CN117876446A - Point cloud registration method and device based on cylinder detection - Google Patents
Point cloud registration method and device based on cylinder detection Download PDFInfo
- Publication number
- CN117876446A CN117876446A CN202410269026.XA CN202410269026A CN117876446A CN 117876446 A CN117876446 A CN 117876446A CN 202410269026 A CN202410269026 A CN 202410269026A CN 117876446 A CN117876446 A CN 117876446A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- registration
- cylinder
- registered
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 95
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000011159 matrix material Substances 0.000 claims abstract description 146
- 238000004364 calculation method Methods 0.000 claims abstract description 14
- 230000005484 gravity Effects 0.000 claims description 18
- 238000013519 translation Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 230000007246 mechanism Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The application provides a point cloud registration method and device based on cylinder detection, which are used for carrying out cylinder detection on an obtained point cloud to be registered and a target point cloud, obtaining a primary registration matrix based on cylinder calculation in the point cloud to be registered and the target point cloud, and carrying out registration on the point cloud to be registered based on the primary registration matrix to obtain the primary registration point cloud. And rotating the initial registration point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating a fine registration matrix between the initial registration point cloud and the target point cloud under each rotation angle. And obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix. In the scheme, primary registration is realized based on the detected cylinder, and fine registration is realized based on the point clouds under a plurality of different rotation angles.
Description
Technical Field
The invention relates to the technical field of three-dimensional point cloud processing, in particular to a point cloud registration method and device based on cylinder detection.
Background
In the field of three-dimensional point cloud processing, registration between point clouds is an important task. Point cloud registration is mainly achieved using features of points in the point cloud or some other geometrical features. Existing point cloud registration methods are typically implemented based on planar features in two point clouds. However, in the point cloud registration mode based on the plane characteristics, at least two planes which are not parallel to each other are needed in the point cloud to realize the registration.
However, in an actual scenario, some point clouds may only have a set of parallel planes, in which case, the existing point cloud registration manner cannot solve the problem of registration of such point clouds, and the registration of such point clouds cannot be effectively achieved.
Disclosure of Invention
The invention aims at providing a point cloud registration method and device based on cylinder detection, which can quickly and accurately realize point cloud registration.
Embodiments of the invention may be implemented as follows:
in a first aspect, the present invention provides a point cloud registration method based on cylinder detection, the method comprising:
acquiring a point cloud to be registered and a target point cloud;
performing cylinder detection on the point cloud to be aligned, and performing cylinder detection on the target point cloud;
Calculating to obtain an initial registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the initial registration matrix to obtain an initial registration point cloud;
rotating the primary registration point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the primary registration point cloud and the target point cloud under each rotation angle;
and obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
In an optional embodiment, the step of calculating the primary registration matrix based on the cylinders in the point cloud to be registered and the target point cloud includes:
determining the cylinder to be registered with the largest radius in all cylinders in the point cloud to be registered, and determining the target cylinder with the largest radius in all cylinders in the target point cloud;
and calculating to obtain an initial registration matrix based on the cylinder to be registered and the target cylinder.
In an alternative embodiment, the step of calculating the primary registration matrix based on the cylinder to be registered and the target cylinder includes:
Obtaining a first direction vector of the axis of the cylinder to be registered and obtaining a second direction vector of the axis of the target cylinder;
determining a first center of gravity point of the cylinder to be registered based on the three-dimensional points of the cylinder to be registered in the point cloud to be registered, and determining a second center of gravity point of the target cylinder based on the three-dimensional points of the target cylinder in the target point cloud;
and calculating according to the first direction vector, the second direction vector, the first gravity center point and the second gravity center point to obtain an initial registration matrix.
In an alternative embodiment, the step of rotating the primary alignment point cloud about the axis of the cylinder by a plurality of different rotation angles includes:
turning the primary alignment point cloud around the centroid of the cylinder and 180 degrees in the axial direction of the cylinder;
and respectively rotating the initial registration point clouds before and after overturning around the axis of the cylinder according to a plurality of different rotation angles.
In an alternative embodiment, the step of obtaining an optimal fine registration matrix of the plurality of fine registration matrices comprises:
for each fine registration matrix, obtaining a fine registration point cloud after fine registration of the initial registration point cloud based on the fine registration matrix;
Calculating an error value between the accurate point cloud and the target point cloud;
and obtaining the minimum error value in the error values, and taking the fine registration matrix corresponding to the minimum error value as an optimal fine registration matrix.
In an alternative embodiment, the step of calculating an error value between the precision point cloud and the target point cloud includes:
aiming at each three-dimensional point in the fine registration point cloud, obtaining a three-dimensional point corresponding to the three-dimensional point in the target point cloud;
calculating the distance between the accurate point cloud and the corresponding three-dimensional point in the target point cloud;
and calculating to obtain an error value between the accurate registration point cloud and the target point cloud according to the distance corresponding to each three-dimensional point in the accurate registration point cloud and the number of the three-dimensional points.
In an optional embodiment, the step of performing cylinder detection on the point cloud to be aligned includes:
detecting the axis of a cylinder in the point cloud to be registered, and determining a projection plane perpendicular to the axis;
projecting the three-dimensional points in the point cloud to be registered onto the projection surface to obtain projection points;
and detecting a circle based on the projection point on the projection surface, and determining a corresponding cylinder based on the circle parameter of the detected circle and the axis parameter of the axis.
In an alternative embodiment, the step of performing circle detection based on the projection point on the projection surface includes:
performing straight line detection based on the projection points on the projection surface, and marking the detected projection points on the straight line as invalid points, wherein the invalid points do not participate in subsequent circle detection;
and detecting the circles of the projection points which are not marked as invalid points on the projection surface.
In an alternative embodiment, the step of performing circle detection on the projection points on the projection surface, which are not marked as invalid points, includes:
selecting a plurality of projection points from the projection points which are not marked as invalid points on the projection surface at random each time, and calculating to obtain a circle parameter based on the plurality of projection points;
and when the calculated circle parameters of the projection points selected by the times exceeding the preset times are consistent, determining that the circle corresponding to the circle parameters is detected.
In a second aspect, the present invention provides a point cloud registration device based on cylinder detection, the device comprising:
the acquisition module is used for acquiring the point cloud to be registered and the target point cloud;
the detection module is used for carrying out cylinder detection on the point cloud to be aligned and carrying out cylinder detection on the target point cloud;
The primary registration module is used for calculating to obtain a primary registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the primary registration matrix to obtain a primary registration point cloud;
the rotation module is used for rotating the initial alignment point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the initial alignment point cloud and the target point cloud under each rotation angle;
the calculation module is used for obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
The beneficial effects of the embodiment of the invention include, for example:
the application provides a point cloud registration method and device based on cylinder detection, which are used for carrying out cylinder detection on an obtained point cloud to be registered and a target point cloud, obtaining a primary registration matrix based on cylinder calculation in the point cloud to be registered and the target point cloud, and carrying out registration on the point cloud to be registered based on the primary registration matrix to obtain the primary registration point cloud. And rotating the initial registration point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating a fine registration matrix between the initial registration point cloud and the target point cloud under each rotation angle. And obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix. In the scheme, primary registration is realized based on the detected cylinder, and fine registration is realized based on the point clouds under a plurality of different rotation angles.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a workpiece model;
fig. 2 is a flowchart of a point cloud registration method based on cylinder detection according to an embodiment of the present application;
FIG. 3 is a flow chart of sub-steps included in step S12 of FIG. 2;
FIG. 4 is a flowchart of sub-steps included in step S123 of FIG. 3;
FIG. 5 is a flow chart of sub-steps included in step S1232 of FIG. 4;
FIG. 6 is a flow chart of sub-steps included in step S13 of FIG. 2;
FIG. 7 is a flowchart of sub-steps included in step S132 of FIG. 6;
FIG. 8 is a schematic diagram of cylinder-based registration;
FIG. 9 is a schematic diagram of post-rotation registration of a primary registration point cloud;
FIG. 10 is a flow chart of sub-steps included in step S14 of FIG. 2;
FIG. 11 is a flow chart of sub-steps included in step S15 of FIG. 2;
FIG. 12 is a flowchart of sub-steps included in step S152 of FIG. 11;
fig. 13 is a functional block diagram of a point cloud registration device based on cylinder detection according to an embodiment of the present application;
fig. 14 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
In the existing point cloud registration method, point cloud registration is generally performed by adopting a plane feature based on the point cloud. Specifically, plane segmentation is performed on point cloud data through a plane segmentation algorithm to obtain plane sheets, plane attribute information is calculated, and then a corresponding relation among the plane sheets is established through the plane sheet attribute information, the plane interrelationship and the rotation translation geometric constraint to obtain a corresponding plane pair set for solving coordinate conversion parameters. And solving the coordinate conversion parameters by using the plane parameters, and selecting an optimal solution according to the established point cloud registration overall consistency measurement, namely a final registration result.
The point cloud registration method needs to have obvious plane characteristics in the point cloud, and at least needs to have two planes which are not parallel to each other to perform registration. When there is only one set of parallel planes in the point cloud, the above approach will not succeed in achieving point cloud registration. For example, as shown in the workpiece point cloud schematic diagram in fig. 1, there are only two planes in the point cloud, and the two planes are parallel to each other, so the point cloud registration of the workpiece cannot be performed by using the point cloud registration method based on the plane characteristics.
Based on the research findings, the application provides a point cloud registration scheme based on cylinder detection, adopts a cylinder detection mode, realizes initial registration of point cloud based on the detected cylinder, and realizes fine registration based on the point cloud under a plurality of different rotation angles. Therefore, the registration between the point clouds can be rapidly and accurately realized on the basis of overcoming the limitation of the conventional point cloud registration based on the plane characteristics.
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of 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, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected 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.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the present invention, it should be noted that, if the terms "first," "second," and the like are used merely to distinguish the descriptions, they are not to be construed as indicating or implying relative importance.
It should be noted that the features of the embodiments of the present invention may be combined with each other without conflict.
Referring to fig. 2, a flowchart of a cylinder detection-based point cloud registration method according to an embodiment of the present application may be implemented by a cylinder detection-based point cloud registration device, which may be implemented by software and/or hardware, and may be configured in an electronic device, which may be a computer device. The detailed steps of the point cloud registration method based on cylinder detection are described as follows.
S11, acquiring a point cloud to be registered and a target point cloud.
S12, performing cylinder detection on the point cloud to be aligned, and performing cylinder detection on the target point cloud.
S13, calculating to obtain an initial registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the initial registration matrix to obtain an initial registration point cloud.
And S14, rotating the initial alignment point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the initial alignment point cloud and the target point cloud under each rotation angle.
And S15, obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
In this embodiment, the point cloud to be registered and the target point cloud are three-dimensional point clouds obtained for the same workpiece. The point cloud to be registered can be extracted from the shot image after the workpiece is shot, and the target point cloud can be obtained by constructing a model of the workpiece on the basis of a modeling tool on computer equipment.
In this embodiment, cylinders, such as circular through holes, cylindrical bumps, and the like, exist in the point cloud to be registered and the target point cloud. Furthermore, there are other three-dimensional points in the point cloud to be registered and the target point cloud than the cylinder.
Firstly, cylinder detection can be performed on the point cloud to be registered and the target point cloud respectively, so as to determine cylinders in the point cloud to be registered and the target point cloud. Therefore, the initial registration matrix can be obtained by calculation based on the corresponding cylinders in the point cloud to be registered and the target point cloud, and initial registration between the point cloud to be registered and the target point cloud can be realized based on the initial registration matrix.
In the initial registration point cloud obtained after the initial registration of the point cloud to be registered, the cylinder of the initial registration point cloud is basically coincident with the cylinder in the target point cloud. However, other three-dimensional points exist in the initial registration point cloud and the target point cloud, and accurate registration of the other three-dimensional points is difficult to achieve in a mode of performing initial registration by using a cylinder alone.
Therefore, in this embodiment, the primary registration point cloud after primary registration may be rotated around the axis of the cylinder by a plurality of different rotation angles, for example, may be rotated by a multiple of 30 degrees, for example, 30 degrees, 60 degrees, 90 degrees, or the like, respectively. After the initial registration point clouds under all rotation angles are obtained, fine registration is carried out on the initial registration point clouds and the target point clouds respectively, and a fine registration matrix is obtained.
The initial registration point cloud is rotated by the method, and the fine registration processing is performed, wherein the registration is mainly performed on other three-dimensional points except for a cylinder in the point cloud. Under different rotation angles, different fine registration matrixes can be obtained, and the fine registration matrix with the best fine registration effect is used as the optimal fine registration matrix.
In this embodiment, a final registration matrix is obtained based on the obtained initial registration matrix and the optimal fine registration matrix. On the basis, the point cloud to be registered can be directly registered to the target point cloud based on the final registration matrix.
According to the point cloud registration scheme based on the cylinder detection, the mode of performing cylinder detection on the point cloud is adopted, primary registration is achieved based on the detected cylinder, and fine registration is achieved based on the point cloud under a plurality of different rotation angles.
Referring to fig. 3, in this embodiment, when the above-mentioned point cloud to be aligned performs cylinder detection, the method may be implemented as follows:
s121, detecting the axis of a cylinder in the point cloud to be registered, and determining a projection plane perpendicular to the axis.
And S122, projecting the three-dimensional points in the point cloud to be registered onto the projection surface to obtain projection points.
And S123, detecting a circle based on the projection point on the projection surface, and determining a corresponding cylinder based on the circle parameter of the detected circle and the axis parameter of the axis.
In this embodiment, the axis of the cylinder existing in the point cloud to be registered is detected first, and then the projection plane perpendicular to the axis is determined. It is considered that the point on the curved surface of the cylinder should be a circle on a projection plane perpendicular to the axis. Therefore, three-dimensional points in the point cloud to be registered are projected onto a projection surface, and then circle detection is performed on the projection surface. And determining the cylinder in the point cloud to be registered based on the detected circle parameters of the circle and the axis parameters of the axis.
In this embodiment, when detecting the axis of the cylinder existing in the point cloud to be registered, a voting mechanism is used for detection and determination. For example, a group of three-dimensional points are randomly selected from the point cloud to be registered at a time, wherein the group of three-dimensional points comprises a plurality of three-dimensional points, and the axis parameter of the axis is calculated based on the selected three-dimensional points. If the axis parameters calculated by the three-dimensional point groups exceeding the preset group number are consistent, the axis corresponding to the detected axis parameters can be determined.
On the basis of this, a projection plane perpendicular to the axis is determined, which may be a plane passing through the origin, and whose normal vector coincides with the direction vector of the axis.
And projecting the three-dimensional points meeting the requirements in the point cloud to be registered onto a projection surface. The three-dimensional point meeting the requirement can be a three-dimensional point with an included angle between the normal vector of the three-dimensional point and the normal vector of the projection surface being larger than a preset included angle. The predetermined included angle may be, for example, 85 °.
For three-dimensional points that have been projected onto the projection surface, the three-dimensional points in the point cloud to be registered may be marked as invalid points, which will no longer participate in subsequent detection. In this way, interference with other cylinder detection can be avoided.
In the present embodiment, when the circle detection is performed based on the projection points on the projection surface, it is considered that some of the projection points on the projection surface form a straight line and some of the projection points form a circle. In order to reduce the interference during the circular detection, referring to fig. 4, in the present embodiment, during the circular detection based on the projection points on the projection surface, the following manner may be adopted:
s1231, performing straight line detection based on the projection points on the projection surface, and marking the detected projection points on the straight line as invalid points, wherein the invalid points do not participate in subsequent circle detection.
S1232, detecting the circles of the projection points which are not marked as invalid points on the projection surface.
In this embodiment, the voting mechanism is also used for straight line detection. For example, a set of proxels is randomly selected from the proxels on the projection surface at a time, the set of proxels including a plurality of proxels. And calculating linear parameters of the straight line based on the selected group of projection points, and judging that the straight line corresponding to the linear parameters is detected if the linear parameters calculated by the projection point groups exceeding the preset group number are consistent.
In order to avoid that the projected points on the straight line interfere with the circle detection, the detected projected points on the straight line are marked as invalid points, and the invalid points do not participate in the subsequent circle detection.
After all straight lines on the projection surface are detected and the projection points on the straight lines are marked as invalid points, the projection points which are not marked as the invalid points on the projection surface are detected in a circle.
Referring to fig. 5, in the present embodiment, when performing circle detection on a projection point on a projection surface, which is not marked as an invalid point, the following manner may be implemented:
s12321, randomly selecting a plurality of projection points from the projection points which are not marked as invalid points on the projection surface at a time, and calculating a circle parameter based on the plurality of projection points.
S12322, when the calculated circle parameters of the projection points selected by the times exceeding the preset times are consistent, determining that the circle corresponding to the circle parameters is detected.
In this embodiment, a voting mechanism is also used for circle detection. After detecting the circle on the projection surface, obtaining the circle parameters corresponding to the circle, wherein the circle parameters comprise the circle center position [ ]x c ,y c ) Radius [ ]r). The circle center position is the projection position of the axis of the cylinder in a two-dimensional coordinate system on the projection plane, and the radius of the circle is the radius of the cylinder.
In addition, the axis parameters of the axis obtained in the present embodiment includeAnd->Wherein->The direction vector representing the axis of the cylinder isxyProjection onto a planexThe angle between the shafts is>Representing the angle between the direction vector and the z-axis.
And determining the cylinder in the point cloud to be registered based on the axis parameter of the axis and the detected circle parameter of the circle.
In addition, it should be noted that the detection manner of the cylinder in the target point cloud is consistent with the detection manner of the cylinder in the point cloud to be aligned, and this embodiment is not described herein.
In this embodiment, in consideration of the actual scene, there are often multiple cylinders in the point cloud to be registered and the target point cloud respectively. Referring to fig. 6, in order to facilitate registration between the two, when calculating an initial registration matrix based on cylinders in a point cloud to be registered and a target point cloud, the following manner may be implemented:
S131, determining the cylinder to be registered with the largest radius in all cylinders in the point cloud to be registered, and determining the target cylinder with the largest radius in all cylinders in the target point cloud.
And S132, calculating to obtain an initial registration matrix based on the cylinder to be registered and the target cylinder.
In this embodiment, in the process of performing cylinder detection on the point cloud to be aligned and the target point cloud, the radius of the cylinder may be obtained. Suppose that the cylinder detected in the target point cloud P includes C P1 、C P2 、…C Pn The cylinder detected in the point cloud to be registered Q comprises C Q1 、C Q2 、…C Qn . Assume that the target cylinder with the largest radius in the target point cloud P isThe cylinder to be registered with the largest radius in the point cloud to be registered Q is +.>。
Therefore, the cylinder with the largest radius in the point cloud to be registered is registered with the cylinder with the largest radius in the target point cloud, so that the cylinder of the point cloud to be registered and the cylinder of the target point cloud are the same cylinder, and the problems of interference and mismatching are avoided.
In addition, in the process of detecting the projection surface and the cylinder, the method is realized by adopting a voting mechanism, and the three-dimensional point on the cylinder curved surface with the largest radius is most likely to be detected first. Therefore, the auxiliary determination of the corresponding relation of the cylinders in the two point clouds can be simultaneously realized at the stage of detecting the cylinders.
Referring to fig. 7, in this embodiment, when calculating the primary registration matrix based on the cylinder to be registered and the target cylinder, it can be achieved by:
s1321, obtaining a first direction vector of the axis of the cylinder to be registered, and obtaining a second direction vector of the axis of the target cylinder.
S1322, determining a first gravity center point of the cylinder to be registered based on the three-dimensional points of the cylinder to be registered in the point cloud to be registered, and determining a second gravity center point of the target cylinder based on the three-dimensional points of the target cylinder in the target point cloud.
S1323, calculating to obtain an initial registration matrix according to the first direction vector, the second direction vector, the first center of gravity point and the second center of gravity point.
From the above, it can be obtained the axis parameterAnd->The first direction vector of the axis of the cylinder to be registered can be calculated based on the axis parameters of the axis. Wherein the first direction vector may be identified as +.>。
First, the components in the directions of the respective coordinate axes can be calculated based on the axis parameters and according to the following formulas:
then, the plurality of obtained components are combined to obtain a first direction vector of the axis.
The second direction vector of the axis of the target cylinder is calculated in a similar manner as described above, and the second direction vector can be identified as 。
In addition, the first gravity center point of the cylinder to be registered is calculated based on the three-dimensional points belonging to the cylinder to be registered in the point cloud to be registered, and can be marked as. Likewise, the second center of gravity of the target cylinder calculated based on the three-dimensional points belonging to the target cylinder in the target point cloud may be labeled +.>。
The primary registration matrix includes a rotation matrix and a translation vector. The rotation matrix comprises a rotation shaft and a rotation angle, and the calculation formulas of the rotation shaft and the rotation angle are as follows:
the translation vector is calculated as follows:
after the initial registration matrix is obtained through calculation, the point cloud to be registered is subjected to initial registration based on the initial registration matrix, and the obtained initial registration point cloud can be marked as Q'.
In this embodiment, the initial registration point cloud and the target point cloud are aligned based on cylinders, as shown in fig. 8, that is, the cylinders between the two may achieve overlapping consistency.
However, there are some other three-dimensional points besides the cylinder in the point cloud to be registered and the target point cloud, and alignment of the other three-dimensional points is also required. In this embodiment, assuming that the other three-dimensional points are not symmetrical about the axis of the cylinder, in the case where the cylinders in the two point clouds are aligned, the other three-dimensional points are not necessarily already aligned. As shown in fig. 9, in which a circle represents the plane of a cylinder and a triangle represents other three-dimensional points, it is also necessary to perform an alignment process on the other three-dimensional points.
In this embodiment, the two cylinders can also be aligned with the two planes of the cylinders upside down, taking into account that the points on the cylinders are mutually symmetrical about a vertical plane perpendicular to the axis. Therefore, in order to avoid such an abnormal situation, referring to fig. 10, in this embodiment, when the rotation process is performed on the axis from the initial alignment point cloud to the cylinder, the following manner may be implemented:
s141, the primary alignment point cloud is turned 180 degrees around the mass center of the cylinder and in the axial direction of the cylinder.
And S142, respectively rotating the initial registration point clouds before and after overturning around the axis of the cylinder according to a plurality of different rotation angles.
In this embodiment, the first alignment point cloud is turned over first, which can be understood as taking a vertical plane perpendicular to the axis of the cylinder as a reference plane, and turning over the three-dimensional points in the point cloud to be aligned with the reference plane.
Since it is not clear whether the cylinder in the primary registration point cloud is upside down relative to the cylinder in the target point cloud, in this embodiment, the primary registration point cloud before and after the overturning is rotated around the axis of the cylinder by a plurality of different rotation angles, respectively.
After the initial registration point cloud under each rotation angle is obtained, a fine registration matrix can be obtained based on the calculation of the initial registration point cloud and the target point cloud under each rotation angle. In this embodiment, the calculation of the fine registration matrix may be performed using an iterative closest point (Iterative Closest Point, ICP) algorithm.
And calculating a corresponding fine registration matrix under each rotation angle, wherein the fine registration matrix obtained under a certain rotation angle can enable the initial registration point cloud and the target point cloud to achieve optimal registration.
Therefore, it is necessary to determine an optimal fine registration matrix among the plurality of fine registration matrices obtained. Referring to fig. 11, in this embodiment, this step may be implemented by:
s151, for each fine registration matrix, obtaining a fine registration point cloud after fine registration of the initial registration point cloud based on the fine registration matrix.
S152, calculating an error value between the accurate point cloud and the target point cloud.
And S153, obtaining the minimum error value in the error values, and taking the fine registration matrix corresponding to the minimum error value as an optimal fine registration matrix.
In this embodiment, it is assumed that the initial alignment point cloud is rotated according to 6 different rotation angles, and the obtained rotated initial alignment point cloud is marked as Q' 1 、Q’ 2 、…Q’ 6 . After the accurate registration matrix is calculated based on the initial registration point clouds under each rotation angle, each initial registration point cloud is accurately registered, and the obtained accurate registration point clouds can be marked as Q '' 1 、Q’’ 2 、…Q’’ 6 。
The error value between each fine registration point cloud and the target point cloud is calculated as follows, and referring to fig. 12 in combination:
s1521, aiming at each three-dimensional point in the fine registration point cloud, obtaining a three-dimensional point corresponding to the three-dimensional point in the target point cloud.
S1522, calculating the distance between the accurate point cloud and the corresponding three-dimensional point in the target point cloud.
S1523, calculating to obtain an error value between the accurate registration point cloud and the target point cloud according to the distance corresponding to each three-dimensional point in the accurate registration point cloud and the number of the three-dimensional points.
In this embodiment, the precision point cloud Q 'is traversed' i And determining the corresponding three-dimensional points in the target point cloud, and calculating the distance between the three-dimensional points.
Fine registration point cloud Q' i The sum of the distances obtained for all three-dimensional points in (a) constitutes the total distance, divided by the fine registration point cloud Q '' i The number of three-dimensional points in the model can obtain an error value, and the error value can be marked as e.
Taking the above example, six error values e can be obtained 1 、e 2 …e 6 And selecting the minimum error value, wherein the fine registration matrix corresponding to the minimum error value is the optimal fine registration matrix.
From the above, in order to more fully consider the registration situation, the primary registration point cloud is turned over. In this embodiment, the above operations are performed by the primary registration point clouds before and after the flipping process, assuming that the minimum error value obtained based on the primary registration point clouds before the flipping process is e i The minimum error value obtained based on the primary registration point cloud after the overturn processing is e j . Then compare e i And e j And taking the fine registration matrix corresponding to the smaller value of the two as the optimal fine registration matrix.
Wherein the optimal fine registration matrix comprises a rotation matrix and a translation vector, and the obtained primary registration matrix also comprises the rotation matrix and the translation vector. In the step of combining the primary registration matrix and the optimal fine registration matrix to obtain a final registration matrix, the rotation matrix in the optimal fine registration matrix may be multiplied by the rotation matrix in the primary registration matrix to obtain the final rotation matrix. The rotation matrix in the optimal fine registration matrix is multiplied by the translation vector in the primary registration matrix, and then the translation vector in the optimal fine registration matrix is added to obtain a final translation vector, which is specifically expressed as follows:
Wherein,R final 、R fine 、R coarse representing the final rotation matrix, the rotation matrix in the optimal fine registration matrix and the rotation matrix in the primary registration matrix respectively,T final 、T fine 、T coarse representing the final translation vector, the translation vector in the optimal fine registration matrix, and the translation vector in the primary registration matrix, respectively.
And forming a final registration matrix by the final rotation matrix and the final translation vector, so that the point cloud to be registered can be registered to the target point cloud based on the final registration matrix.
In the point cloud registration scheme based on cylinder detection provided in this embodiment, the cylinders in the point cloud are used to perform point cloud registration, and the final rotation matrix and translation vector are determined by aligning the axes of the cylinders and rotating around the axes by a certain angle and performing fine registration. When the cylinder is detected, a mode that a projection plane perpendicular to the axis of the cylinder in the detection point cloud is adopted, and then circular detection is carried out on the projection plane is adopted.
In addition, when primary registration is carried out, registration is realized based on the cylinder with the largest radius in the two point clouds, so that the cylinder based on the two point clouds is ensured to be the same cylinder. Moreover, since the cylinder detection is realized based on a voting mechanism, the most points on the curved surface of the cylinder with the largest radius are most likely to be detected first, and the stage of detecting the cylinder also assists in determining the corresponding relation of the cylinders in two point clouds.
Based on the same inventive concept, please refer to fig. 13, which is a schematic diagram illustrating functional modules of a point cloud registration device based on cylinder detection according to an embodiment of the present application, where the functional modules of the point cloud registration device based on cylinder detection may be divided according to the above-mentioned method embodiment. For example, each functional module may be divided corresponding to each function, or two or more functions may be integrated in one processing module. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present application, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.
For example, in the case of dividing each functional module by corresponding each function, the point cloud registration apparatus based on cylinder detection shown in fig. 13 is only one apparatus schematic. The point cloud registration device based on cylinder detection can comprise an acquisition module, a detection module, a primary registration module, a rotation module and a calculation module, and the functions of each functional module of the point cloud registration device based on cylinder detection are respectively described in detail below.
The acquisition module is used for acquiring the point cloud to be registered and the target point cloud;
it will be appreciated that the acquisition module may be used to perform step S11 described above, and reference may be made to the details of step S11 for a detailed implementation of the acquisition module.
The detection module is used for carrying out cylinder detection on the point cloud to be aligned and carrying out cylinder detection on the target point cloud;
it will be appreciated that the detection module may be used to perform step S12 described above, and reference may be made to the details of step S12 for a detailed implementation of the detection module.
The primary registration module is used for calculating to obtain a primary registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the primary registration matrix to obtain a primary registration point cloud;
it will be appreciated that the primary registration module may be used to perform step S13 described above, and reference may be made to step S13 for details regarding implementation of the primary registration module.
The rotation module is used for rotating the initial alignment point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the initial alignment point cloud and the target point cloud under each rotation angle;
It will be appreciated that the rotation module may be used to perform step S14 described above, and reference may be made to the details of step S14 described above for a detailed implementation of the rotation module.
The calculation module is used for obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
It will be appreciated that the calculation module may be used to perform step S15 described above, and reference may be made to the details of step S15 for a detailed implementation of the calculation module.
In one possible implementation, the initial registration module may be configured to:
determining the cylinder to be registered with the largest radius in all cylinders in the point cloud to be registered, and determining the target cylinder with the largest radius in all cylinders in the target point cloud;
and calculating to obtain an initial registration matrix based on the cylinder to be registered and the target cylinder.
In one possible implementation, the initial registration module may be configured to:
obtaining a first direction vector of the axis of the cylinder to be registered and obtaining a second direction vector of the axis of the target cylinder;
determining a first center of gravity point of the cylinder to be registered based on the three-dimensional points of the cylinder to be registered in the point cloud to be registered, and determining a second center of gravity point of the target cylinder based on the three-dimensional points of the target cylinder in the target point cloud;
And calculating according to the first direction vector, the second direction vector, the first gravity center point and the second gravity center point to obtain an initial registration matrix.
In one possible implementation, the rotation module may be configured to:
turning the primary alignment point cloud around the centroid of the cylinder and 180 degrees in the axial direction of the cylinder;
and respectively rotating the initial registration point clouds before and after overturning around the axis of the cylinder according to a plurality of different rotation angles.
In one possible implementation, the computing module may be configured to:
for each fine registration matrix, obtaining a fine registration point cloud after fine registration of the initial registration point cloud based on the fine registration matrix;
calculating an error value between the accurate point cloud and the target point cloud;
and obtaining the minimum error value in the error values, and taking the fine registration matrix corresponding to the minimum error value as an optimal fine registration matrix.
In one possible implementation, the computing module may be configured to:
aiming at each three-dimensional point in the fine registration point cloud, obtaining a three-dimensional point corresponding to the three-dimensional point in the target point cloud;
calculating the distance between the accurate point cloud and the corresponding three-dimensional point in the target point cloud;
And calculating to obtain an error value between the accurate registration point cloud and the target point cloud according to the distance corresponding to each three-dimensional point in the accurate registration point cloud and the number of the three-dimensional points.
In one possible implementation manner, the detection module may be used for:
detecting the axis of a cylinder in the point cloud to be registered, and determining a projection plane perpendicular to the axis;
projecting the three-dimensional points in the point cloud to be registered onto the projection surface to obtain projection points;
and detecting a circle based on the projection point on the projection surface, and determining a corresponding cylinder based on the circle parameter of the detected circle and the axis parameter of the axis.
In one possible implementation manner, the detection module may be used for:
performing straight line detection based on the projection points on the projection surface, and marking the detected projection points on the straight line as invalid points, wherein the invalid points do not participate in subsequent circle detection;
and detecting the circles of the projection points which are not marked as invalid points on the projection surface.
In one possible implementation manner, the detection module may be used for:
selecting a plurality of projection points from the projection points which are not marked as invalid points on the projection surface at random each time, and calculating to obtain a circle parameter based on the plurality of projection points;
And when the calculated circle parameters of the projection points selected by the times exceeding the preset times are consistent, determining that the circle corresponding to the circle parameters is detected.
Referring to fig. 14, a block diagram of an electronic device provided in an embodiment of the present application may be a computer device, etc., where the electronic device includes a memory, a processor, and a communication module. The memory, the processor and the communication module are electrically connected with each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines.
Wherein the memory is used for storing programs or data. The Memory may be, but is not limited to, random access Memory (Random Access Memory, RAM), read Only Memory (ROM), programmable Read Only Memory (Programmable Read-Only Memory, PROM), erasable Read Only Memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable Read Only Memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.
The processor is used for reading/writing data or programs stored in the memory and executing the point cloud registration method based on cylinder detection provided by any embodiment of the application.
The communication module is used for establishing communication connection between the electronic equipment and other communication terminals through a network and is used for receiving and transmitting data through the network.
It should be understood that the structure shown in fig. 14 is merely a schematic structural diagram of an electronic device, and that the electronic device may also include more or fewer components than those shown in fig. 14, or have a different configuration than that shown in fig. 14.
Further, the embodiment of the application also provides a computer readable storage medium, and the computer readable storage medium stores machine executable instructions, which when executed implement the point cloud registration method based on cylinder detection provided in the above embodiment.
In particular, the computer readable storage medium can be a general purpose storage medium, such as a mobile disk, a hard disk, etc., and when the computer program on the computer readable storage medium is executed, the above-mentioned point cloud registration method based on cylinder detection can be performed. With respect to the processes in the computer readable storage medium and the executable instructions thereof involved when executed, reference is made to the relevant descriptions of the above method embodiments, which are not described in detail herein.
In summary, according to the method and the device for registering point clouds based on cylinder detection provided by the embodiments of the present application, cylinder detection is performed on the acquired point clouds to be registered and the target point clouds, an initial registration matrix is obtained based on cylinder calculation in the point clouds to be registered and the target point clouds, and registration is performed on the point clouds to be registered based on the initial registration matrix to obtain an initial registration point cloud. And rotating the initial registration point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating a fine registration matrix between the initial registration point cloud and the target point cloud under each rotation angle. And obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix. In the scheme, primary registration is realized based on the detected cylinder, and fine registration is realized based on the point clouds under a plurality of different rotation angles.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A point cloud registration method based on cylinder detection, the method comprising:
acquiring a point cloud to be registered and a target point cloud;
performing cylinder detection on the point cloud to be aligned, and performing cylinder detection on the target point cloud;
calculating to obtain an initial registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the initial registration matrix to obtain an initial registration point cloud;
rotating the primary registration point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the primary registration point cloud and the target point cloud under each rotation angle;
and obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
2. The method for point cloud registration based on cylinder detection according to claim 1, wherein the step of calculating an initial registration matrix based on cylinders in the point cloud to be registered and the target point cloud comprises:
determining the cylinder to be registered with the largest radius in all cylinders in the point cloud to be registered, and determining the target cylinder with the largest radius in all cylinders in the target point cloud;
and calculating to obtain an initial registration matrix based on the cylinder to be registered and the target cylinder.
3. The method for point cloud registration based on cylinder detection according to claim 2, wherein the step of calculating an initial registration matrix based on the cylinder to be registered and a target cylinder comprises:
obtaining a first direction vector of the axis of the cylinder to be registered and obtaining a second direction vector of the axis of the target cylinder;
determining a first center of gravity point of the cylinder to be registered based on the three-dimensional points of the cylinder to be registered in the point cloud to be registered, and determining a second center of gravity point of the target cylinder based on the three-dimensional points of the target cylinder in the target point cloud;
And calculating according to the first direction vector, the second direction vector, the first gravity center point and the second gravity center point to obtain an initial registration matrix.
4. The method of cylinder detection-based point cloud registration of claim 2, wherein the step of rotating the primary registration point cloud about the axis of the cylinder by a plurality of different rotation angles comprises:
turning the primary alignment point cloud around the centroid of the cylinder and 180 degrees in the axial direction of the cylinder;
and respectively rotating the initial registration point clouds before and after overturning around the axis of the cylinder according to a plurality of different rotation angles.
5. The method of cylinder detection-based point cloud registration according to claim 1, wherein the step of obtaining an optimal fine registration matrix of a plurality of fine registration matrices comprises:
for each fine registration matrix, obtaining a fine registration point cloud after fine registration of the initial registration point cloud based on the fine registration matrix;
calculating an error value between the accurate point cloud and the target point cloud;
and obtaining the minimum error value in the error values, and taking the fine registration matrix corresponding to the minimum error value as an optimal fine registration matrix.
6. The cylinder detection-based point cloud registration method of claim 5, wherein the step of calculating an error value between the precisely-registered point cloud and the target point cloud comprises:
aiming at each three-dimensional point in the fine registration point cloud, obtaining a three-dimensional point corresponding to the three-dimensional point in the target point cloud;
calculating the distance between the accurate point cloud and the corresponding three-dimensional point in the target point cloud;
and calculating to obtain an error value between the accurate registration point cloud and the target point cloud according to the distance corresponding to each three-dimensional point in the accurate registration point cloud and the number of the three-dimensional points.
7. The method for registering point cloud based on cylinder detection according to claim 1, wherein the step of performing cylinder detection on the point cloud to be registered comprises:
detecting the axis of a cylinder in the point cloud to be registered, and determining a projection plane perpendicular to the axis;
projecting the three-dimensional points in the point cloud to be registered onto the projection surface to obtain projection points;
and detecting a circle based on the projection point on the projection surface, and determining a corresponding cylinder based on the circle parameter of the detected circle and the axis parameter of the axis.
8. The method of cylinder detection-based point cloud registration as recited in claim 7, wherein said step of performing circle detection based on projected points on said projection surface comprises:
performing straight line detection based on the projection points on the projection surface, and marking the detected projection points on the straight line as invalid points, wherein the invalid points do not participate in subsequent circle detection;
and detecting the circles of the projection points which are not marked as invalid points on the projection surface.
9. The method of cylinder detection-based point cloud registration as recited in claim 8, wherein the step of performing circle detection on projected points on the projection surface that are not marked as invalid points comprises:
selecting a plurality of projection points from the projection points which are not marked as invalid points on the projection surface at random each time, and calculating to obtain a circle parameter based on the plurality of projection points;
and when the calculated circle parameters of the projection points selected by the times exceeding the preset times are consistent, determining that the circle corresponding to the circle parameters is detected.
10. A point cloud registration device based on cylinder detection, the device comprising:
the acquisition module is used for acquiring the point cloud to be registered and the target point cloud;
The detection module is used for carrying out cylinder detection on the point cloud to be aligned and carrying out cylinder detection on the target point cloud;
the primary registration module is used for calculating to obtain a primary registration matrix based on the cylinder in the point cloud to be registered and the target point cloud, and registering the point cloud to be registered based on the primary registration matrix to obtain a primary registration point cloud;
the rotation module is used for rotating the initial alignment point cloud around the axis of the cylinder according to a plurality of different rotation angles, and calculating to obtain a fine registration matrix between the initial alignment point cloud and the target point cloud under each rotation angle;
the calculation module is used for obtaining an optimal fine registration matrix in the plurality of fine registration matrices, and combining the initial registration matrix and the optimal fine registration matrix to obtain a final registration matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410269026.XA CN117876446B (en) | 2024-03-11 | 2024-03-11 | Point cloud registration method and device based on cylinder detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410269026.XA CN117876446B (en) | 2024-03-11 | 2024-03-11 | Point cloud registration method and device based on cylinder detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117876446A true CN117876446A (en) | 2024-04-12 |
| CN117876446B CN117876446B (en) | 2024-05-24 |
Family
ID=90594753
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410269026.XA Active CN117876446B (en) | 2024-03-11 | 2024-03-11 | Point cloud registration method and device based on cylinder detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117876446B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118134979A (en) * | 2024-05-07 | 2024-06-04 | 法奥意威(苏州)机器人***有限公司 | Point cloud registration method and device based on single plane characteristics and electronic equipment |
| CN118196109A (en) * | 2024-05-20 | 2024-06-14 | 法奥意威(苏州)机器人***有限公司 | Arc-shaped weld joint identification method and device |
| CN118196172A (en) * | 2024-05-20 | 2024-06-14 | 法奥意威(苏州)机器人***有限公司 | Cylinder axis detection method and device, storage medium and electronic equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140244219A1 (en) * | 2013-02-28 | 2014-08-28 | Autodesk, Inc. | Method of creating a pipe route line from a point cloud in three-dimensional modeling software |
| KR101708042B1 (en) * | 2015-12-01 | 2017-02-21 | 중앙대학교 산학협력단 | Cylinder estimation device and method by ransac algorithm in point cloud |
| CN107578400A (en) * | 2017-07-26 | 2018-01-12 | 西南交通大学 | A kind of contact net device parameter detection method of BIM and three-dimensional point cloud fusion |
| CN110276790A (en) * | 2019-06-28 | 2019-09-24 | 易思维（杭州）科技有限公司 | Point cloud registration method based on shape constraining |
| CN110443836A (en) * | 2019-06-24 | 2019-11-12 | 中国人民解放军战略支援部队信息工程大学 | A kind of point cloud data autoegistration method and device based on plane characteristic |
| CN113052881A (en) * | 2021-03-24 | 2021-06-29 | 王程 | Automatic registration method for extracting pole point in indoor three-dimensional point cloud |
| CN113643270A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Image registration method and device based on point cloud data |
| CN117173227A (en) * | 2023-11-01 | 2023-12-05 | 法奥意威(苏州)机器人***有限公司 | Point cloud registration method and device based on plane fitting and electronic equipment |
-
2024
- 2024-03-11 CN CN202410269026.XA patent/CN117876446B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140244219A1 (en) * | 2013-02-28 | 2014-08-28 | Autodesk, Inc. | Method of creating a pipe route line from a point cloud in three-dimensional modeling software |
| KR101708042B1 (en) * | 2015-12-01 | 2017-02-21 | 중앙대학교 산학협력단 | Cylinder estimation device and method by ransac algorithm in point cloud |
| CN107578400A (en) * | 2017-07-26 | 2018-01-12 | 西南交通大学 | A kind of contact net device parameter detection method of BIM and three-dimensional point cloud fusion |
| CN110443836A (en) * | 2019-06-24 | 2019-11-12 | 中国人民解放军战略支援部队信息工程大学 | A kind of point cloud data autoegistration method and device based on plane characteristic |
| CN110276790A (en) * | 2019-06-28 | 2019-09-24 | 易思维（杭州）科技有限公司 | Point cloud registration method based on shape constraining |
| CN113052881A (en) * | 2021-03-24 | 2021-06-29 | 王程 | Automatic registration method for extracting pole point in indoor three-dimensional point cloud |
| CN113643270A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Image registration method and device based on point cloud data |
| CN117173227A (en) * | 2023-11-01 | 2023-12-05 | 法奥意威(苏州)机器人***有限公司 | Point cloud registration method and device based on plane fitting and electronic equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118134979A (en) * | 2024-05-07 | 2024-06-04 | 法奥意威(苏州)机器人***有限公司 | Point cloud registration method and device based on single plane characteristics and electronic equipment |
| CN118196109A (en) * | 2024-05-20 | 2024-06-14 | 法奥意威(苏州)机器人***有限公司 | Arc-shaped weld joint identification method and device |
| CN118196172A (en) * | 2024-05-20 | 2024-06-14 | 法奥意威(苏州)机器人***有限公司 | Cylinder axis detection method and device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117876446B (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117876446B (en) | Point cloud registration method and device based on cylinder detection | |
| CN107810522B (en) | Real-time, model-based object detection and pose estimation | |
| US20150146964A1 (en) | Inspection apparatus, method, and computer program product for machine vision inspection | |
| GB2526342A (en) | Point cloud matching method | |
| CN112085798B (en) | Camera calibration method and device, electronic equipment and storage medium | |
| KR20140014298A (en) | Planar mapping and tracking for mobile devices | |
| CN113643270B (en) | Image registration method and device based on point cloud data | |
| CN117173227B (en) | Point cloud registration method and device based on plane fitting and electronic equipment | |
| Wujanz et al. | Identification of stable areas in unreferenced laser scans for deformation measurement | |
| CN112686950B (en) | Pose estimation method, pose estimation device, terminal equipment and computer readable storage medium | |
| CN111882657B (en) | Three-dimensional reconstruction scale recovery method, device, system and computer equipment | |
| CN114120149B (en) | Oblique photogrammetry building feature point extraction method and device, electronic equipment and medium | |
| CN113362446B (en) | Method and device for reconstructing object based on point cloud data | |
| CN113077521A (en) | Camera calibration method and device | |
| CN117314987B (en) | Point cloud registration method and device | |
| CN113407030B (en) | Visual positioning method, related device, equipment and storage medium | |
| CN112631200A (en) | Machine tool axis measuring method and device | |
| JP3849228B2 (en) | 3D position / orientation detection device for circular feature of parts, detection method thereof, and recording medium thereof | |
| CN108801226B (en) | Plane inclination testing method and equipment | |
| CN115205558B (en) | Multi-mode image matching method and device with rotation and scale invariance | |
| Wang et al. | A binocular vision method for precise hole recognition in satellite assembly systems | |
| CN111633337B (en) | Reflection eliminating method and device for laser welding seam measurement | |
| CN116295466A (en) | Map generation method, map generation device, electronic device, storage medium and vehicle | |
| CN111223139A (en) | Target positioning method and terminal equipment | |
| CN112763974B (en) | Sensor deployment method, device and medium in positioning system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |