阅读说明：本技术 一种激光点云特征的提取方法 (Method for extracting laser point cloud characteristics ) 是由 纪刚 朱慧 杨丰拓 安帅 柴琦 蒋云飞 于 2020-04-01 设计创作，主要内容包括：本发明公开了一种激光点云特征的提取方法,包括如下步骤：选取中心点P<Sub>c</Sub>,以该点为球心,r为半径,形成球形区域S<Sub>1</Sub>；求S<Sub>1</Sub>内所有邻域点组成的多面体的质心P<Sub>c</Sub>’的坐标；在S<Sub>1</Sub>内重新划分一个同心的球形区域S<Sub>2</Sub>,S<Sub>2</Sub>的半径小于球形区域S<Sub>1</Sub>的半径,并求S<Sub>2</Sub>内所有邻域点组成的多面体的质心P<Sub>c</Sub>”的坐标；在P<Sub>c</Sub>、P<Sub>c</Sub>’、P<Sub>c</Sub>”三个点组成的平面内建立坐标系,坐标系原点为P<Sub>c</Sub>点；将球形区域S<Sub>1</Sub>内所有点的坐标都转换到新建的坐标系下,然后将所有的邻域点分别和球心P<Sub>c</Sub>连接组成向量,将所有的向量进行矢量叠加,求得的矢量即为中心点P<Sub>c</Sub>的特征。本发明所公开的方法能够快速提取出点云的特征,并且对于初始值的选定不是非常敏感,提高了鲁棒性,对于不同场景具有较好的适应性。(The invention discloses a method for extracting laser point cloud characteristics, which comprises the following steps: selecting a center point P c Taking the point as the center of the sphere and r as the radius, a spherical area S is formed 1 (ii) a Calculating S 1 Centroid P of polyhedron formed by all neighborhood points in polyhedron c The coordinates of'; at S 1 Inner subdivision into a concentric spherical region S 2 ，S 2 Is smaller than the spherical area S 1 And calculating S 2 Centroid P of polyhedron formed by all neighborhood points in polyhedron c "of the coordinates; at P c 、P c '、P c ' A coordinate system is established in a plane consisting of three points, and the origin of the coordinate system is P c Point; a spherical area S 1 All the coordinates of the inner points are converted into a newly-built coordinate system, and then all the neighborhood points are respectively matched with the sphere center P c Connecting to form vectors, and performing vector superposition on all the vectors to obtain a vector which is the central point P c The characteristics of (1). The method disclosed by the invention can be used for quickly extracting the characteristics of the point cloud, is not very sensitive to the selection of the initial value, improves the robustness and has better adaptability to different scenes.)

1. The method for extracting the laser point cloud features is characterized by comprising the following steps of:

step one, selecting a central point P_cTaking the point as the center of the sphere and r as the radius, a spherical area S is formed₁(ii) a Finding a spherical region S₁Centroid P of polyhedron formed by all neighborhood points in polyhedron_cThe coordinates of';

step two, in the spherical area S₁Inner subdivision into a concentric spherical region S₂Spherical region S₂Is smaller than the spherical area S₁And finding the spherical area S₂Centroid P of polyhedron formed by all neighborhood points in polyhedron_c"of the coordinates;

step three, in P_c、P_c'、P_c' A coordinate system is established in a plane consisting of three points, and the origin of the coordinate system is P_cPoint, the direction of the X-axis being point P_cPoint of direction P_cThe direction of the' Y axis is perpendicular to the X axis and close to P_c"the direction, according to X, Y axis to establish the right hand coordinate system, the direction of Z axis is also uniquely determined;

step four, the spherical area S is processed₁All the coordinates of the inner points are converted into a newly-built coordinate system, and then all the neighborhood points are respectively matched with the sphere center P_cConnecting to form vectors, and performing vector superposition on all the vectors to obtain a vector which is the central point P_cThe characteristics of (1).

2. The method for extracting laser point cloud features of claim 1, wherein in the first step, P is_nIs the center of sphere P_c(x₁,y₁,z₁) All neighborhood points of (1) are set as the ith neighborhood point P_niHas the coordinates of (x)_ni,y_ni,z_ni) Then the centroid P_c'(x₂,y₂,z₂) The coordinates of (a) are:

where k is the neighborhood point P_nThe number of (2).

3. The method for extracting laser point cloud features of claim 2, wherein in the second step, a spherical area S is initially set₂Has a radius of a spherical region S₁Is a half of the radius of (A), is provided with a spherical area S₂Inner j neighborhood point P_njHas the coordinates of (x)_nj,y_nj,z_nj) Then the centroid P_c”(x₃,y₃,z₃) The coordinates of (a) are:

wherein m is a spherical region S₂Number of neighborhood points contained in, if m<3, then enlarging the spherical area S₂Until m ≧ 3.

4. The method for extracting the laser point cloud features of claim 3, wherein three point coordinates in the third step are as follows: p_c(x₁,y₁,z₁)，P_c'(x₂,y₂,z₂)，P_c”(x₃,y₃,z₃) Make a vector The normal to the plane is perpendicular to the two vectors, so the normal vector

a＝(y₂-y₁)(z₃-z₁)-(y₃-y₁)(z₂-z₁)

b＝(x₃-x₁)(z₂-z₁)-(x₂-y₁)(z₃-z₁)

c＝(x₂-x₁)(y₃-y₁)-(x₃-x₁)(y₂-y₁)

The plane equation is then:

a(x-x₁)+b(y-y₁)+c(z-z₁)＝0

the method is simplified as follows:

ax+by+cz+d＝0

wherein:

d＝-a·x₁-b·y₁-c·z₁

let the origin of the coordinate system be P_cThe direction vector of the X axis isThe direction vector of the Z axis isThe direction vector of the Y axis is:thus, the new coordinate system establishment is completed.

5. The method for extracting the laser point cloud features of claim 4, wherein in the fourth step, the coordinates of the point P in the new coordinate system are as follows:

ⁿP＝T^oP

wherein, the corner mark n represents the coordinate under the new coordinate system, the corner mark o represents the coordinate of the original coordinate system, and T is a transformation matrix:

the conversion of the point to the new coordinate system can be written as:

then, all neighborhood points P are set under the coordinate system_nRespectively with the center of sphere P_cConnected together to form k vectors, and the k vectors are vector-superposed again to obtain the final vector tau which is the desired point P_cThe characteristics of (1).

Technical Field

The invention relates to a method for extracting laser point cloud characteristics.

Background

The point cloud is a data source widely used in many fields such as machine vision, three-dimensional reconstruction, machine learning and the like at present, and if the point cloud data is to be fully utilized, extraction of point cloud features is very important.

The existing method for extracting the point cloud features, such as PFH, is large in calculation amount and is not suitable for scenes with high real-time requirements, such as SLAM and the like, an improved algorithm FPFH reduces calculation complexity, but the method is essentially a histogram and describes the angle relationship between points, a probability histogram is finally counted, distance information is not available, and the problem of the scale of the point cloud is solved. And many current algorithms do not have rotational invariance.

Disclosure of Invention

In order to solve the technical problems, the invention provides a method for extracting laser point cloud features, so as to achieve the purposes of quickly extracting the features of the point cloud, not being very sensitive to the selection of an initial value and having better adaptability to different scenes.

In order to achieve the purpose, the technical scheme of the invention is as follows:

a method for extracting laser point cloud features comprises the following steps:

step one, selecting a central point P_cTaking the point as the center of the sphere and r as the radius, a spherical area S is formed₁(ii) a Finding a spherical region S₁Centroid P of polyhedron formed by all neighborhood points in polyhedron_cThe coordinates of';

step two, in the spherical area S₁Inner subdivision into a concentric spherical region S₂Ball, ballShaped region S₂Is smaller than the spherical area S₁And finding the spherical area S₂Centroid P of polyhedron formed by all neighborhood points in polyhedron_c"of the coordinates;

step three, in P_c、P_c'、P_c' A coordinate system is established in a plane consisting of three points, and the origin of the coordinate system is P_cPoint, the direction of the X-axis being point P_cPoint of direction P_cThe direction of the' Y axis is perpendicular to the X axis and close to P_c"the direction, according to X, Y axis to establish the right hand coordinate system, the direction of Z axis is also uniquely determined;

step four, the spherical area S is processed₁All the coordinates of the inner points are converted into a newly-built coordinate system, and then all the neighborhood points are respectively matched with the sphere center P_cConnecting to form vectors, and performing vector superposition on all the vectors to obtain a vector which is the central point P_cThe characteristics of (1).

In the above scheme, in step one, P_nIs the center of sphere P_c(x₁,y₁,z₁) All neighborhood points of (1) are set as the ith neighborhood point P_niHas the coordinates of (x)_ni,y_ni,z_ni) Then the centroid P_c'(x₂,y₂,z₂) The coordinates of (a) are:

where k is the neighborhood point P_nThe number of (2).

In the above scheme, in the second step, the spherical area S is initially set₂Has a radius of a spherical region S₁Is a half of the radius of (A), is provided with a spherical area S₂Inner j neighborhood point P_njHas the coordinates of (x)_nj,y_nj,z_nj) Then the centroid P_c”(x₃,y₃,z₃) The coordinates of (a) are:

wherein m is a ballShaped region S₂Number of neighborhood points contained in, if m<3, then enlarging the spherical area S₂Until m ≧ 3.

In the above scheme, in the third step, the coordinates of the three points are as follows: p_c(x₁,y₁,z₁)，P_c'(x₂,y₂,z₂)，P_c”(x₃,y₃,z₃) Make a vectorThe normal to the plane is perpendicular to the two vectors, so the normal vector

a＝(y₂-y₁)(z₃-z₁)-(y₃-y₁)(z₂-z₁)

b＝(x₃-x₁)(z₂-z₁)-(x₂-y₁)(z₃-z₁)

c＝(x₂-x₁)(y₃-y₁)-(x₃-x₁)(y₂-y₁)

The plane equation is then:

a(x-x₁)+b(y-y₁)+c(z-z₁)＝0

the method is simplified as follows:

ax+by+cz+d＝0

wherein:

d＝-a·x₁-b·y₁-c·z₁

let the origin of the coordinate system be P_cThe direction vector of the X axis isThe direction vector of the Z axis isThe direction vector of the Y axis is:thus, the new coordinate system establishment is completed.

In the above scheme, in the fourth step, the coordinates of the point P in the new coordinate system are:

ⁿP＝T^oP

wherein, the corner mark n represents the coordinate under the new coordinate system, the corner mark o represents the coordinate of the original coordinate system, and T is a transformation matrix:

the conversion of the point to the new coordinate system can be written as:

then, all neighborhood points P are set under the coordinate system_nRespectively with the center of sphere P_cConnected together to form k vectors, and the k vectors are vector-superposed again to obtain the final vector tau which is the desired point P_cThe characteristics of (1).

Through the technical scheme, the method for extracting the laser point cloud characteristics has the following beneficial effects:

the method is different from the characteristic point characterization mode of the histogram of PFH and FPFH, and the method characterizes the characteristics by directly constructing a space vector mode. The centroid of the two spherical areas is calculated, a three-dimensional space coordinate system is constructed by the centroid, and new vectors are constructed by the vectors of all the points in the coordinate system, so that the characteristics of the current point are represented, the rotation invariance is ensured, the influence of noise points is reduced, and the robustness is improved. The method can quickly extract the characteristics of the point cloud, is not very sensitive to the selection of the initial value, and has good adaptability to different scenes.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.

Fig. 1 is a schematic flow chart of a method for extracting laser point cloud features disclosed in the embodiment of the present invention;

FIG. 2 is a view of a spherical region S established₁A schematic diagram of (a);

FIG. 3 shows a spherical region S established₂And a schematic of a new coordinate system;

fig. 4 is a schematic diagram of vector superposition.

Detailed Description

The technical solution 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.

The invention provides a method for extracting laser point cloud characteristics, which comprises the following steps of:

step one, as shown in FIG. 2, select a center point P_cTaking the point as the center of the sphere and r as the radius, a spherical area S is formed₁(ii) a Finding a spherical region S₁Centroid P of polyhedron formed by all neighborhood points in polyhedron_cThe coordinates of';

P_nis the center of sphere P_c(x₁,y₁,z₁) All neighborhood points of (1) are set as the ith neighborhood point P_niHas the coordinates of (x)_ni,y_ni,z_ni) Then the centroid P_c'(x₂,y₂,z₂) The coordinates of (a) are:

where k is the neighborhood point P_nThe number of (2).

Step two, as shown in FIG. 3, in the spherical region S₁Inner subdivision into a concentric spherical region S₂Spherical region S₂Is smaller than the spherical area S₁And finding the spherical area S₂Centroid P of polyhedron formed by all neighborhood points in polyhedron_c"of the coordinates;

initially setting a spherical region S₂Has a radius of a spherical region S₁Is a half of the radius of (A), is provided with a spherical area S₂Inner j neighborhood point P_njHas the coordinates of (x)_nj,y_nj,z_nj) Then the centroid P_c”(x₃,y₃,z₃) The coordinates of (a) are:

wherein m is a spherical region S₂Number of neighborhood points contained in, if m<3, then enlarging the spherical area S₂Until m ≧ 3.

Step three, in P_c、P_c'、P_c' A coordinate system is established in a plane consisting of three points, and the origin of the coordinate system is P_cPoint, the direction of the X-axis being point P_cPoint of direction P_cThe direction of the' Y axis is perpendicular to the X axis and close to P_c"the direction, according to X, Y axis to establish the right hand coordinate system, the direction of Z axis is also uniquely determined;

the three-point coordinates are as follows: p_c(x₁,y₁,z₁)，P_c'(x₂,y₂,z₂)，P_c”(x₃,y₃,z₃) Make a vector The normal to the plane is perpendicular to the two vectors, so the normal vector

a＝(y₂-y₁)(z₃-z₁)-(y₃-y₁)(z₂-z₁)

b＝(x₃-x₁)(z₂-z₁)-(x₂-y₁)(z₃-z₁)

c＝(x₂-x₁)(y₃-y₁)-(x₃-x₁)(y₂-y₁)

The plane equation is then:

a(x-x₁)+b(y-y₁)+c(z-z₁)＝0

the method is simplified as follows:

ax+by+cz+d＝0

wherein:

d＝-a·x₁-b·y₁-c·z₁

let the origin of the coordinate system be P_cThe direction vector of the X axis isThe direction vector of the Z axis isThe direction vector of the Y axis is:thus, the new coordinate system establishment is completed.

Step four, the spherical area S is processed₁All the coordinates of the inner points are converted into a newly-built coordinate system, and then all the neighborhood points are respectively matched with the sphere center P_cConnecting and forming vectors, as shown in FIG. 4, performing vector superposition on all vectors, and obtaining the vector as the central point P_cThe characteristics of (1).

The coordinates of point P in the new coordinate system are:

ⁿP＝T^oP

wherein, the corner mark n represents the coordinate under the new coordinate system, the corner mark o represents the coordinate of the original coordinate system, and T is a transformation matrix:

the conversion of the point to the new coordinate system can be written as:

then, all neighborhood points P are set under the coordinate system_nRespectively with the center of sphere P_cConnected together to form k vectors, and the k vectors are vector-superposed again to obtain the final vector tau which is the desired point P_cThe characteristics of (1).

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.