangle linked list, and the minimum line number may be determined according to all column numbers of all grid lines in a preset number of rectangular grids of all first point cloud datar _min Maximum line numberr _max Minimum column numberc _min And maximum column numberc _max Thus, a target rectangular grid consisting of a preset number of rectangular grids can be obtained according to the minimum line number, the maximum line number, the minimum column number and the maximum column number, as shown in fig. 2, P1 is second random point cloud data, P2 is third random point cloud data, and the target rectangular grid is R _x A bar grid is added to the slash grid. Traversing rectangular regionsR _x Determining first point cloud data included in the target rectangular grid as target point cloud data, thereby obtaining all target point cloud data, randomly determining one target point cloud data from the target point cloud data, and forming a triangle by the target point cloud data PTemp, the P1 and the P2 points as shown in figure 3T _ri The circumscribed circles of the triangle can be found, and as shown in fig. 4, the grid area covered by the circumscribed circles can be obtained, and the vertical bar grid area in fig. 4. Then the target point cloud data in the coverage range of the circumcircle can be traversed, and the triangle is performedT _ri Empty circle detection of (2): if no point cloud data falls into the circumscribed circle and no triangle exists in the triangle linked list of the rectangular grid corresponding to the target point cloud data PTemp, the triangleT _ri Satisfying the networking condition of Delaunay triangle net, popping edges P1-P2, storing triangle into triangle chain table of rectangle grid where target point cloud data PTemp is located, and popping the rest of edges P2-PTemp and PTemp-P1 and repeating from presetRandomly determining second random point cloud data with undetermined grid row and column numbers in the first number of point cloud data, and determining third random point cloud data according to the second random point cloud data; the third random point cloud data is the first point cloud data nearest to the second random point cloud data and the subsequent steps; if the point cloud data is detected to fall into the circumscribed circle, from the next target point cloud data PTemp in the re-detection area Rx until the stack is empty, all the stored triangles can be obtained according to all triangle linked lists of all the rectangular grids included in the target rectangular grid, and then all the second point cloud data can be obtained according to all the target point cloud data of all the triangles.

In some embodiments of the present invention, step S104 includes:

judging the elevation value of each second point cloud data according to the preset elevation difference value, and determining the preset number of characteristic point cloud data belonging to the characteristic points in the preset number of second point cloud data;

dividing all characteristic point cloud data included in all rectangular grids corresponding to each node according to preset child node values to obtain a preset number of new nodes; and obtaining the R tree structure according to the preset number of new nodes.

It should be noted that after obtaining all triangles, each vertex in all triangles can be traversed, so as to detect the artificial ground object of the road. For the artificial ground object, the ground objects such as traffic signal lamps, traffic sign plates, street lamps and the like have obvious rod-shaped structures, and basically all the street lamps and the traffic sign plates are approximately perpendicular to the central line of the road, so that the angle between the main direction of the road and the main direction of the artificial ground object can be utilized to describe the direction relationship between the artificial ground object and the road.

Further, the number of the preset number can be setDetecting the second point cloud data, so as to obtain edge points of the artificial ground object, counting clusters after the edge points of the artificial ground object are clustered, marking 2 edge points of each scanning line, marking edge point clusters with the number smaller than a threshold value as noise points, marking edge point clusters with the number larger than the threshold value as constraint points as the edge points to be optimized, and restraining and optimizing the noise points. The nearest M ipsilateral constraint points can also be selected respectively forward and backward as constraint points of the noise point according to the scan line number of the noise point, for example, 10 ipsilateral constraint points. And then the included angle of the vectors of the front constraint point and the rear constraint point can be calculated. Assume that there are two points、/>The azimuth angles of the two points are respectively +.>The calculation is as shown in formula (6):

（6）

in the method, in the process of the invention,dthe vertical distance threshold for the two-point line is a constant.

If the azimuth angle is greater than the preset angle threshold, the constraint points are regarded as edges belonging to the same artificial ground object. Otherwise, only the first 10 constraint points are selected as constraint points. The preset angle threshold may be set according to practical situations, which is not limited in the embodiments of the present invention. The method comprises the steps of carrying out RANSAC fitting on selected constraint points to obtain an artificial ground object edge line, calculating the distance from the point on a scanning line where a noise point is located to the artificial ground object edge line, selecting the point with the smallest distance as the artificial ground object edge point after constraint optimization, and obtaining the characteristic points of the street lamp, the traffic signal lamp and the traffic sign board based on the relative position relation with the road direction, wherein the characteristic points are required to be further refined. The process of extracting the road lamp, the traffic signal lamp and the traffic sign board feature points is not limited to the process, and can be obtained by other technologies.

In a specific embodiment of the present invention, a height difference value may be set, each point cloud data includes a corresponding height value, and the greater the likelihood that the point cloud data with a large height value belongs to a street lamp, a traffic signal lamp, and a traffic sign, so the height value of each second point cloud data may be determined according to the preset height difference value, to determine a preset number of feature point cloud data belonging to a feature point in the preset number of second point cloud data, for example, assuming that a is a road laser point data set (i.e., a data set of the preset number of feature point cloud data),Sharpif the road lamp, the traffic signal lamp and the traffic sign board point set are adopted, the points meeting the following detection functions areSharpThe elements of the set, the probe function is shown in formula (7):

（7）

in the method, in the process of the invention,Sharpis a street lamp, a traffic signal lamp and a traffic sign board set;the current point cloud data; />Neighboring points of the current point cloud data; />Is->Elevation of the dot; />Is->Elevation of the dot; />Is->To->A distance;is a high difference value for a given range. I.e. for a given->A point, if there is a point nearby +.>Satisfy the relation->Then (I)>The points are characteristic points of the street lamp, the traffic signal lamp and the traffic sign board, and the preset number of characteristic point cloud data belonging to the characteristic points in the preset number of second point cloud data. The preset number of the characteristic point cloud data belonging to the characteristic points in the preset number of the second point cloud data can be obtained through the process.

Further, the outer surrounding rectangle of the preset number of feature point cloud data can be determined according to the positions of the preset number of feature point cloud data in the coordinate system, so that the general trend of the road point cloud can be determined, for example, taking the trend of the X axis as an example, if the longer side of the outer surrounding rectangle is parallel or nearly parallel to the X axis, the road can be inferred to mainly extend along the X axis direction; conversely, if the longer side is parallel or nearly parallel to the Y-axis, it may be indicated that the road runs primarily along the Y-axis. All grids of the same column (row) can be organized into a slice to serve as one node of the first layer R tree, so that all nodes of all rows or all columns are obtained, the preset sub-node value P of each R tree node can be set, each P sub-node in a new layer of nodes is combined into one group to serve as one node of one layer on the R tree in the X or Y direction, the last remaining sub-nodes which are less than P are combined into one node to obtain all regenerated new nodes, the number Q of the new layer of R tree nodes can be calculated, if Q is more than or equal to P, the process is carried out again, each P sub-node in the new layer of nodes is combined into one group to serve as one node of one layer on the R tree in the X or Y direction, the last remaining sub-nodes which are less than P are combined into one node, all regenerated new nodes are obtained, and subsequent steps are obtained, and otherwise, the algorithm exits. Further, all grids may be organized into a multi-layered R-tree structure after the steps described above.

In some embodiments of the present invention, step S104 includes:

according to the R tree structure, determining the number of points stored by a root node, the number of layers of each node, the total number of points of all the characteristic point cloud data, the number of grids of all the nodes and the maximum number of points of the characteristic point cloud data contained in a single grid;

obtaining grid layer sampling total points corresponding to each rectangular grid according to the number of points stored by the root node, the total number of points, all grid numbers, the maximum number of points and the layer number of each node;

obtaining a sampling interval corresponding to each rectangular grid according to the number and the area of the points included in each rectangular grid and the sum of the total number and the area of the grid layer sampling;

thinning each rectangular grid according to the sampling interval of each rectangular grid to obtain third point cloud data with preset number in the R tree structure; the amount of third point cloud data of each rectangular grid is proportional to its area.

According to the embodiment of the invention, the data volume to be transmitted and processed can be reduced by performing thinning and compression on the point cloud data, so that the rendering performance is improved. In the LOD structure of the point cloud, the nodes of a high level represent the whole shape and the rough characteristics, the nodes of a low level represent the details and the local characteristics, object models of different detail levels are constructed through LOD technology, grids with higher density can be distributed to higher levels, and more detail information is reserved; while less dense grids may be assigned to lower levels, reducing detail and improving rendering efficiency.

In a specific embodiment of the present invention, after obtaining the R-tree structure, the number of layers where each node is located may also be obtainedTotal point number of all characteristic point cloud data +.>All mesh number of all nodes->And maximum point number of feature point cloud data included in single grid +.>. And further can calculate the sampling interval of two adjacent layers +.>The calculations are shown in equations (8), (9) and (10):

（8）

in the method, in the process of the invention,the number of leaf nodes of the index structure is equal to that of the leaf nodes of the index structure when the index structure is uniformly divided.

（9）

（10）

In the method, in the process of the invention,ceilthe function is a round-up function,the number of layers of the index structure is equal to that of the index structure when the index structure is divided.

Entry into the root node store where the index structure may be setThe specific setting condition can be set according to the actual condition, so that the total sampling point number of the grid layer can be calculated according to the point number>The calculation is as shown in formula (11):

（11）

traversing all rectangular grids, and calculating the area of each rectangular grid according to the minimum surrounding rectangular vertex coordinates of each rectangular gridAnd the sum of the areas of all meshes +.>The calculations are shown in equations (12) and (13):

)（12）

（13）

in the method, in the process of the invention,for the number of grids->，/>，/>，/>Is the firstiThe outer perimeter of each rectangular grid encloses rectangular vertex coordinates.

Thereby the sampling interval of each rectangular grid can be calculated The calculation is as shown in equation (14):

（14）

in the method, in the process of the invention,the number of points is the point cloud data in the ith rectangular grid.

And thinning the sub-nodes of each rectangular grid according to the calculated sampling interval of each rectangular grid, and obtaining the third point cloud data with the preset number in the R tree structure after the thinning is completed, wherein the number of the points sampled in each rectangular grid is in direct proportion to the area of the third point cloud data and is irrelevant to the point cloud density. Because the point cloud density of a single rectangular grid is relatively uniform, the depth is greater than in the index structureCan be used for creating LOD nodes in a mode of sampling at equal intervals, wherein the sampling interval is +.>. Similarly, the point cloud density of all the grids after sampling is substantially uniform, and thus, the depth is less than +.>The LOD node is created by equally-spaced sampling method, and the sampling interval is equal to +.>。

In some embodiments of the present invention, after step S104, further includes:

judging each third point cloud data in the R tree structure respectively, and judging whether the layer number of the third point cloud data is smaller than or equal to the grid layer number of the rectangular grid where the third point cloud data is located;

If not, the third point cloud data is saved in the depth-first traversal order.

According to the point cloud data classified by the characteristic points of the artificial ground object, all the node data are stored in the same file according to the traversing sequence of the index structure, and in this way, progressive loading and rendering of the point cloud can be realized, and a high-quality visual effect is obtained.

In a specific embodiment of the present invention, for a node in the R tree structure where the number of layers of the third point cloud data is less than or equal to the number of layers of the grid of the rectangular grid where the third point cloud data is located, the third point cloud data of the node may be stored using a breadth-first traversal order. And for nodes with the grid layer number larger than the rectangular grid where the third point cloud data is located, storing the third point cloud data of the nodes by using the depth-first traversal sequence.

In some embodiments of the present invention, step S105 includes:

when the third point cloud data with the preset number are rendered, if the viewpoint is far and the coverage area of the view port is large, rendering the third point cloud data with the breadth-first traversal and storage to obtain a point cloud result in the visible range;

if the viewpoint is closer and the coverage area of the view port is smaller, rendering the third point cloud data stored in the depth-first traversal to obtain a point cloud result in the visual field;

According to the point cloud result, obtaining an artificial ground object, detecting the artificial ground object, and determining the change of elements on a vehicle running road; changes include complete deletions or additions of artificial features and partial deletions of artificial feature components.

The embodiment of the invention can take the analysis range of the visual field of the road environment as a constraint condition, and can meet the visualization, interaction and scheduling processing of the point cloud data by classifying, indexing and managing the point cloud data in the same position and at different periods, and the road change monitoring of the vehicle-mounted laser point cloud determines the change of the road by obtaining the difference of the road point cloud data at a plurality of time points.

In the specific embodiment of the invention, when the third point cloud data with the preset number is rendered, if the viewpoint is far and the coverage area of the view port is large, the third point cloud data with the breadth-first traversal and preservation are rendered, and a point cloud result in the visible range is obtained; if the viewpoint is closer and the coverage area of the view port is smaller, rendering the third point cloud data stored in the depth-first traversal to obtain a point cloud result in the visual field; the method comprises the steps that through a point cloud result in a visible range, the artificial ground objects of roads in the same position and different periods are objectively identified based on geometric surface features, and if the artificial ground objects in the same position are subjected to space position and sudden total missing or new addition of structures, the area can be judged to be possibly subjected to change of road elements; for judging the partial missing of the artificial ground object part, the center position of the artificial ground object can be fitted based on the incomplete scattered points by using a template matching method, and the change scale of the artificial ground object is analyzed through the visibility contrast, so that the change information of the road elements is judged.

In order to better implement the road element change monitoring method in the embodiment of the present invention, correspondingly, on the basis of the road element change monitoring method, the embodiment of the present invention further provides a road element change monitoring device, as shown in fig. 5, where the road element change monitoring device includes:

a data acquisition module 501, configured to acquire initial point cloud data of a vehicle; filtering the initial point cloud data to obtain first point cloud data with preset number;

the grid dividing module 502 is configured to perform grid division on a preset number of first point cloud data to obtain a preset number of rectangular grids, and calculate coordinate values of each first point cloud data to obtain grid rows and column numbers of each first point cloud data in the preset number of rectangular grids;

the triangle detection module 503 is configured to obtain triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, and detect all triangles to obtain the preset number of second point cloud data;

the structure determining module 504 is configured to separate the preset number of second point cloud data to obtain a preset number of nodes, thereby obtaining an R tree structure composed of the preset number of nodes, and dilute the preset number of second point cloud data according to the R tree structure to obtain preset number of third point cloud data in the R tree structure;

The change determining module 505 is configured to analyze the third point cloud data of the preset number to obtain a change of the element on the vehicle driving road.

The road element change monitoring device provided in the foregoing embodiment may implement the technical solution described in the foregoing road element change monitoring method embodiment, and the specific implementation principle of each module or unit may refer to the corresponding content in the foregoing road element change monitoring method embodiment, which is not repeated herein.

As shown in fig. 6, the present invention further provides an electronic device 600 accordingly. The electronic device 600 comprises a processor 601, a memory 602 and a display 603. Fig. 6 shows only a portion of the components of the electronic device 600, but it should be understood that not all of the illustrated components are required to be implemented and that more or fewer components may be implemented instead.

The memory 602 may be an internal storage unit of the electronic device 600 in some embodiments, such as a hard disk or memory of the electronic device 600. The memory 602 may also be an external storage device of the electronic device 600 in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the electronic device 600.

Further, the memory 602 may also include both internal storage units and external storage devices of the electronic device 600. The memory 602 is used for storing application software and various types of data for installing the electronic device 600.

The processor 601 may in some embodiments be a central processing unit (Central Processing Unit, CPU), microprocessor or other data processing chip for executing program code or processing data stored in the memory 602, such as the road element change monitoring method of the present invention.

The display 603 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like in some embodiments. The display 603 is used for displaying information at the electronic device 600 and for displaying a visual user interface. The components 601-603 of the electronic device 600 communicate with each other via a system bus.

In some embodiments of the present invention, when the processor 601 executes the road element change monitoring program in the memory 602, the following steps may be implemented:

Dividing the grid of the preset number of first point cloud data to obtain a preset number of rectangular grids, and calculating coordinate values of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids;

obtaining triangles corresponding to each first point cloud data according to the preset number of first point cloud data, all grid rows and all column numbers, and detecting all triangles to obtain the preset number of second point cloud data;

It should be understood that: the processor 601 may perform other functions in addition to the above functions when executing the road element change monitoring program in the memory 602, and in particular reference may be made to the description of the corresponding method embodiments above.

Further, the type of the electronic device 600 is not particularly limited, and the electronic device 600 may be a mobile phone, a tablet computer, a personal digital assistant (personal digitalassistant, PDA), a wearable device, a laptop (laptop), or other portable electronic devices. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices that carry IOS, android, microsoft or other operating systems. The portable electronic device described above may also be other portable electronic devices, such as a laptop computer (laptop) or the like having a touch-sensitive surface, e.g. a touch panel. It should also be appreciated that in other embodiments of the invention, the electronic device 600 may not be a portable electronic device, but rather a desktop computer having a touch-sensitive surface (e.g., a touch panel).

Accordingly, the embodiments of the present application further provide a computer readable storage medium, where the computer readable storage medium is used to store a computer readable program or instruction, and when the program or instruction is executed by a processor, the steps or functions of the road element change monitoring method provided in the foregoing method embodiments can be implemented.

Those skilled in the art will appreciate that all or part of the flow of the methods of the embodiments described above may be accomplished by way of a computer program stored in a computer readable storage medium to instruct related hardware (e.g., a processor, a controller, etc.). The computer readable storage medium is a magnetic disk, an optical disk, a read-only memory or a random access memory.

The method and the device for monitoring the change of the road element provided by the invention are described in detail, and specific examples are applied to the description of the principle and the implementation mode of the invention, and the description of the examples is only used for helping to understand the method and the core idea of the invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present invention, the present description should not be construed as limiting the present invention.

Claims

1. A road element change monitoring method, characterized by comprising:

analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road;

The step of performing grid segmentation on the first point cloud data with the preset number to obtain a rectangular grid with the preset number comprises the following steps:

performing grid segmentation on the preset number of first point cloud data according to the first maximum value, the first minimum value, the second maximum value, the second minimum value and the data quantity of the preset number of first point cloud data to obtain a preset number of rectangular grids;

the calculating the coordinate value of each first point cloud data to obtain grid row and column numbers of each first point cloud data in the preset number of rectangular grids comprises the following steps:

if yes, grid row and column numbers of each first point cloud data in the preset number of rectangular grids are obtained;

the rectangular grid comprises a corresponding triangular linked list;

2. The method for monitoring the change of road elements according to claim 1, wherein the filtering the initial point cloud data to obtain a preset number of first point cloud data includes:

3. The method for monitoring the change of the road element according to claim 1, wherein the step of separating the second point cloud data of the preset number to obtain the preset number of nodes, thereby obtaining an R tree structure composed of the preset number of nodes includes:

4. The method for monitoring the change of the road element according to claim 3, wherein the thinning the second point cloud data with the preset number according to the R tree structure to obtain third point cloud data with the preset number in the R tree structure includes:

5. The method for monitoring the change of the road element according to claim 1, further comprising, after the thinning the preset number of second point cloud data according to the R tree structure to obtain the preset number of third point cloud data in the R tree structure:

if not, the third point cloud data is saved in the depth-first traversal order.

6. The method for monitoring the change of the road element according to claim 5, wherein the analyzing the third point cloud data of the preset number to obtain the change of the element on the vehicle driving road comprises:

7. A road element change monitoring device, characterized by comprising:

the change determining module is used for analyzing the third point cloud data of the preset number to obtain the change of the elements on the vehicle driving road;

The grid segmentation module is further used for determining a first maximum value and a first minimum value of an x coordinate and a second maximum value and a second minimum value of a y coordinate according to the coordinate values of the preset number of first point cloud data; performing grid segmentation on the preset number of first point cloud data according to the first maximum value, the first minimum value, the second maximum value, the second minimum value and the data quantity of the preset number of first point cloud data to obtain a preset number of rectangular grids; judging whether all the grid row and column numbers are determined by the point cloud data in the first point cloud data of the preset number; if not, randomly determining second random point cloud data without determining grid row and column numbers from the first point cloud data with the preset number, and determining third random point cloud data according to the second random point cloud data; the third random point cloud data is first point cloud data closest to the second random point cloud data; calculating the second random point cloud data and the third random point cloud data according to the grid line number, the first maximum value, the first minimum value, the second maximum value, the second minimum value and the grid column number of the preset number of rectangular grids to obtain grid line numbers and column numbers of the second random point cloud data in the preset number of rectangular grids; if yes, grid row and column numbers of each first point cloud data in the preset number of rectangular grids are obtained;

The rectangular grid comprises a corresponding triangular linked list;

the triangle detection module is further configured to determine a minimum line number, a maximum line number, a minimum column number, and a maximum column number from the all grid lines and the all column numbers, and obtain a target rectangular grid composed of a preset number of rectangular grids according to the minimum line number, the maximum line number, the minimum column number, and the maximum column number; determining first point cloud data included in the target rectangular grid as target point cloud data, and obtaining all target point cloud data; determining triangles corresponding to each target point cloud data according to the second random point cloud data and the third random point cloud data respectively to obtain all triangles of all the target point cloud data; judging whether the circumscribed circle of the triangle corresponding to the target point cloud data comprises the point cloud data or not, and judging whether a triangle linked list of the rectangular grid corresponding to the target point cloud data comprises a corresponding triangle or not; if not, storing the triangle into the triangle linked list of the rectangular grid corresponding to the target point cloud data; obtaining all stored triangles according to all triangle linked lists of all rectangular grids included in the target rectangular grid, and obtaining all second point cloud data according to all target point cloud data of all triangles.