CN113196336A

CN113196336A - Point cloud density quantification method and device and storage medium

Info

Publication number: CN113196336A
Application number: CN201980078380.1A
Authority: CN
Inventors: 王闯
Original assignee: SZ DJI Technology Co Ltd
Current assignee: SZ DJI Technology Co Ltd
Priority date: 2019-11-29
Filing date: 2019-11-29
Publication date: 2021-07-30
Also published as: WO2021102967A1

Abstract

The invention provides a point cloud density quantification method, a point cloud density quantification device and a storage medium, wherein the method comprises the following steps: aiming at point cloud data of point cloud density to be calculated, dividing an area where the point cloud data is located into a plurality of non-overlapping graphic units, wherein each graphic unit is composed of at least three adjacent point cloud points in the point cloud data; calculating a point cloud density characterization for each of the graphical units based on respective geometric parameters of the plurality of graphical units, the point cloud density characterization for each of the graphical units being monotonically related to at least one of the geometric parameters of itself; and calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density representation of each graphic unit. The point cloud density quantization scheme provided by the embodiment of the invention provides a graph calculation scheme capable of accurately quantizing the density and uniformity of point cloud data, so that a proper laser radar product can be selected according to actual needs based on the scheme to generate point cloud data meeting requirements.

Description

Point cloud density quantification method and device and storage medium

Description

Technical Field

The present invention relates generally to the field of laser detection technology, and more particularly, to a method, an apparatus, and a storage medium for quantifying point cloud density.

Background

The core function of a three-dimensional point cloud detection system including a laser radar is to output three-dimensional point cloud data, one of core indexes of the three-dimensional point cloud data is point cloud density, and the point cloud density can represent the distribution of points in a unit three-dimensional space. In general, the higher the point cloud density of a lidar at the same time, the higher the resolution and efficiency of its three-dimensional measurement. Another indicator of three-dimensional point cloud data is point cloud uniformity, which can characterize the consistency of point cloud density in space. Because the point clouds of the laser radar are not all arranged at equal intervals, the point cloud densities of the laser radar are different in different areas, and the point cloud densities of the laser radar in different areas have smaller difference, so that the uniformity of the point clouds can be considered to be higher. In most applications, the higher the uniformity of the lidar point cloud, the lower the probability of missing small objects during measurement.

Due to different design principles and scanning modes, the point cloud density and uniformity of different laser radar products are greatly different. The characterization of the point cloud density commonly used in the industry at present is a beam or a point number, the former represents how many laser scanning lines exist, and 16, 32, 40, or 64 laser scanning lines exist commonly, and the latter represents how many point numbers exist in a unit time. However, both of the two characterization methods are relatively rough, and the actual density and uniformity of the point cloud cannot be really and effectively evaluated, and the former method is only suitable for the multi-line scanning type laser radar. And if the accurate density and uniformity of the point cloud cannot be obtained, a proper laser radar product cannot be selected according to actual needs to generate point cloud data meeting the requirements.

Disclosure of Invention

In order to solve the above problem, embodiments of the present invention provide a point cloud density quantization scheme, which can accurately quantize the density of point cloud data and is applicable to point cloud data generated by various point cloud detection systems. The following briefly describes the quantization scheme of the point cloud density proposed by the present invention, and more details will be described in the following detailed description with reference to the accompanying drawings.

According to an aspect of the present invention, there is provided a method for quantifying a point cloud density, the method comprising: aiming at point cloud data of point cloud density to be calculated, dividing an area where the point cloud data is located into a plurality of non-overlapping graphic units, wherein each graphic unit is composed of at least three adjacent point cloud points in the point cloud data; calculating a point cloud density characterization for each of the graphical units based on respective geometric parameters of the plurality of graphical units, the point cloud density characterization for each of the graphical units being monotonically related to at least one of the geometric parameters of itself; and calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density representation of each graphic unit.

According to another aspect of the present invention, there is provided an apparatus for quantifying a point cloud density, the apparatus comprising a memory and a processor, the memory having stored thereon a computer program for execution by the processor, the computer program, when executed by the processor, performing the method for quantifying a point cloud density as described above.

According to a further aspect of the present invention, a storage medium is provided, on which a computer program is stored, which computer program, when executed, performs the above method of quantifying a point cloud density.

The point cloud density quantification method, the point cloud density quantification device and the storage medium provided by the embodiment of the invention provide a graph calculation scheme capable of accurately quantifying the density of point cloud data, and the method can be suitable for point cloud data generated by various point cloud detection systems, so that not only can a proper laser radar product be selected to generate point cloud data meeting requirements according to actual needs based on the scheme, but also the point cloud density obtained through quantification has important reference values for testing and iteration of the laser radar product.

Drawings

Fig. 1 shows a schematic comparison of point cloud density and uniformity for different lidar.

Fig. 2 shows a schematic flow diagram of a method of quantifying a point cloud density according to an embodiment of the invention.

FIG. 3 shows a schematic diagram of a method for quantifying point cloud density according to an embodiment of the present invention, constructing a graphical unit for computing a point cloud density representation for a lidar point cloud array.

FIG. 4 shows a schematic flow diagram of a method of quantifying the uniformity of a point cloud in accordance with an embodiment of the present invention.

Fig. 5A to 5D are schematic diagrams illustrating a point cloud density quantization method according to an embodiment of the present invention, which constructs a graph unit for calculating a point cloud density representation for different lidar point cloud arrays.

Fig. 6 shows density distribution diagrams of four types of laser radars corresponding to fig. 5A to 5D.

Fig. 7 shows a schematic block diagram of a device for quantifying the density of a point cloud according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. It is to be understood that the described embodiments are merely a subset of embodiments of the invention and not all embodiments of the invention, with the understanding that the invention is not limited to the example embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the invention described herein without inventive step, shall fall within the scope of protection of the invention.

In the following description, numerous specific details are set forth in order to provide a more thorough understanding of the present invention. It will be apparent, however, to one skilled in the art, that the present invention may be practiced without one or more of these specific details. In other instances, well-known features have not been described in order to avoid obscuring the invention.

It is to be understood that the present invention may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. As used herein, the term "and/or" includes any and all combinations of the associated listed items.

In order to provide a thorough understanding of the present invention, detailed steps and detailed structures will be set forth in the following description in order to explain the present invention. The following detailed description of the preferred embodiments of the invention, however, the invention is capable of other embodiments in addition to those detailed.

The 3D laser radar is a perception system which utilizes laser to scan and measure distance so as to acquire three-dimensional information in surrounding scenes, and the basic principle is as follows: actively transmitting laser pulses to a detected object, capturing laser echo signals and calculating the distance of a detected object according to the time difference between the transmission and the reception of the laser; obtaining angle information of the measured object based on the known emission direction of the laser; by high-frequency transmission and reception, massive distance and angle information of the detection points can be acquired, and the information is called point cloud. Three-dimensional information of surrounding scenes can be reconstructed based on the point cloud. One of the core indicators of 3D lidar is the density of the point cloud (referred to herein simply as point cloud density), which is typically expressed in D_pThe representation may be defined as the number of points in the unit three-dimensional space. In general, the higher the point cloud density of a lidar at the same time, the higher the resolution and efficiency of its three-dimensional measurement. Another core indicator of 3D lidar is the uniformity of the point cloud (referred to herein simply as point cloud uniformity), which is commonly expressed in U_pRepresenting, characterizable point clouds in spaceConsistency of density. Because the point clouds of the laser radar are not all arranged at equal intervals, the point clouds of the laser radar are different in density in different areas, and the point clouds of the laser radar in different areas have smaller difference, so that the uniformity of the point clouds can be considered to be higher. In most applications, the higher the uniformity of the lidar point cloud, the lower the probability of missing small objects during measurement. Due to different design principles and scanning modes, the density and uniformity of point cloud of different laser radar products are greatly different. As shown in fig. 1, the upper half of fig. 1 shows image information acquired based on a 128-line lidar, and the lower half of fig. 1 shows image information acquired based on a 64-line lidar, which have different point cloud densities and point cloud uniformities.

The characterization of the point cloud density commonly used in the industry at present is a beam or a point number, the former represents how many laser scanning lines exist, and 16, 32, 40, or 64 laser scanning lines exist commonly; the latter represents how many points per unit time. The former is generally only suitable for multi-line scanning laser radar, but has the disadvantage of only having the total number, and no information such as the angle between lines and the number of points on the lines, so the method is very rough for estimating the density of point cloud. Even with the same line beam lidar, the actual point cloud densities may differ significantly. The latter represents the number of points in a unit time, but has the disadvantage that the distribution characteristics of the points cannot be represented only by the total number of points, and even if the laser radar with the same number of points has different distribution density, the density of some areas may be high, and the density of some areas is low, so that the reliability of the index is not high. Even if the line bundles and the point numbers are integrated, the space between the lines and the space between the points cannot be represented, so that the density and the uniformity of the point cloud cannot be accurately evaluated. In addition, there is also an evaluation method in which a plurality of uniform grids are divided in a space, a point cloud exists in the grids, the statistical value is added by 1, and the point cloud does not exist, and finally the total number of the grids with the point cloud is obtained for comparison. However, this method has the disadvantage that, for the same point cloud, the size of the mesh division has a great influence on the result, and even in comparison, the opposite conclusion is reached. Therefore, the method cannot really estimate the true and accurate point cloud density. Generally speaking, there is still no accurate method for evaluating the density and uniformity of the point cloud in the industry, which can satisfy the two characteristics of accurate quantification and universal applicability.

Based on the point cloud density quantization scheme, the density and the uniformity of point cloud data can be accurately quantized, and the point cloud density quantization scheme can be suitable for point cloud data generated by various point cloud detection systems. The following detailed description refers to the accompanying drawings.

FIG. 2 shows a schematic flow diagram of a method 200 of quantifying a point cloud density according to an embodiment of the invention. As shown in fig. 2, a method 200 for quantifying the density of a point cloud according to an embodiment of the present invention may include the following steps:

in step S210, for point cloud data of a point cloud density to be calculated, a region in which the point cloud data is located is divided into a plurality of non-overlapping graphic units, each graphic unit being composed of at least three point cloud points adjacent to each other in the point cloud data.

In step S220, a point cloud density characterization of each of the graphic units is calculated based on respective geometric parameters of the graphic units, and the point cloud density characterization of each of the graphic units is monotonically related to at least one of the geometric parameters of itself.

In step S230, the point cloud density of the point cloud data of the point cloud density to be calculated is calculated based on the point cloud density characterization of each graphic unit.

In the embodiment of the invention, the area where the point cloud data is located is firstly divided into a plurality of graphic units which are not overlapped with each other, the geometric characteristics of each graphic unit are calculated based on the geometric parameters of each graphic unit to be used as the point cloud density representation of the graphic unit, and the point cloud density of the whole point cloud array is calculated based on the point cloud density representation of each image unit. The method can provide accurate characterization of the point cloud density. This method is described in further detail below.

In the embodiment of the invention, the point cloud density and the point cloud uniformity are only related to the angle and the frequency of laser emission and are not related to the point cloud depth, so that in order to simplify the problem analysis process, the point cloud depths can be all set to be the same value D0, and all the point clouds are approximately distributed on a sphere with a laser radar as the sphere center.

In an embodiment of the present invention, the dividing the region where the point cloud data is located into a plurality of non-overlapping graphic units in step S210 may include: mapping the point cloud data to the same surface; and dividing an area of the point cloud data on the surface into a plurality of graphic units which do not overlap with each other. In this embodiment, the point cloud data is projected onto a surface (e.g., a two-dimensional plane or other curved surface) and the density of the point cloud is represented based on a graphical unit of point cloud points on the surface, thereby simplifying the calculation process.

Illustratively, three-dimensional point cloud data is projected onto a two-dimensional plane, forming a point cloud array on the two-dimensional plane, as shown in FIG. 3. For the 9 point cloud points shown in fig. 3, 3 adjacent point cloud points may be selected from the 9 point cloud points at a time to form a plurality of triangles that do not overlap with each other. For example, any point P1 in the point cloud may be selected, and two points P2 and P3 adjacent thereto may be selected in a certain order to form a triangle R1. Similarly, another triangle may be formed by points P1, P3, and P4, another triangle may be formed by points P2, P3, and P5, yet another triangle may be formed by points P3, P4, and P6, and so on (for brevity, the reference numerals and the triangles formed by all of these 9 points are not identified) until all cloud points are formed into triangles with their neighbors and the triangles do not overlap each other. In fig. 3, three point cloud points are described as an example to form a triangle, but it should be understood that this is merely exemplary, and in the embodiment of the present invention, four or more adjacent points may be selected to form a corresponding graphic unit as long as the formed graphic units do not overlap with each other.

Based on the plurality of graphic units obtained in step S210, a point cloud density characterization for each of the graphic units may be calculated based on respective geometric parameters of the plurality of graphic units, the point cloud density characterization for each of the graphic units being monotonically related to at least one of the geometric parameters thereof, as shown in step S220.

In one example, the point cloud density characterization for each of the graphical units may be equal to the square root of its area multiplied by its perimeter. Next to the example above with reference to fig. 3, for a triangle R1 composed of point cloud points P1, P2, and P3, the lengths of its three sides are a, b, and c, respectively, and the area is denoted as S. Since the point cloud includes coordinate data of each point, the area S of the triangle R1 is easily calculated according to the heleny formula, as shown in the following formula (1):

the point cloud density representation D of triangle R1 is then_pCan be represented by the following formula (2):

it is clear that the smaller the area S, the higher the point cloud density, under the same arrangement. In addition, under the same area S, the side lengths a, b and c of the triangle are more equal, i.e. the point cloud arrangement is more completely uniform, i.e. D_pThe smaller. D_pThe smaller the representative point cloud, the denser the representative point cloud.

By using a similar method, the respective areas S of all the triangles and the respective point cloud density representations D obtained in fig. 3 can be calculated_p. Assuming that N triangles are obtained in total, and the point cloud density representations of the N triangles are respectively marked as D_p1To D_pnThen, the point cloud density of the point cloud data as a whole can be calculated based on the point cloud density representations of all the triangles. For example, the average value of the point cloud density representations of all the triangles may be used as the point cloud density of the point cloud data as a whole, or the maximum value of the point cloud density representations of all the triangles may also be used as the point cloud density of the point cloud data as a whole. In an exemplary manner, the first and second electrodes are,the point cloud density representations of all the triangles can also be combined into a data set, and the data set is directly used as the representation of the point cloud density of the whole point cloud data.

In another example, the point cloud density characterization for each of the graphical units may be equal to the square root of its perimeter multiplied by its area. Still referring to the example of fig. 3, for a triangle R1 composed of point cloud points P1, P2, and P3, the three sides are a, b, and c, respectively, and the area is S, which may be as shown in equation (1). In this example, the point cloud density of triangle R1 characterizes D_pCan be represented by the following formula (3):

similarly, the point cloud density characterization of all the triangles in fig. 3 can be obtained based on equation (3), so as to obtain the point cloud density of the point cloud data as a whole.

In yet another example, the point cloud density representation of each of the graphical units may be equal to its area multiplied by its perimeter. Still referring to the example of fig. 3, for a triangle R1 composed of point cloud points P1, P2, and P3, the three sides are a, b, and c, respectively, and the area is S, which may be as shown in equation (1). In this example, the point cloud density of triangle R1 characterizes D_pCan be represented by the following formula (4):

D _p(a + b + c) formula (4)

Similarly, the point cloud density characterization of all the triangles in fig. 3 can be obtained based on equation (4), so as to obtain the point cloud density of the point cloud data as a whole.

In other examples, the point cloud density characterization of each of the graphic elements may be any other suitable calculation method as long as the obtained point cloud density characterization has a certain monotonic correlation with any one or several geometric parameters of the graphic element. Furthermore, as described above, the description of three point cloud points forming a triangle is merely exemplary, and four or more adjacent points may be selected to form a corresponding graphic unit, as long as the formed graphic units do not overlap with each other. Assuming that a graphic unit is composed of four adjacent points as an example, the area of the graphic unit is S, and the side lengths are a, b, c, and d, respectively, the foregoing formula (2) is modified to the following formula (5):

the aforementioned formula (3) is modified to the following formula (6):

the aforementioned formula (4) is modified to the following formula (7):

D _p(a + b + c + d) formula (7)

The above exemplarily illustrates an exemplary process of mapping point cloud data onto the same surface, then dividing an area of the point cloud data on the surface into a plurality of graphic units that do not overlap with each other, and calculating an overall density of the point cloud data based on a point cloud density characterization of each graphic unit.

In another embodiment of the present invention, the region in which the point cloud data is located may also be directly divided into a plurality of non-overlapping graphic units in a three-dimensional space, and each graphic unit is a stereo graphic unit. In this embodiment, the point cloud density characterization of each graphic unit may be calculated based on the geometric parameters of each solid graphic unit, and then the point cloud density of the whole point cloud data may be obtained based on the point cloud density characterizations of all solid graphic units obtained by segmentation.

Illustratively, the geometric parameters of each stereographic element may include, but are not limited to, a side length, an area, and/or a volume of the stereographic element. For example, the geometric parameters of each solid graphical unit may include, but are not limited to, the distance from the geometric center point of the graphical unit to the respective vertex of the graphical unit and the distance from the geometric center point to the respective edge of the graphical unit. The relational expression of the point cloud density representation and the geometric parameters of each solid graphic unit can be designed in a similar manner as described above, and similarly, it is only required to satisfy that a certain unidirectional correlation exists between the point cloud density representation and any one or more geometric parameters of each solid graphic unit, and for brevity, the description is omitted here.

The above exemplarily describes the method of quantifying the density of the point cloud according to the embodiment of the present invention. Based on the above description, the method for quantizing point cloud density according to the embodiment of the present invention provides a graph calculation scheme that can accurately quantize the density of point cloud data, and can be applied to point cloud data generated by various point cloud detection systems, so that a suitable laser radar product can be selected according to actual needs based on the scheme to generate point cloud data meeting requirements.

Further, the method for quantifying the point cloud density according to the embodiment of the invention may further include the following steps: calculating a point cloud uniformity of the point cloud data based on the point cloud density representations of the plurality of graphical units. The method including this step may be used as a method for quantifying the uniformity of the point cloud according to an embodiment of the present invention. This is described below with reference to fig. 4.

FIG. 4 shows a schematic flow diagram of a method 400 of quantifying the uniformity of a point cloud in accordance with an embodiment of the present invention. As shown in fig. 4, a method 400 for quantifying uniformity of a point cloud according to an embodiment of the present invention may comprise the following steps:

in step S410, for point cloud data of a point cloud density to be calculated, a region in which the point cloud data is located is divided into a plurality of non-overlapping graphic units, each graphic unit being composed of at least three point cloud points adjacent to each other in the point cloud data.

In step S420, a point cloud density representation of each of the graphic units is calculated based on respective geometric parameters of the plurality of graphic units, the point cloud density representation of each of the graphic units being monotonically related to at least one of the geometric parameters of itself.

In step S430, a point cloud uniformity of the point cloud data is calculated based on the point cloud density representations of the plurality of graphical units.

Step S410 and step S420 are the same as step S210 and step S220, respectively, and are not described herein again for brevity. In the embodiment of the present invention, based on the point cloud density representations of the plurality of graphic units obtained in step S420, the point cloud uniformity of the point cloud data as a whole can be calculated. In one example, the calculating a point cloud uniformity of the point cloud data based on the point cloud density characterization of the plurality of graphical units may include: respectively normalizing the point cloud density representations of the plurality of graphic units; and evaluating a standard deviation of the normalized point cloud density representations of the plurality of graphic units to obtain a point cloud uniformity of the point cloud data.

For example, still referring to the example shown in fig. 3, the region where the point cloud is located is divided into N non-overlapping graphic units, and the point cloud density representations thereof are respectively denoted as D_p1To D_pnThen, the data set may be normalized first, and then the normalized point cloud density is characterized by a standard deviation δ, i.e., the point cloud uniformity U of the point cloud data is obtained_p. Illustratively, respectively normalizing the point cloud density representations of the plurality of graphical units may comprise: dividing the point cloud density representations of the plurality of graph elements by an average of the point cloud density representations of the plurality of graph elements, respectively; or dividing the point cloud density representations of the plurality of graphic units by the maximum value in the point cloud density representations of the plurality of graphic units respectively. Following the example above, that is, for D_p1To D_pnAre respectively divided by D_p1To D_pnAverage of all values, or divided by D, respectively_p1To D_pnMaximum of all values. Since some lidar has a high uniformity, although the average density is low, the point cloud density D is high_pThe point cloud density representation of each graphic unit is larger after the point cloud area is divided, and if the standard deviation of the point cloud density representation of each graphic unit is directly used as the point cloud uniformity U_pWill get a larger U_pIn fact, however, the point cloud density is low but high, and in this case, the point cloud density characterization of each graphic unit is normalized and then the standard deviation is calculated to obtain a more accurate value of the point cloud uniformity. Of course, in some scenarios, the standard deviation of the point cloud density characterization of each graphic unit can also be directly used as the point cloud uniformity U_p。

The above exemplarily describes the method for quantifying the uniformity of the point cloud according to the embodiment of the present invention. Based on the above description, the method for quantizing the point cloud uniformity according to the embodiment of the present invention provides a graph calculation scheme that can accurately quantize the uniformity of point cloud data, and is applicable to point cloud data generated by various point cloud detection systems, so that a suitable laser radar product can be selected according to actual needs based on the scheme to generate point cloud data meeting requirements.

The method for quantifying the point cloud density and the point cloud uniformity according to the embodiment of the invention can be applied to any laser radar, because the method can divide the area where the point cloud data generated by any laser radar is located into a plurality of non-overlapping graphic units, and calculate the point cloud density and the point cloud uniformity of the generated point cloud data based on the point cloud density representations of the graphic units. For different laser radars, when the point cloud density and the point cloud uniformity are obtained according to the method for quantizing the point cloud density and the point cloud uniformity, appropriate laser radar products can be selected as required to generate corresponding point cloud data by comparing the respective point cloud density and the point cloud uniformity.

The following calculations and comparisons are made using four lidar arrays as examples. To simplify the calculation, illustratively, the point cloud is projected onto a plane and the segmented graphical elements are triangles. As shown in fig. 5A to 5D. Fig. 5A to 5D are schematic diagrams illustrating a point cloud density quantization method according to an embodiment of the present invention, which constructs a graph unit for calculating a point cloud density representation for different lidar point cloud arrays. For comparison, the lidar point cloud arrays shown in fig. 5A to 5D are all 3 lines, and the number of points is 9. The point cloud density and uniformity of the lidar shown in fig. 5A to 5D are indistinguishable in the original way of characterizing the point cloud density in terms of line number and point number. It is clear, however, that the point cloud density and uniformity of the point cloud arrays of the lidar shown in fig. 5A through 5D are different. The point cloud density and uniformity of the point cloud arrays of the lidar shown in fig. 5A-5D are calculated as follows according to the above method according to an embodiment of the present invention:

the lidar 1 shown in FIG. 5A has a triangle consisting of points P1, P2 and P3 with side lengths of 1, 1 and

then, the area of the triangle is 0.5, and the point cloud density characterization D of the triangle is calculated according to the formula (2) in the foregoing_pThen:

due to the characteristics of the scanning mode of the laser radar 1, the D of all the triangles obtained by the division_pBoth 2.4142. Based on this, the point cloud density of the laser radar 1 is 2.4142, and the point cloud uniformity is 0.

The lidar 2 shown in FIG. 5B has a triangle consisting of points P1, P2 and P3 with side lengths of 1, 0.5 and

then, the area of the triangle is 0.25, and the point cloud density characterization D of the triangle is calculated according to the formula (2)_pThen:

due to the characteristics of the scanning mode of the laser radar 2, the D of all the triangles obtained by the division_pBoth 1.309. Based on this, the point cloud density of the laser radar 2 is 1.309, pointsThe cloud uniformity was 0.

In the lidar 3 shown in fig. 5C, the side lengths of the triangle formed by the points P1, P2 and P3 are 1, 1 and 1, respectively, then the area of the triangle is 0.433, and the point cloud density characterization D of the triangle is calculated according to the formula (2)_pThen:

due to the characteristics of the scanning mode of the laser radar 3, the D of all the triangles obtained by the division_pBoth 1.9741. Based on this, the point cloud density of the laser radar 3 is 1.9741, and the point cloud uniformity is 0.

In the laser radar 4 shown in fig. 5D, the side lengths of the triangle formed by the points P1, P2 and P3 are 1, 1.34536 and 1.005, respectively, and then the area of the triangle is 0.5, and the point cloud density characterization D is calculated according to the formula (2)_pThen:

due to the scanning pattern of the lidar 4, the divided remaining triangles have a size D_pCan be respectively as follows: 2.1771, 2.1771, 2.3691, 2.6099, 2.4142, 2.4142, and 2.6099. Based on this, the point cloud density of the laser radar 4 is 2.3926, and the point cloud uniformity is 0.0687.

By adopting the point cloud density and point cloud uniformity quantification method provided by the embodiment of the invention, accurate values of the point cloud density and the point cloud uniformity of each of the laser radar 1 to the laser radar 4 can be obtained, so that comparison can be performed. In a further embodiment of the present invention, a density distribution map of the point cloud data may also be drawn and presented based on the point cloud density characterization of the plurality of graphic units obtained by the segmentation. For example, taking the point cloud arrays of four types of lidar shown in fig. 5A to 5D as an example, the point cloud density representations of the respective graphic units are plotted into a point cloud density distribution diagram as shown in fig. 6.

As shown in fig. 6, the point cloud densities of each of lidar 1, lidar 2, and lidar 3 are shown by lines 610, 620, and 630 in fig. 6, respectively, and the values of the point cloud densities of the graphic elements of lidar 4 are not completely the same, and thus a plurality of lines 640 occur. In general, the smaller the point cloud density characterization value is, the higher the representative point cloud density is, so as can be clearly seen from fig. 6, the point cloud density of the laser radar 2 is the highest; lidar 1 and lidar 4 are very close, but the former is slightly lower. Furthermore, as is apparent from fig. 6, as the data volume gradually increases, the density distribution map based on the point cloud density characterization values of the plurality of graph elements gradually exhibits a characteristic similar to a "frequency spectrum". In general, the wider the "band", the worse the point cloud uniformity; the narrower the band, the better the point cloud uniformity. For a symmetrically distributed "spectrum", the center value can be directly calculated, and the smaller the center value, the higher the point cloud density. For asymmetric "spectra," the weighting can be performed according to the requirements of a particular scene, and a weighted center value can be calculated. Likewise, a smaller weighted center value represents a higher point cloud density. In actual use, certain frequency bands can be removed, and calculation is carried out, so that the point cloud density and point cloud uniformity indexes which are more in line with actual requirements are obtained.

Thus, in a further embodiment of the present invention, another method of calculating point cloud density and point cloud uniformity may be provided, namely by calculating a density profile of a point cloud density representation of a graph element obtained by segmentation. Specifically, a density center value of a density distribution map of the point cloud data may be calculated and taken as a point cloud density of the point cloud data. For the calculation of the point cloud uniformity, the density distribution width of the density distribution map of the point cloud data can be calculated and taken as the point cloud uniformity of the point cloud data. Further, point cloud density representations of the plurality of graphical units may be weighted based on user settings, and a point cloud density and/or a point cloud uniformity of the point cloud data may be calculated based on the weighted point cloud density representations of the plurality of graphical units.

In a further embodiment of the invention, it may further comprise presenting to the user at least one of: the point cloud density of the point cloud data, the point cloud uniformity of the point cloud data, and the density distribution map of the point cloud data. In this embodiment, the calculation result and/or the density distribution map drawing result may be presented to the user, so that the user can intuitively know the index conditions of the point cloud density and the point cloud uniformity of the lidar or the index comparison conditions of the point cloud density and the point cloud uniformity of different lidar.

The point cloud density and point cloud uniformity quantification methods according to the embodiments of the present invention are exemplarily described above, and since the point cloud density and point cloud uniformity calculations need to be characterized based on the point cloud densities of a plurality of graphic units obtained by segmentation, the point cloud density and point cloud uniformity quantification methods may also be collectively referred to as point cloud density quantification methods. Based on the above description, the method for quantizing point cloud density according to the embodiment of the present invention provides a graph calculation scheme capable of accurately quantizing density and uniformity of point cloud data, and is applicable to point cloud data generated by various point cloud detection systems, so that a suitable lidar product can be selected according to actual needs based on the scheme to generate point cloud data meeting requirements, and the point cloud density and the point cloud uniformity obtained by quantization also have important reference values for testing and iteration of the lidar product.

The point cloud density quantization apparatus provided by another aspect of the present invention is described below with reference to fig. 7. Fig. 7 shows a schematic block diagram of an apparatus 700 for quantizing point cloud density of point cloud data according to an embodiment of the present invention. The apparatus 700 for quantifying the point cloud density comprises a memory 710 and a processor 720.

The memory 710 stores a program for implementing corresponding steps in the method for quantifying point cloud density according to the embodiment of the present invention. The processor 720 is configured to execute the program stored in the memory 710 to perform the corresponding steps of the method for quantifying the point cloud density according to the embodiment of the present invention.

In one embodiment, the program when executed by the processor 720 causes the apparatus 700 for quantifying the density of a point cloud to perform the steps of: aiming at point cloud data of point cloud density to be calculated, dividing an area where the point cloud data is located into a plurality of non-overlapping graphic units, wherein each graphic unit is composed of at least three adjacent point cloud points in the point cloud data; calculating a point cloud density characterization for each of the graphical units based on respective geometric parameters of the plurality of graphical units, the point cloud density characterization for each of the graphical units being monotonically related to at least one of the geometric parameters of itself; and calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density representation of each graphic unit.

In one embodiment, the segmenting of the region in which the point cloud data is located into a plurality of graphics units that do not overlap with each other, which is performed by the device 700 for quantifying point cloud density when the program is executed by the processor 720, further comprises: mapping the point cloud data to the same surface; and dividing an area of the point cloud data on the surface into a plurality of graphic units which do not overlap with each other.

In one embodiment, the point cloud density characterization for each of the graphical units is equal to the square root of its area multiplied by its perimeter.

In one embodiment, the point cloud density representation of each of the graphical units is equal to the square root of its perimeter multiplied by its area.

In one embodiment, the point cloud density characterization of each of the graphical units is equal to its area multiplied by its perimeter.

In one embodiment, the faces are planar and the graphical elements are triangles.

In one embodiment, the graphic element is a solid graphic element, and the geometric parameter includes a side length, an area and/or a volume of the solid graphic element.

In one embodiment, the geometric parameters include distances from a geometric center point of the graphic element to respective vertices of the graphic element and distances from the geometric center point to respective edges of the graphic element.

In one embodiment, the geometric parameters further include distances from geometric center points of the graphic elements to respective faces of the graphic elements.

In one embodiment, the point cloud density of the point cloud data for which the point cloud density is to be calculated based on the point cloud density characterization for each of the graphic units, which is performed by the point cloud density quantifying device 700 when the program is executed by the processor 720, includes: taking the average value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated; or taking the maximum value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: calculating a point cloud uniformity of the point cloud data based on the point cloud density representations of the plurality of graphical units.

In one embodiment, the point cloud uniformity calculation of the point cloud data based on the point cloud density characterization of the plurality of graphical units performed by the point cloud density quantification apparatus 700 when the program is executed by the processor 720 includes: respectively normalizing the point cloud density representations of the plurality of graphic units; and evaluating a standard deviation of the normalized point cloud density representations of the plurality of graphic units to obtain a point cloud uniformity of the point cloud data.

In one embodiment, the normalization that the means for quantifying the density of the point cloud 700 performs when the program is run by the processor 720 comprises: dividing the point cloud density representations of the plurality of graph elements by an average of the point cloud density representations of the plurality of graph elements, respectively; or dividing the point cloud density representations of the plurality of graphic units by the maximum value in the point cloud density representations of the plurality of graphic units respectively.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: and drawing and presenting a density distribution map of the point cloud data based on the point cloud density representations of the plurality of graphic units.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: and calculating a density center value of the density distribution map of the point cloud data, and taking the density center value as the point cloud density of the point cloud data.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: and calculating the density distribution width of the density distribution map of the point cloud data, and taking the density distribution width as the point cloud uniformity of the point cloud data.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: weighting the point cloud density representations of the plurality of graphic units based on user settings, and calculating the point cloud density and/or point cloud uniformity of the point cloud data based on the weighted point cloud density representations of the plurality of graphic units.

In one embodiment, the program when executed by the processor 720 further causes the apparatus for quantifying a point cloud density 700 to perform the steps of: presenting to the user at least one of: the point cloud density of the point cloud data, the point cloud uniformity of the point cloud data, and the density distribution map of the point cloud data.

Furthermore, according to an embodiment of the present invention, there is also provided a storage medium on which program instructions are stored, which when executed by a computer or a processor are used for executing the respective steps of the method for quantifying point cloud density of an embodiment of the present invention. The storage medium may include, for example, a memory card of a smart phone, a storage component of a tablet computer, a hard disk of a personal computer, a Read Only Memory (ROM), an Erasable Programmable Read Only Memory (EPROM), a portable compact disc read only memory (CD-ROM), a USB memory, or any combination of the above storage media. The computer-readable storage medium may be any combination of one or more computer-readable storage media.

In one embodiment, the computer program instructions, when executed by a computer, may perform a method of quantifying a point cloud density according to an embodiment of the invention.

In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the steps of: aiming at point cloud data of point cloud density to be calculated, dividing an area where the point cloud data is located into a plurality of non-overlapping graphic units, wherein each graphic unit is composed of at least three adjacent point cloud points in the point cloud data; calculating a point cloud density characterization for each of the graphical units based on respective geometric parameters of the plurality of graphical units, the point cloud density characterization for each of the graphical units being monotonically related to at least one of the geometric parameters of itself; and calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density representation of each graphic unit.

In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the segmenting the area in which the point cloud data is located into a plurality of graphical units that do not overlap with each other, further comprising: mapping the point cloud data to the same surface; and dividing an area of the point cloud data on the surface into a plurality of graphic units which do not overlap with each other.

In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the calculating of the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density characterization of each of the graphical units, comprising: taking the average value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated; or taking the maximum value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: calculating a point cloud uniformity of the point cloud data based on the point cloud density representations of the plurality of graphical units.

In one embodiment, the computer program instructions, when executed by a computer or processor, cause the computer or processor to perform the computing of point cloud uniformity of the point cloud data based on point cloud density characterizations of the plurality of graphical elements, comprising: respectively normalizing the point cloud density representations of the plurality of graphic units; and evaluating a standard deviation of the normalized point cloud density representations of the plurality of graphic units to obtain a point cloud uniformity of the point cloud data.

In one embodiment, the normalization, which when executed by a computer or processor, causes the computer or processor to perform, comprises: dividing the point cloud density representations of the plurality of graph elements by an average of the point cloud density representations of the plurality of graph elements, respectively; or dividing the point cloud density representations of the plurality of graphic units by the maximum value in the point cloud density representations of the plurality of graphic units respectively.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: and drawing and presenting a density distribution map of the point cloud data based on the point cloud density representations of the plurality of graphic units.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: and calculating a density center value of the density distribution map of the point cloud data, and taking the density center value as the point cloud density of the point cloud data.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: and calculating the density distribution width of the density distribution map of the point cloud data, and taking the density distribution width as the point cloud uniformity of the point cloud data.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: weighting the point cloud density representations of the plurality of graphic units based on user settings, and calculating the point cloud density and/or point cloud uniformity of the point cloud data based on the weighted point cloud density representations of the plurality of graphic units.

In one embodiment, the computer program instructions, when executed by a computer or processor, further cause the computer or processor to perform the steps of: presenting to the user at least one of: the point cloud density of the point cloud data, the point cloud uniformity of the point cloud data, and the density distribution map of the point cloud data.

Based on the above description, the method, the device and the storage medium for quantizing the point cloud density according to the embodiments of the present invention provide a graph calculation scheme that can accurately quantize the density and the uniformity of the point cloud data, and can be applied to the point cloud data generated by various point cloud detection systems, so that a suitable lidar product can be selected according to actual needs based on the scheme to generate point cloud data meeting requirements, and the quantized point cloud density and point cloud uniformity also have important reference values for the test and iteration of the lidar product.

Although the illustrative embodiments have been described herein with reference to the accompanying drawings, it is to be understood that the foregoing illustrative embodiments are merely exemplary and are not intended to limit the scope of the invention thereto. Various changes and modifications may be effected therein by one of ordinary skill in the pertinent art without departing from the scope or spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as set forth in the appended claims.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described device embodiments are merely illustrative, and for example, the division of the units is only one logical functional division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another device, or some features may be omitted, or not executed.

In the description provided herein, numerous specific details are set forth. It is understood, however, that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description.

Similarly, it should be appreciated that in the description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the invention and aiding in the understanding of one or more of the various inventive aspects. However, the method of the present invention should not be construed to reflect the intent: that is, the claimed invention requires more features than are expressly recited in a claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with the claims themselves being directed to separate embodiments of the present invention.

It will be understood by those skilled in the art that all of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and all of the processes or elements of any method or apparatus so disclosed, may be combined in any combination, except combinations where such features are mutually exclusive. The features disclosed in this specification (including any accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will appreciate that while some embodiments described herein include some features included in other embodiments, rather than other features, combinations of features of different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art will appreciate that a microprocessor or Digital Signal Processor (DSP) may be used in practice to implement some or all of the functionality of some of the modules according to embodiments of the present invention. The present invention may also be embodied as apparatus programs (e.g., computer programs and computer program products) for performing a portion or all of the methods described herein. Such programs implementing the present invention may be stored on a computer-readable storage medium or may be in the form of one or more signals. Such a signal may be downloaded from an internet website or provided on a carrier signal or in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by one and the same item of hardware. The usage of the words first, second and third, etcetera do not indicate any ordering. These words may be interpreted as names.

The above description is only for the specific embodiment of the present invention or the description thereof, and the protection scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and the changes or substitutions should be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

A method of quantifying point cloud density, the method comprising:

aiming at point cloud data of point cloud density to be calculated, dividing an area where the point cloud data is located into a plurality of non-overlapping graphic units, wherein each graphic unit is composed of at least three adjacent point cloud points in the point cloud data;

calculating a point cloud density characterization for each of the graphical units based on respective geometric parameters of the plurality of graphical units, the point cloud density characterization for each of the graphical units being monotonically related to at least one of the geometric parameters of itself; and

and calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density representation of each graphic unit.
The method of claim 1, wherein the segmenting the region in which the point cloud data is located into a plurality of graphical units that do not overlap with each other, further comprises:

mapping the point cloud data to the same surface; and

and dividing the area of the point cloud data on the surface into a plurality of graphic units which do not overlap with each other.
The method of claim 2, wherein the point cloud density representation of each of the graphical units is equal to the square root of its area multiplied by its perimeter.
The method of claim 2, wherein the point cloud density representation of each of the graphical units is equal to the square root of its perimeter multiplied by its area.
The method of claim 2, wherein the point cloud density representation of each of the graphical units is equal to its area multiplied by its perimeter.
The method of any of claims 2-5, wherein the face is a plane and the graphical elements are triangles.
The method of claim 1, wherein the graphic element is a solid graphic element, and the geometric parameter comprises a side length, an area and/or a volume of the solid graphic element.
The method of claim 1, wherein the geometric parameters comprise distances from a geometric center point of the graphical element to respective vertices of the graphical element and from the geometric center point to respective edges of the graphical element.
The method of claim 8, wherein the geometric parameters further comprise distances from geometric center points of the graphical elements to respective faces of the graphical elements.
The method of any of claims 1-9, wherein the calculating the point cloud density of the point cloud data of the point cloud density to be calculated based on the point cloud density characterization of each of the graphical units comprises:

taking the average value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated; or

And taking the maximum value of the point cloud density representations of the plurality of graphic units as the point cloud density of the point cloud data of the point cloud density to be calculated.
The method according to any one of claims 1-10, further comprising:

calculating a point cloud uniformity of the point cloud data based on the point cloud density representations of the plurality of graphical units.
The method of claim 11, wherein calculating the point cloud uniformity of the point cloud data based on the point cloud density characterization of the plurality of graphical units comprises:

respectively normalizing the point cloud density representations of the plurality of graphic units; and

and evaluating a standard deviation of the normalized point cloud density representations of the plurality of graphic units to obtain a point cloud uniformity of the point cloud data.
The method of claim 12, wherein the normalizing comprises:

dividing the point cloud density representations of the plurality of graph elements by an average of the point cloud density representations of the plurality of graph elements, respectively; or

And dividing the point cloud density representations of the plurality of graphic units by the maximum value in the point cloud density representations of the plurality of graphic units respectively.
The method according to any one of claims 1-10, further comprising:

and drawing and presenting a density distribution map of the point cloud data based on the point cloud density representations of the plurality of graphic units.
The method of claim 14, further comprising:

and calculating a density center value of the density distribution map of the point cloud data, and taking the density center value as the point cloud density of the point cloud data.
The method of claim 14, further comprising:

and calculating the density distribution width of the density distribution map of the point cloud data, and taking the density distribution width as the point cloud uniformity of the point cloud data.
The method of any one of claims 1-9 or 11-14, further comprising:

weighting the point cloud density representations of the plurality of graphic units based on user settings, and calculating the point cloud density and/or point cloud uniformity of the point cloud data based on the weighted point cloud density representations of the plurality of graphic units.
The method according to any one of claims 1-17, further comprising presenting to a user at least one of: the point cloud density of the point cloud data, the point cloud uniformity of the point cloud data, and the density distribution map of the point cloud data.
An apparatus for quantification of point cloud density, characterized in that the apparatus comprises a memory and a processor, the memory having stored thereon a computer program for execution by the processor, the computer program, when executed by the processor, performing the method of quantification of point cloud density according to any one of claims 1-18.
A storage medium, characterized in that the storage medium has stored thereon a computer program which, when running, executes a method of quantification of a point cloud density according to any one of claims 1-18.