CN115035206A

CN115035206A - Compression method and decompression method of laser point cloud and related devices

Info

Publication number: CN115035206A
Application number: CN202210497782.9A
Authority: CN
Inventors: 林辉; 卢维; 王政; 李铭
Original assignee: Zhejiang Huaray Technology Co Ltd
Current assignee: Zhejiang Huaray Technology Co Ltd
Priority date: 2022-05-09
Filing date: 2022-05-09
Publication date: 2022-09-09
Anticipated expiration: 2042-05-09
Also published as: CN115035206B

Abstract

The application provides a compression method, a decompression method and a related device of laser point cloud, and relates to the technical field of automatic navigation. The compression method comprises the steps of determining a target straight line in a laser point cloud frame based on the position of a laser scanning point in the laser point cloud frame, and dividing the laser point cloud frame based on the target straight line to obtain each area, so that the compression information of the laser point cloud frame is determined to be first compression information corresponding to each target straight line and second compression information corresponding to each area, wherein the first compression information corresponding to the target straight line comprises straight line information of the target straight line and starting and stopping point cloud serial numbers of the laser scanning point on the target straight line; and the second compressed information corresponding to the area comprises the scanning distance and the starting-stopping point cloud serial number of each laser scanning point in the area. The calculation amount of the straight line extraction is small, the realization is simple, and the point cloud information on the straight line can be greatly reduced, so that the compression cost and the compressed data amount can be greatly reduced, the compression efficiency is improved, and the high-efficiency compression of the laser point cloud is realized.

Description

Compression method and decompression method of laser point cloud and related devices

Technical Field

The embodiment of the application relates to the technical field of automatic navigation, in particular to a compression method and a decompression method of laser point cloud and a related device.

Background

2D laser navigation is a key technology in an Automatic Guided Vehicle (AGV), along with the continuous increase of application requirements of mobile robots, the number of robot devices adopting 2D laser navigation is more and more, the running time of executing tasks is longer and longer, the data volume of 2D laser point clouds is also increased, and the storage and transmission requirements of more and more mass point cloud data put forward higher and higher requirements on the compression and the decompression of the 2D laser point clouds.

At present, the commonly adopted laser point cloud compression method is to convert the laser point cloud into an image or a video, and then to compress the image or the video by adopting a traditional image/video compression algorithm. However, the data amount of the laser point cloud is generally less than that of the image or the video, and the compression cost of the image/video compression algorithm is high, so that the compression cost of compressing the laser point cloud by using the method is high, and the compression efficiency is low.

Disclosure of Invention

In order to solve the existing technical problems, embodiments of the present application provide a compression method, a decompression method, and a related apparatus for laser point cloud, which can improve the compression efficiency of compressing the laser point cloud and reduce the compression cost.

In order to achieve the above purpose, the technical solution of the embodiment of the present application is implemented as follows:

in a first aspect, an embodiment of the present application provides a method for compressing a laser point cloud, where the method includes:

determining a target straight line in a laser point cloud frame based on the positions of laser scanning points in the laser point cloud frame, wherein the number of the laser scanning points on the target straight line is greater than a first set number threshold, the point cloud serial numbers of the laser scanning points on the target straight line are continuous, the point cloud serial numbers of the laser scanning points are set based on a scanning sequence, and the laser scanning points on different target straight lines are different;

taking first compressed information corresponding to each target straight line and second compressed information corresponding to each area as compressed information of the laser point cloud frame, wherein the areas are obtained by dividing based on the target straight lines, and the first compressed information corresponding to the target straight lines comprises straight line information of the target straight lines, the maximum point cloud serial number value and the minimum point cloud serial number value of laser scanning points positioned on the target straight lines; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number; and the scanning distance of the laser scanning point is the distance between the laser scanning point and a laser emitting device of the laser radar.

According to the method for compressing the laser point cloud provided by the embodiment of the application, the target straight line can be extracted from the laser point cloud frame based on the position of the laser scanning point in the laser point cloud frame, the laser point cloud frame is compressed according to the extracted target straight line, and the compression information of the laser point cloud frame is obtained, wherein the compression information comprises first compression information corresponding to each target straight line and second compression information corresponding to each area, the areas are obtained based on target straight line division, and the first compression information corresponding to the target straight line comprises straight line information of the target straight line, the maximum point cloud serial number value and the minimum point cloud serial number value of the laser scanning point on the target straight line; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number. Because the calculation amount of the straight line extraction is small, the realization is simple, so that the laser point cloud is compressed by adopting a straight line extraction mode, the compression cost can be greatly reduced, and the compression efficiency is improved; and when compressing the laser scanning point on the straight line, the obtained compressed information only retains the point cloud number maximum value and the point cloud number minimum value of the laser scanning point on the straight line and the straight line information of the straight line, thereby greatly reducing the data volume and realizing high-efficiency compression.

In an alternative embodiment, the determining a target straight line in the laser point cloud frame based on the position of the laser scanning point in the laser point cloud frame includes:

dividing laser scanning points included in a laser point cloud frame into a plurality of point cloud sets, wherein each laser scanning point is located in one point cloud set, and the point cloud sets comprise the laser scanning points with continuous sequence numbers;

aiming at any one target point cloud set with the number of laser scanning points larger than a second set number threshold, determining a target straight line corresponding to the target point cloud set, wherein the target straight line is the straight line with the minimum sum of distances between the laser scanning points included in the target point cloud set; or the target straight line is the straight line which comprises the most laser scanning points in the target point cloud set.

In this embodiment, the laser scanning points included in the laser point cloud frame may be first divided into a plurality of point cloud sets, and then a target straight line corresponding to a target point cloud set is determined from any one of the target point cloud sets whose number of the laser scanning points is greater than a second set number threshold, where the target straight line may be a straight line having a minimum sum of distances between the target point cloud set and the laser scanning points included in the target point cloud set, or may be a straight line including the laser scanning points in the target point cloud set at the maximum. Therefore, a larger number of target straight lines which better meet the set requirements can be extracted from the laser point cloud frame.

In an alternative embodiment, the dividing the laser scanning points included in the laser point cloud frame into a plurality of point cloud sets includes:

taking laser scanning points included in the laser point cloud frame as a point cloud set, and circularly executing the following operations until the point cloud set with the number of the laser scanning points smaller than a second set number threshold exists:

determining a fitting straight line in the point cloud set by a least square method according to the positions of the laser scanning points in the point cloud set, and determining target scanning points meeting set distance conditions according to the distance between each laser scanning point in the point cloud set and the fitting straight line;

dividing the laser scanning points positioned in front of the point cloud serial number of the target scanning point and the target scanning point into one point cloud set, and dividing the laser scanning points positioned behind the point cloud serial number of the target scanning point into another point cloud set.

In this embodiment, the laser scanning points included in the laser point cloud frame may be regarded as a point cloud set, and then the following operations are executed in a loop until there is a point cloud set whose number of laser scanning points is less than a second set number threshold: according to the positions of laser scanning points in the point cloud sets, a fitting straight line is determined in the point cloud sets through a least square method, target scanning points meeting set distance conditions are determined according to the distance between each laser scanning point and the fitting straight line in the point cloud sets, the laser scanning points located in front of the point cloud serial numbers of the target scanning points and the target scanning points are divided into one point cloud set, and the laser scanning points located behind the point cloud serial numbers of the target scanning points are divided into another point cloud set. Therefore, the laser scanning points included in the laser point cloud frame can be reasonably divided into the corresponding point cloud sets, and when the target straight line is extracted from the divided point cloud sets, the more appropriate target straight line can be extracted.

In an optional embodiment, the line information of the target line includes a line parameter and a line number, and the line information of the target line includes the line number set based on the sequence of the acquired laser point cloud frames; or

The target straight line comprises a collinear straight line and a non-collinear straight line, the straight line information of the non-collinear straight line comprises straight line parameters and straight line serial numbers, and the straight line information of the collinear straight line comprises the straight line serial numbers of the corresponding non-collinear straight lines;

wherein a collinear line of the target lines is determined by:

randomly selecting a laser point cloud frame from a plurality of continuous laser point cloud frames as a key frame, and using other laser point cloud frames in the plurality of continuous laser point cloud frames as common frames, wherein the linear sequence number included in the linear information of the non-collinear line is set based on the sequence of the acquired laser point cloud frames in the plurality of laser point cloud frames;

and taking a target straight line which meets a set collinear condition with at least one target straight line included in the key frame as a collinear straight line, wherein the target straight line which meets the set collinear condition with the collinear straight line is a non-collinear straight line corresponding to the collinear straight line.

In this embodiment, when a single laser point cloud frame is compressed, the line information of the target line included in the first compressed information corresponding to the target line in the obtained compressed information may include line parameters and line serial numbers, and the line serial numbers included in the line information of the target line are set based on the sequence of the acquired laser point cloud frames. When a plurality of continuous laser point cloud frames are compressed, the extracted target straight line may include a collinear straight line and a non-collinear straight line, then the straight line information of the non-collinear straight line included in the first compressed information corresponding to the non-collinear straight line in the obtained compressed information may include a straight line parameter and a straight line sequence number, and the straight line information of the collinear straight line included in the first compressed information corresponding to the collinear straight line may include a straight line sequence number of the corresponding non-collinear straight line. The collinear straight line is a target straight line which meets the set collinear condition with at least one target straight line included by a key frame in the continuous multiple laser point cloud frames in each target straight line included by a common frame in the continuous multiple laser point cloud frames. When a plurality of continuous laser point cloud frames are compressed, straight lines included by common frames in the plurality of continuous laser point cloud frames and straight lines included by key frames in the plurality of continuous laser point cloud frames are subjected to straight line fusion, and the straight lines in the common frames meeting the collinear condition are removed from compressed information, so that the number of the extracted straight lines can be reduced, the data volume is further reduced, and the compression efficiency is improved.

In an optional embodiment, the regarding, as collinear straight lines, among the target straight lines included in the normal frame, target straight lines that satisfy a set collinear condition with at least one target straight line included in the key frame includes:

for each target straight line included in the common frame, respectively performing the following operations:

projecting the target straight lines included in the common frame to the reference frame of the key frame, and respectively determining the distance between the target straight lines included in the common frame and at least one target straight line included in the key frame;

and if the distance between the target straight line included in the common frame and the key target straight line in at least one target straight line included in the key frame meets a set threshold condition, taking the target straight line included in the common frame as a collinear straight line.

In this embodiment, by projecting each target straight line included in the normal frame to a reference frame of the key frame, and determining a distance between the target straight line included in the normal frame and at least one target straight line included in the key frame, respectively, when the distance between one target straight line included in the normal frame and the key target straight line included in the key frame satisfies a set threshold condition, the target straight line included in the normal frame may be regarded as a collinear straight line. By calculating the distance between the straight line in the common frame and the straight line in the key frame, the target straight line which is collinear with the straight line in the key frame can be determined in the common frame, so that the straight line fusion can be completed, the straight line in the common frame which meets the collinear condition is removed from the compressed information, and the number of the extracted straight lines is reduced.

In an alternative embodiment, the set threshold condition is that the first dot line distance is less than a second set threshold distance, and the second dot line distance is less than a second set threshold distance;

the first point line distance is the distance between a laser scanning point corresponding to the maximum value of the point cloud serial number on a target straight line included in the common frame and the key target straight line; the second wire distance is the distance between the laser scanning point corresponding to the minimum value of the point cloud serial number on the target straight line included in the common frame and the key target straight line.

In this embodiment, the threshold condition is set such that the first dot-line distance is less than the second set threshold distance, and the second dot-line distance is less than the second set threshold distance. The first point line distance is the distance between a laser scanning point corresponding to the maximum value of the point cloud serial number on a target straight line included in the common frame and a key target straight line; the second wire distance is the distance between the laser scanning point corresponding to the minimum value of the point cloud serial numbers on the target straight lines included in the common frame and the key target straight line. Because whether the two straight lines are collinear or not can be determined according to the distance from the point to the straight line, the complex calculation amount for determining the collinearity of the two straight lines is reduced, and the calculation is simple and quick to realize.

In a second aspect, an embodiment of the present application provides a method for decompressing a laser point cloud, where the method includes:

acquiring compressed information of a laser point cloud frame; the compressed information comprises first compressed information corresponding to each target straight line and second compressed information corresponding to each area; the target straight line is determined in the laser point cloud frame by a compression end based on the position of a laser scanning point in the laser point cloud frame; the region is obtained by dividing based on the target straight line;

determining the point cloud serial numbers of the laser scanning points on the target straight line according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points on the target straight line, which are included in the first compressed information, determining the scanning angle corresponding to the point cloud serial numbers of the laser scanning points on the target straight line according to the incidence relation between the point cloud serial numbers of the laser scanning points and the scanning angle, and determining the position of each laser scanning point on the target straight line in the laser point cloud frame based on the straight line information of the target straight line and the scanning angle of each laser scanning point; and

determining the point cloud serial numbers of the laser scanning points in the area according to the maximum point cloud serial number and the minimum point cloud serial number of the laser scanning points in the area, which are included by the second compressed information, determining the scanning angle corresponding to the point cloud serial number of each laser scanning point in the area according to the association relation, and determining the position of each laser scanning point in the area in the laser point cloud frame based on the determined scanning angle and the scanning distance of each laser scanning point included by the second compressed information.

According to the decompression method of the laser point cloud provided by the embodiment of the application, after compressed information of a laser point cloud frame is obtained, the point cloud serial number of each laser scanning point on the target straight line can be determined according to the maximum point cloud serial number and the minimum point cloud serial number of the laser scanning point on the target straight line, which are included in first compressed information corresponding to each target straight line in the compressed information, the scanning angle corresponding to the point cloud serial number of each laser scanning point on the target straight line is determined according to the incidence relation between the point cloud serial number of each laser scanning point and the scanning angle, and the position of each laser scanning point on the target straight line in the laser point cloud frame is determined based on the straight line information of the target straight line and the scanning angle of each laser scanning point; and determining the point cloud serial numbers of the laser scanning points in the areas according to the maximum point cloud serial number and the minimum point cloud serial number of the laser scanning points in the areas, which are included in the second compressed information corresponding to the areas in the compressed information, determining the scanning angles corresponding to the point cloud serial numbers of the laser scanning points in the areas according to the association relation, and determining the positions of the laser scanning points in the areas in the laser point cloud frames based on the determined scanning angles and the scanning distances of the laser scanning points included in the second compressed information. Because can be according to the some cloud segmentation classification of laser point cloud, decompress from the compressed information that the straight line corresponds or the compressed information that the region corresponds and obtain each laser scanning point in the laser point cloud frame to can realize accomplishing decompressing to the laser point cloud high-efficiently fast.

In an optional embodiment, the determining, based on the line information of the target straight line and the scanning angle of each laser scanning point, the position of each laser scanning point on the target straight line in the laser point cloud frame includes:

if the line information of the target line comprises line parameters and line serial numbers, determining the positions of the laser scanning points on the target line in the laser point cloud frame based on the line parameters of the target line and the scanning angles of the laser scanning points;

if the line information of the target line comprises a line serial number, determining that the target line is a collinear line, determining line parameters of the non-collinear line according to the non-collinear line corresponding to the line serial number, and determining the positions of the laser scanning points on the target line in the laser point cloud frame based on the line parameters of the non-collinear line and the scanning angles of the laser scanning points, wherein a set collinear condition is met between the collinear line and the corresponding non-collinear line.

In this embodiment, if the line information of the target line includes the line parameter and the line serial number, the position of each laser scanning point on the target line in the laser point cloud frame may be determined based on the line parameter of the target line and the scanning angle of each laser scanning point; and if the line information of the target line comprises a line serial number, determining that the target line is a collinear line, determining a line parameter of the non-collinear line according to the non-collinear line corresponding to the line serial number, and determining the position of each laser scanning point on the target line in the laser point cloud frame based on the line parameter of the non-collinear line and the scanning angle of each laser scanning point. Therefore, when the straight line information comprises the straight line parameters and the straight line serial numbers, the first compressed information of the straight line corresponding to the straight line information can be directly decompressed; when the straight line information only comprises the straight line serial number, the straight line corresponding to the straight line information can be determined to be a collinear straight line, the straight line parameter of the non-collinear straight line is obtained according to the non-collinear straight line corresponding to the straight line serial number, decompression of the first compressed information corresponding to the collinear straight line is completed based on the straight line parameter, and the efficiency of decompressing the laser point cloud is further improved.

In a third aspect, an embodiment of the present application further provides a compression apparatus for laser point cloud, including:

the system comprises a straight line extraction unit, a target straight line extraction unit and a target image acquisition unit, wherein the straight line extraction unit is used for determining a target straight line in a laser point cloud frame based on the position of laser scanning points in the laser point cloud frame, the number of the laser scanning points on the target straight line is greater than a first set number threshold, the point cloud serial numbers of the laser scanning points on the target straight line are continuous, the point cloud serial numbers of the laser scanning points are set based on a scanning sequence, and the laser scanning points on different target straight lines are different;

the information compression unit is used for taking first compression information corresponding to each target straight line and second compression information corresponding to each area as compression information of the laser point cloud frame, wherein the areas are obtained by dividing based on the target straight lines, and the first compression information corresponding to the target straight lines comprises straight line information of the target straight lines, the maximum point cloud serial number value and the minimum point cloud serial number value of laser scanning points positioned on the target straight lines; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number; and the scanning distance of the laser scanning point is the distance between the laser scanning point and a laser emitting device of the laser radar.

In an optional embodiment, the straight line extracting unit is further configured to:

aiming at any one target point cloud set with the number of laser scanning points larger than a second set number threshold value, determining a target straight line corresponding to the target point cloud set, wherein the target straight line is the straight line with the minimum sum of distances between the laser scanning points included in the target point cloud set; or the target straight line is the straight line which comprises the most laser scanning points in the target point cloud set.

determining a fitting straight line in the point cloud set through a least square method according to the positions of the laser scanning points in the point cloud set, and determining target scanning points meeting set distance conditions according to the distance between each laser scanning point in the point cloud set and the fitting straight line;

wherein a collinear line of the target lines is determined by:

In an optional embodiment, the information compressing unit is further configured to:

and if the distance between the target straight line included in the common frame and the key target straight line in the at least one target straight line included in the key frame meets a set threshold condition, taking the target straight line included in the common frame as a collinear straight line.

In a fourth aspect, an embodiment of the present application further provides a decompression apparatus for a laser point cloud, including:

the information acquisition unit is used for acquiring compressed information of the laser point cloud frame; the compressed information comprises first compressed information corresponding to each target straight line and second compressed information corresponding to each area; the target straight line is determined in the laser point cloud frame by a compression end based on the position of a laser scanning point in the laser point cloud frame; the region is obtained by dividing based on the target straight line;

the information decompression unit is used for determining the point cloud serial numbers of the laser scanning points on the target straight line according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points on the target straight line, determining the scanning angles corresponding to the point cloud serial numbers of the laser scanning points on the target straight line according to the association relationship between the point cloud serial numbers and the scanning angles of the laser scanning points, and determining the positions of the laser scanning points on the target straight line in the laser point cloud frame based on the straight line information of the target straight line and the scanning angles of the laser scanning points; and

In an optional embodiment, the information decompression unit is specifically configured to:

In a fifth aspect, the present application further provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, the method for compressing a laser point cloud of the first aspect is implemented.

In a sixth aspect, the present application further provides an electronic device, including a memory and a processor, where the memory stores a computer program executable on the processor, and when the computer program is executed by the processor, the processor is enabled to implement the method for compressing the laser point cloud of the first aspect.

In a seventh aspect, an embodiment of the present application further provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method for decompressing a laser point cloud in the second aspect is implemented.

In an eighth aspect, the present application further provides an electronic device, which includes a memory and a processor, where the memory stores a computer program that is executable on the processor, and when the computer program is executed by the processor, the processor is enabled to implement the method for decompressing the laser point cloud of the second aspect.

For technical effects brought by any one implementation manner of the third aspect, the fifth aspect, and the sixth aspect, reference may be made to technical effects brought by a corresponding implementation manner of the first aspect, and details are not described here.

For technical effects brought by any one implementation manner of the fourth aspect, the seventh aspect, and the eighth aspect, reference may be made to technical effects brought by a corresponding implementation manner of the second aspect, and details are not repeated here.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.

Fig. 1 is a flowchart of a method for compressing a laser point cloud according to an embodiment of the present disclosure;

fig. 2 is a schematic diagram illustrating a determination of a target straight line in a laser point cloud frame according to an embodiment of the present disclosure;

fig. 3 is a flowchart of another method for compressing a laser point cloud according to an embodiment of the present disclosure;

fig. 4 is a flowchart of a decompression method of a laser point cloud according to an embodiment of the present disclosure;

fig. 5 is a flowchart of another method for decompressing a laser point cloud according to an embodiment of the present disclosure;

fig. 6 is a schematic structural diagram of a compressing apparatus for laser point cloud according to an embodiment of the present disclosure;

fig. 7 is a schematic structural diagram of a decompression device for laser point cloud according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be described in further detail with reference to the accompanying drawings, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

It should be noted that references to the terms "comprising" and "having," and variations thereof, in the context of this application are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

The embodiment of the application provides a compression method of laser point cloud, as shown in fig. 1, comprising the following steps:

step S101, determining a target straight line in the laser point cloud frame based on the position of the laser scanning point in the laser point cloud frame.

The number of the laser scanning points on the target straight line is larger than a first set number threshold, the point cloud serial numbers of the laser scanning points on the target straight line are continuous, the point cloud serial numbers of the laser scanning points are set based on a scanning sequence, and the laser scanning points on different target straight lines are different.

Specifically, the laser scanning points included in the laser point cloud frame may be regarded as a point cloud set, and the following operations are cyclically performed until there is a point cloud set in which the number of laser scanning points is smaller than a second set number threshold:

determining a fitting straight line in the point cloud set by a least square method according to the positions of the laser scanning points in the point cloud set, and determining target scanning points meeting set distance conditions according to the distance between each laser scanning point in the point cloud set and the fitting straight line; dividing the laser scanning points located in front of the point cloud serial numbers of the target scanning points and the target scanning points into one point cloud set, and dividing the laser scanning points located behind the point cloud serial numbers of the target scanning points into another point cloud set.

The least squares method is a mathematical optimization technique. It finds the best functional match of the data by minimizing the sum of the squares of the errors. The parameters of the linear equation are determined by the least square method, and the sum of squares of the errors between these determined data and the actual data is minimized. Least squares are fitting n points to minimize the total error from the fitted line. The solution is to minimize the vertical error of each point to the line.

Each laser scanning point is located in one point cloud set, and the point cloud set comprises the laser scanning points with continuous sequence numbers.

After laser scanning points included in the laser point cloud frame are divided into a plurality of point cloud sets, a target straight line corresponding to the target point cloud set is determined for any target point cloud set with the number of the laser scanning points larger than a second set number threshold, wherein the target straight line is the straight line with the minimum sum of distances between the target straight line and all the laser scanning points included in the target point cloud set, or the target straight line is the straight line containing the laser scanning points in the target point cloud set at the maximum.

Step S102, using first compressed information corresponding to each target straight line and second compressed information corresponding to each area as compressed information of a laser point cloud frame, wherein the first compressed information corresponding to the target straight line comprises straight line information of the target straight line, and the maximum value and the minimum value of point cloud serial numbers of laser scanning points on the target straight line; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number.

The area is obtained by dividing based on a target straight line, and the scanning distance of the laser scanning point is the distance between the laser scanning point and a laser emitting device of the laser radar.

For example, as shown in fig. 2, the laser point cloud frame includes 39 laser scanning points in total, and the number near each laser scanning point in fig. 2 represents the point cloud number of the corresponding laser scanning point, and the point cloud numbers are arranged according to the scanning order of the laser scanning points. Through the straight line determination mode, 3 item calibration straight lines including L1, L2 and L3 can be determined in the laser point cloud frame, based on the three items calibration straight lines, laser scanning points in the laser point cloud frame can be divided into 3 target straight lines and 2 areas respectively, namely the laser scanning points 1-13 can be divided into the area 1, the laser scanning points 14-23 can be divided into the target straight line L1, the laser scanning points 24-29 can be divided into the target straight line L2, the laser scanning points 30-33 can be divided into the area 2, and the laser scanning points 34-39 can be divided into the target straight line L3.

Then, respectively obtaining compressed information of the laser point cloud frame for the laser scanning points in 3 target straight lines and 2 areas, wherein the second compressed information corresponding to the area 1 comprises scanning distances of the laser scanning points 1-13, 1 and 13; the first compressed information corresponding to the target straight line L1 includes straight line information of the target straight line L1, 14 and 23; the first compressed information corresponding to the target straight line L2 includes the straight line information of the target straight line L2, 24, and 29; the second compression information corresponding to the area 2 comprises scanning distances of the laser scanning points 30-33, 30 and 33; the first compressed information corresponding to the target straight line L3 includes the straight line information of the target straight line L3, 34, and 39.

The straight line information of the target straight line comprises straight line parameters and straight line serial numbers, and the straight line serial numbers of the straight line information of the target straight line are set based on the sequence of the collected laser point cloud frames. Or the target straight line comprises a collinear straight line and a non-collinear straight line, the straight line information of the non-collinear straight line comprises straight line parameters and a straight line serial number, and the straight line information of the collinear straight line comprises a straight line serial number of the corresponding non-collinear straight line.

Wherein a collinear straight line of the target straight lines is determined by:

randomly selecting a laser point cloud frame from a plurality of continuous laser point cloud frames as a key frame, and using other laser point cloud frames in the plurality of continuous laser point cloud frames as common frames, wherein the linear sequence number included in the linear information of the non-collinear linear is set based on the sequence of the collected laser point cloud frames in the plurality of laser point cloud frames;

for each target straight line included in the normal frame, the following operations are respectively performed:

projecting the target straight line included by the common frame to the reference frame of the key frame, and respectively determining the distance between the target straight line included by the common frame and at least one target straight line included by the key frame; if the distance between the target straight line included in the common frame and the key target straight line in the at least one target straight line included in the key frame meets the set threshold condition, taking the target straight line included in the common frame as a collinear straight line;

the target straight line satisfying the set collinear condition with the collinear straight line is a non-collinear straight line corresponding to the collinear straight line.

For example, the normal frame includes A3-entry calibration line including a1, a2 and A3, the key frame includes A3-entry calibration line including B1, B2 and B3, the target lines a1, a2 and A3 are respectively projected under the reference system of the key frame, and distances d11, d12 and d13 between the target line a1 and the target line B1, the target line B2 and the target line B3 are respectively determined; distances d21, d22 and d23 between the target straight line a2 and the target straight line B1, the target straight line B2 and the target straight line B3, respectively; distances d31, d32 and d33 between the target straight line a3 and the target straight line B1, the target straight line B2 and the target straight line B3, respectively. If the distance d11 between the target straight line a1 and the target straight line B1 satisfies the set threshold condition, it may be determined that the target straight line a1 is collinear with the target straight line B1, and the target straight line a1 is taken as a collinear straight line of the target straight line B1.

The threshold setting condition is that the first point-line distance is smaller than the second set threshold distance, and the second point-line distance is smaller than the second set threshold distance. The first point line distance is the distance between a laser scanning point corresponding to the maximum value of the point cloud serial number on a target straight line included in the common frame and a key target straight line; the second wire distance is the distance between the laser scanning point corresponding to the minimum value of the point cloud serial numbers on the target straight lines included in the common frame and the key target straight line.

In some embodiments, the method for compressing the laser point cloud provided by the present application is suitable for an AGV using 2D laser navigation, and the operating environment thereof is an indoor structured environment. The AGV is usually equipped with a 2D laser, a wheel-type odometer, and an Inertial Measurement Unit (IMU) for calculating and positioning, and can output the pose of each frame of laser point cloud in the same map reference system

Where (x, y) is the 2D position and θ is the rotation angle.

Specifically, the method for compressing the laser point cloud provided by the embodiment of the present application may also be implemented according to the process shown in fig. 3, as shown in fig. 3, including the following steps:

step S301, determining a target straight line in the laser point cloud frame based on the position of the laser scanning point in the laser point cloud frame.

One frame of 2D laser point cloud f ═ p ₁ ,p ₂ ,…,p _n N laser scanning points are included, and each scanning point describes a specific 2D coordinate. Laser point clouds generally have two common coordinate representations: cartesian coordinate system, polar coordinate system. In the embodiment of the present application, the laser point cloud may be represented by a polar coordinate system, that is, each laser scanning point in the laser point cloud f may be represented as p _i ＝{θ _i ,r _i In which θ _i The scanning angle is the angle between the connecting line between the laser scanning point and the laser emitting device of the laser radar and the horizontal plane of the laser emitting device, r _i The scanning distance is the distance between the laser scanning point and the laser emitting device of the laser radar. The polar coordinate system may be converted to a Cartesian coordinate system by the following equation:

since the laser radar is usually scanned at an equal angular resolution, the difference between the scanning angles corresponding to adjacent laser scanning points in the laser point cloud f is a fixed value θ _inc And starting angle theta of laser radar scanning _min End angle theta _max Are all known. The reference system fid and the scanning starting angle theta of the laser point cloud are generally set _min Angle of scan end theta _max Angular resolution theta _inc Minimum scanning distance r _min Maximum scanning distance r _max Frame head of the laser point cloud is combined: h ═ fid, θ _min ,θ _max ,θ _inc ,r _min ,r _max }. Therefore, the laser point cloud can also be represented as f ═ { H | r ₁ ,r ₂ ,…,r _n Central point p of _i Has a scanning angle of theta _min +(i-1)·θ _inc 。

And determining a target straight line in the laser point cloud frame according to the position of the laser scanning point in the laser point cloud frame. And the determined target straight line simultaneously meets the following three conditions: the method comprises the following steps that firstly, laser scanning points on different target straight lines are different; secondly, the number of laser scanning points on the target straight line is larger than a first set number threshold; and thirdly, the point cloud serial numbers of the laser scanning points on the target straight line are continuous. And the point cloud serial numbers of the laser scanning points in the laser point cloud frame are set based on the scanning sequence.

Since the straight-line equation of the target straight line is ax + by + c is 0, the straight-line parameter of the target straight line can be expressed as l ═ a, b, c]So that each target straight line determined in the laser point cloud frame can be expressed as { l } ₁ ,l ₂ ,…,l _m }。

Step S302, according to the determined target straight lines, dividing laser scanning points in the laser point cloud frame into a straight line point cloud section and a non-straight line point cloud section, and carrying out intra-frame coding on the laser point cloud frame based on the obtained straight line point cloud section and the non-straight line point cloud section to obtain compressed information of the laser point cloud frame.

After each target straight line in the laser point cloud frame is determined, the laser scanning points positioned on the target straight line and the laser scanning points not positioned on the target straight line can be correspondingly obtainedThe laser scanning spot of (1). By utilizing the equiangular resolution scanning characteristic of the laser point cloud and combining the obtained straight line result, all laser scanning points in the laser point cloud frame can be segmented in sequence to obtain f ═ s ₁ ,s ₂ ,…,s _h }. Where s represents a segment, including two types: a linear point cloud segment and a non-linear point cloud segment.

The linear point cloud segment may be represented as s _h ＝{i,j|lid _h I, j represents the start sequence number and the end sequence number of the laser scanning point, lid _h The linear serial number of the target linear to which the laser scanning point belongs is represented; the non-linear point cloud segment may be represented as s _h ＝{i,j|r _i ,r _i+1 ,…,r _j Wherein i, j represents the starting sequence number and the ending sequence number of the laser scanning point, r _i Is the scan distance.

The laser point cloud frame is subjected to intra-frame coding, and the obtained compression information of the laser point cloud frame can be as follows:

f＝{H|l ₁ ,l ₂ ,…,l _m |s ₁ ,s ₂ ,…,s _h }

wherein l _m ＝[a _m ,b _m ,c _m ]，

Step S303, caching the compressed information of the laser point cloud frame.

Step S304, determining whether the number of the laser point cloud frames continuously cached reaches a set frame number threshold; if yes, go to step S305; if not, step S303 is performed.

Because the compression ratio of a single laser point cloud frame is relatively limited, in order to more fully utilize the characteristic of high coincidence degree of the laser point clouds in a plurality of continuous laser point cloud frames, the continuous laser point cloud frames need to be further compressed together, so that the obtained compressed information of the laser point cloud frames can be cached firstly, and the next step of operation is carried out after the number of the laser point cloud frames cached continuously reaches a set frame number threshold N frame.

Step S305, randomly selecting a laser point cloud frame from a plurality of continuous laser point cloud frames reaching a set frame number threshold as a key frame, and using other laser point cloud frames in the plurality of continuous laser point cloud frames as common frames.

When the number of the laser point cloud frames which are continuously cached reaches a set frame number threshold, one laser point cloud frame can be randomly selected from the plurality of continuous laser point cloud frames to serve as a key frame, and other laser point cloud frames in the plurality of continuous laser point cloud frames are used as common frames.

Preferably, the first one can be selected from the N laser point cloud frames buffered continuously

Frame as Key frame, noted f _K The rest frames are common frames and are marked as f _t . Wherein, the key frame f _K The corresponding pose is xi _K Normal frame f _t The corresponding pose is xi _t . And recording continuous N laser point cloud frames as a point cloud group G { { ξ { (xi) } ₁ ,f ₁ },…,{ξ _K ,f _K },…,{ξ _N ,f _N }}。

And step S306, performing linear fusion on each target straight line included in the common frame and each target straight line included in the key frame respectively, and determining a linear fusion result.

In a group of point cloud frames G, because the frames are continuous, target straight lines between the common frames and the key frames are overlapped, so that straight line fusion can be performed. The method for linear fusion comprises the following steps:

(1) obtaining a common frame f in G _t ；

(2) In the normal frame f _t Obtaining a target straight line

And point cloud segments belonging to the target straight line are obtained.

Projecting the target straight line to the key frame f _K May be:

determining the polar coordinates of the first and the last laser scanning points of the point cloud segment belonging to the target straight line

And

then, it can be projected to the key frame f separately _K The reference system of (a) is obtained:

(3) at key frame f _K Obtaining a target straight line

And separately calculate

Is in line with the target

The distance between them.

First, a laser beam is scanned to a spot

And

converting the polar coordinate system into a Cartesian coordinate system to respectively obtain corresponding Cartesian coordinates

And

then, when the target is straight

OfEquation of a line of

Time, laser scanning spot

And

to the target straight line

May be a distance of

If d is satisfied _b <th _overlap ，d _e <th _overlap Condition, then determining

Collinearity and adding to a collinearity look-up Table _overlap Of, wherein, th _overlap Is a preset distance threshold. If the condition is not met, the next target straight line of the key frame is continuously traversed until the condition is met or all the target straight lines included in the key frame are completely traversed.

(4) And (4) repeating the step (2) and the step (3) until all the target straight lines included in the common frame and all the target straight lines included in the key frame are subjected to straight line fusion.

(5) And (4) repeating the steps (1) to (4) until all the target straight lines included by all the common frames in the G and all the target straight lines included by the key frames are subjected to straight line fusion.

And S307, performing interframe coding on the continuous multiple laser point cloud frames according to the obtained linear fusion result to obtain compressed information of the continuous multiple laser point cloud frames.

After all the target straight lines included in all the common frames in G are subjected to straight line fusion with all the target straight lines included in the key frame, interframe coding can be further performed on each laser point cloud frame in G:

(1) the key frame f _K All included target lines are recorded within G.

(2) Obtaining a common frame f in G _t Go through all its projections to the key frame f _K Target straight line under reference system, not in collinear look-up Table _overlap Is recorded in G and will be in the collinear look-up Table Table _overlap Key frame f for target straight line in (1) _K The corresponding target straight line inside is replaced.

(3) Modifying normal frames f _t The compression information of (2): and (3) removing the frame header H and the linear information of all target straight lines, traversing all point cloud segments, modifying the linear serial number associated with the linear point cloud segment into the linear serial number modified in the step (2), and not processing the non-linear point cloud segment.

(4) And (4) repeating the step (2) and the step (3) until all the common frames in the G are processed.

Performing interframe coding on a plurality of continuous laser point cloud frames in the point cloud group G to obtain compression information of the plurality of continuous laser point cloud frames in the point cloud group G as follows:

G＝{l ₁ ,l ₂ ,…,l _q |{ξ ₁ ,f ₁ },…,{ξ _K ,f _K },…,{ξ _N ,f _N }}

wherein, { l ₁ ,l ₂ ,…,l _q All are keyframe reference frames.

Step S308, determining whether all the laser point cloud frames are completely compressed; if yes, go to step S309; if not, step S303 is performed.

Step S309, add frame header information to the compressed information of all laser point cloud frames.

If all the laser point cloud frames are not compressed, the steps S303 to S307 are repeated. If all the laser point cloud frames are completely compressed, frame header information H can be added to the finally obtained compressed information of all the laser point cloud frames, and the obtained compressed information of all the laser point cloud frames PC is: PC ═ H | G ₁ ,G ₂ ,…,G _N }。

Compared with the prior art, the method for compressing the laser point cloud can extract a straight line from a single laser point cloud frame, replace original laser scanning points with the straight line, greatly reduce point cloud segmentation information on the straight line into a state that only the sequence number of a segmentation starting laser scanning point, the sequence number of a termination laser scanning point and the sequence number of the straight line are reserved by utilizing the scanning characteristic of the equal-angular resolution of a laser radar, and accordingly achieve efficient compression of the laser point cloud in the frame. Meanwhile, the method for compressing the laser point cloud can also utilize the characteristic of high coincidence degree of the laser point cloud in continuous laser point cloud frames to divide a point cloud sequence, namely all the laser point cloud frames into point cloud groups with fixed frame numbers, select key frames in the point cloud groups, project straight lines in common frames to a key frame reference system and perform straight line fusion, so that the number of the straight lines is further reduced, the compression efficiency in the time sense is improved, and the efficient compression of the laser point cloud between frames is realized.

After compressing the laser point cloud, correspondingly, an embodiment of the present application further provides a method for decompressing the laser point cloud, as shown in fig. 4, including the following steps:

step S401, acquiring compressed information of a laser point cloud frame; the compressed information includes first compressed information corresponding to each target straight line and second compressed information corresponding to each region.

The target straight line is determined in the laser point cloud frame by the position of the compression end on the basis of the laser scanning point in the laser point cloud frame, and the region is obtained by dividing based on the target straight line.

Step S402, according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points on the target straight line, which are included in the first compressed information, the point cloud serial numbers of the laser scanning points on the target straight line are determined, according to the incidence relation between the point cloud serial numbers and the scanning angles of the laser scanning points, the scanning angles corresponding to the point cloud serial numbers of the laser scanning points on the target straight line are determined, and based on the straight line information of the target straight line and the scanning angles of the laser scanning points, the positions of the laser scanning points on the target straight line in the laser point cloud frame are determined.

Specifically, after the scanning angle corresponding to the point cloud number of each laser scanning point on the target straight line is determined, if the straight line information of the target straight line includes the straight line parameter and the straight line number, the position of each laser scanning point on the target straight line in the laser point cloud frame is determined based on the straight line parameter of the target straight line and the scanning angle of each laser scanning point.

And if the line information of the target line comprises a line serial number, determining that the target line is a collinear line, determining a line parameter of the non-collinear line according to the non-collinear line corresponding to the line serial number, and determining the position of each laser scanning point on the target line in the laser point cloud frame based on the line parameter of the non-collinear line and the scanning angle of each laser scanning point. And the collinear straight line and the corresponding non-collinear straight line meet the set collinear condition.

Step S403, determining the point cloud number of each laser scanning point in the area according to the maximum point cloud number and the minimum point cloud number of the laser scanning point located in the area, which are included in the second compressed information, and determining the scanning angle corresponding to the point cloud number of each laser scanning point in the area according to the association relationship, and determining the position of each laser scanning point located in the area in the laser point cloud frame based on the determined scanning angle and the scanning distance of each laser scanning point included in the second compressed information.

In some embodiments, corresponding to the step S309 of adding the compressed information of the frame header information, the method for decompressing a laser point cloud provided in the embodiment of the present application may also be implemented according to the process shown in fig. 5, as shown in fig. 5, including the following steps:

step S501, acquiring compressed information of all laser point cloud frames, and reading frame header information from the compressed information.

The compressed laser point cloud sequence shares the same frame header information, so that the reading is only needed once, and the frame header information is H ═ fid, theta _min ,θ _max ,θ _inc ,r _min ,r _max }。

Step S502, reading the compressed information of a plurality of continuous laser point cloud frames reaching the set frame number threshold.

Reading compressed information G ═ l of a plurality of continuous laser point cloud frames in the point cloud group G ₁ ,l ₂ ,…,l _q |{ξ ₁ ,f ₁ },…,{ξ _K ,f _K },…,{ξ _N ,f _N }}。

Step S503, reading the compression information of a single laser point cloud frame in the continuous multiple laser point cloud frames.

And reading the compression information { ξ) of a single laser point cloud frame from the point cloud group G _t ,f _t }。

Step S504, according to the compression information of the single laser point cloud frame, the position of each laser scanning point in the laser point cloud frame is determined.

The process of determining the position of each laser scanning point in the laser point cloud frame from the compressed information of the single laser point cloud frame is as follows:

(1) one point cloud segment s of a single laser point cloud frame is read.

(2) If the point cloud segment s is a nonlinear point cloud segment, i.e. s ═ i, j | r _i ,r _i+1 ,…,r _j Directly calculating the coordinates of each laser scanning point in the non-linear point cloud segment:

x _i ＝r _i ·cos(θ _min +(i-1)·θ _inc )

y _i ＝r _i ·sin(θ _min +(i-1)·θ _inc )

(3) if the point cloud segment s is a straight-line point cloud segment, i.e. s ═ i, j | lid _h In which lid _h The linear serial number of the target linear under the key frame reference system can be firstly passed through the formula

Projecting the target straight line to a reference system of a current laser point cloud frame, and then calculating the coordinates of each laser scanning point on the target straight line:

θ _i ＝θ _min +(i-1)·θ _inc

(4) and (4) repeating the step (2) and the step (3) until all point cloud segments in the single laser point cloud frame are decompressed.

Determining whether all the laser point cloud frames in the point cloud group G are completely decompressed, if not, repeatedly executing the step S503 and the step S504 until all the laser point cloud frames in the point cloud group G are completely decompressed, determining whether all the point cloud groups G in the point cloud sequence PC are completely decompressed, and if not, repeating the step S502 to the step S504 until all the point cloud groups G in the point cloud sequence PC are completely decompressed.

Compared with the prior art, the decompression method of the laser point cloud can determine to recover the laser scanning points from the original scanning distance or from the straight lines in the point cloud group according to the point cloud segmentation categories, so that the decompression of the laser point cloud can be completed quickly and efficiently.

The laser point cloud compression method shown in fig. 1 is based on the same inventive concept, and the embodiment of the application also provides a laser point cloud compression device. Because the device is a device corresponding to the laser point cloud compression method, and the principle of solving the problems of the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and repeated parts are not described again.

Fig. 6 shows a schematic structural diagram of a compressing apparatus for laser point cloud provided in an embodiment of the present application, and as shown in fig. 6, the compressing apparatus for laser point cloud includes a straight line extracting unit 601 and an information compressing unit 602.

The line extraction unit 601 is configured to determine a target line in a laser point cloud frame based on the positions of laser scanning points in the laser point cloud frame, where the number of laser scanning points on the target line is greater than a first set number threshold, point cloud serial numbers of the laser scanning points on the target line are consecutive, the point cloud serial numbers of the laser scanning points are set based on a scanning sequence, and the laser scanning points on different target lines are different;

an information compression unit 602, configured to use first compression information corresponding to each target straight line and second compression information corresponding to each area as compression information of a laser point cloud frame, where the areas are obtained by dividing based on the target straight line, and the first compression information corresponding to the target straight line includes straight line information of the target straight line, a maximum point cloud serial number value and a minimum point cloud serial number value of a laser scanning point located on the target straight line; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number; the scanning distance of the laser scanning point is the distance between the laser scanning point and a laser emitting device of the laser radar.

In an alternative embodiment, the straight line extracting unit 601 is further configured to:

dividing laser scanning points included in a laser point cloud frame into a plurality of point cloud sets, wherein each laser scanning point is located in one point cloud set, and each point cloud set comprises laser scanning points with continuous sequence numbers;

aiming at any one target point cloud set with the number of the laser scanning points larger than a second set number threshold, determining a target straight line corresponding to the target point cloud set, wherein the target straight line is the straight line with the minimum sum of distances between the laser scanning points included in the target point cloud set; or the target straight line is the straight line which contains the most laser scanning points in the target point cloud set.

taking laser scanning points included in the laser point cloud frame as a point cloud set, and executing the following operations in a circulating manner until the point cloud set with the number of the laser scanning points smaller than a second set number threshold exists:

determining a fitting straight line in the point cloud set by a least square method according to the positions of the laser scanning points in the point cloud set, and determining target scanning points meeting a set distance condition according to the distance between each laser scanning point in the point cloud set and the fitting straight line;

dividing the laser scanning points located in front of the point cloud serial numbers of the target scanning points and the target scanning points into one point cloud set, and dividing the laser scanning points located behind the point cloud serial numbers of the target scanning points into another point cloud set.

In an optional embodiment, the line information of the target line includes line parameters and a line number, and the line number included in the line information of the target line is set based on the sequence of the acquired laser point cloud frames; or

wherein a collinear line of the target lines is determined by:

and taking the target straight line which meets the set collinear condition with at least one target straight line included in the key frame as a collinear straight line, wherein the target straight line which meets the set collinear condition with the collinear straight line is a non-collinear straight line corresponding to the collinear straight line.

In an alternative embodiment, the information compressing unit 602 is further configured to:

projecting the target straight line included by the common frame to the reference frame of the key frame, and respectively determining the distance between the target straight line included by the common frame and at least one target straight line included by the key frame;

and if the distance between the target straight line included in the common frame and the key target straight line in at least one target straight line included in the key frame meets the set threshold condition, taking the target straight line included in the common frame as a collinear straight line.

In an alternative embodiment, the threshold condition is set such that the first dot-line distance is less than a second set threshold distance, and the second dot-line distance is less than the second set threshold distance;

the first point-line distance is the distance between a laser scanning point corresponding to the maximum value of the point cloud serial number on a target straight line included in the common frame and a key target straight line; the second wire distance is the distance between the laser scanning point corresponding to the minimum value of the point cloud serial number on the target straight line included in the common frame and the key target straight line.

The laser point cloud decompression method shown in fig. 4 is based on the same inventive concept, and the embodiment of the application also provides a laser point cloud decompression device. Because the device is a device corresponding to the decompression method of the laser point cloud, and the principle of solving the problems of the device is similar to that of the method, the implementation of the device can refer to the implementation of the method, and repeated parts are not described again.

Fig. 7 shows a schematic structural diagram of a decompression apparatus for laser point cloud provided in an embodiment of the present application, and as shown in fig. 7, the decompression apparatus for laser point cloud includes an information acquisition unit 701 and an information decompression unit 702.

The information acquisition unit 701 is used for acquiring compressed information of a laser point cloud frame; the compressed information comprises first compressed information corresponding to each target straight line and second compressed information corresponding to each area; the target straight line is determined in the laser point cloud frame by the position of the compression end on the basis of the laser scanning point in the laser point cloud frame; the regions are obtained by dividing based on the target straight line;

the information decompression unit 702 is configured to determine a point cloud number of each laser scanning point on the target straight line according to a point cloud number maximum value and a point cloud number minimum value of the laser scanning point located on the target straight line, which are included in the first compressed information, determine a scanning angle corresponding to the point cloud number of each laser scanning point located on the target straight line according to an association relationship between the point cloud number of each laser scanning point and the scanning angle, and determine a position of each laser scanning point located on the target straight line in the laser point cloud frame based on the straight line information of the target straight line and the scanning angle of each laser scanning point; and

and determining the point cloud serial numbers of the laser scanning points in the area according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points in the area, which are included in the second compressed information, determining the scanning angle corresponding to the point cloud serial numbers of the laser scanning points in the area according to the association relation, and determining the position of each laser scanning point in the area in the laser point cloud frame based on the determined scanning angle and the scanning distance of each laser scanning point included in the second compressed information.

In an alternative embodiment, the information decompressing unit 702 is specifically configured to:

if the straight line information of the target straight line comprises straight line parameters and a straight line serial number, determining the positions of the laser scanning points on the target straight line in the laser point cloud frame based on the straight line parameters of the target straight line and the scanning angles of the laser scanning points;

and if the line information of the target line comprises a line serial number, determining that the target line is a collinear line, determining the line parameters of the non-collinear line according to the non-collinear line corresponding to the line serial number, and determining the position of each laser scanning point on the target line in a laser point cloud frame based on the line parameters of the non-collinear line and the scanning angle of each laser scanning point, wherein the collinear line and the corresponding non-collinear line meet the set collinear condition.

The electronic equipment is based on the same inventive concept as the method embodiment, and the embodiment of the application also provides the electronic equipment. The electronic device may be used to compress or decompress a laser point cloud. In one embodiment, the electronic device may be a server, a terminal device, or other electronic devices. In this embodiment, the electronic device may be configured as shown in fig. 8, and include a memory 801, a communication module 803, and one or more processors 802.

A memory 801 for storing computer programs executed by the processor 802. The memory 801 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, programs required for running an instant messaging function, and the like; the storage data area can store various instant messaging information, operation instruction sets and the like.

The memory 801 may be a volatile memory (volatile memory), such as a random-access memory (RAM); the memory 801 may also be a non-volatile memory (non-volatile memory) such as, but not limited to, a read-only memory (rom), a flash memory (flash memory), a Hard Disk Drive (HDD) or a solid-state drive (SSD), or the memory 801 may be any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. The memory 801 may be a combination of the above memories.

The processor 802 may include one or more Central Processing Units (CPUs), a digital processing unit, and the like. The processor 802 is configured to implement the above-described compression method or decompression method of the laser point cloud when calling the computer program stored in the memory 801.

The communication module 803 is used for communicating with the terminal device and other servers.

The embodiment of the present application does not limit the specific connection medium among the memory 801, the communication module 803 and the processor 802. In fig. 8, the memory 801 and the processor 802 are connected by a bus 804, the bus 804 is represented by a thick line in fig. 8, and the connection manner between other components is merely illustrative and not limited. The bus 804 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 8, but this is not intended to represent only one bus or type of bus.

According to an aspect of the application, a computer program product or computer program is provided, comprising computer instructions, the computer instructions being stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to cause the computer device to execute the compression method or the decompression method of the laser point cloud in the above embodiments.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The above description is only for the specific embodiments of the present application, but the scope of the present application 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 application, and shall be covered by the scope of the present application.

Claims

1. A method for compressing a laser point cloud, comprising:

taking first compression information corresponding to each target straight line and second compression information corresponding to each area as compression information of the laser point cloud frame, wherein the areas are obtained by dividing based on the target straight line, and the first compression information corresponding to the target straight line comprises straight line information of the target straight line, the maximum point cloud serial number value and the minimum point cloud serial number value of laser scanning points on the target straight line; the second compressed information corresponding to the area comprises the scanning distance of each laser scanning point in the area, the maximum value of the point cloud serial number and the minimum value of the point cloud serial number; and the scanning distance of the laser scanning point is the distance between the laser scanning point and a laser emitting device of the laser radar.

2. The method of claim 1, wherein determining a target straight line in the laser point cloud frame based on a location of a laser scanning point in the laser point cloud frame comprises:

3. The method of claim 2, wherein dividing the laser scanning points comprised by the laser point cloud frame into a plurality of point cloud sets comprises:

4. The method of claim 1, wherein the line information of the target line includes line parameters and line numbers, the line information of the target line including line numbers being set based on an order of the acquired laser point cloud frames; or

wherein a collinear line of the target lines is determined by:

randomly selecting a laser point cloud frame from a plurality of continuous laser point cloud frames as a key frame, and using other laser point cloud frames in the plurality of continuous laser point cloud frames as common frames, wherein the linear sequence number included in the linear information of the non-collinear linear is set based on the sequence of the acquired laser point cloud frames in the plurality of laser point cloud frames;

5. The method according to claim 4, wherein the regarding, as collinear straight lines, target straight lines among the target straight lines included in the normal frame, which satisfy a set collinear condition with at least one target straight line included in the key frame, comprises:

6. The method of claim 5, wherein the set threshold condition is that a first point-to-line distance is less than a second set threshold distance, and the second point-to-line distance is less than a second set threshold distance;

the first point line distance is the distance between a laser scanning point corresponding to the maximum value of the point cloud serial number on a target straight line included in the common frame and the key target straight line; and the second wire distance is the distance between a laser scanning point corresponding to the minimum value of the point cloud serial number on the target straight line included in the common frame and the key target straight line.

7. A decompression method of laser point cloud is characterized by comprising the following steps:

according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points in the area, which are included by the second compressed information, the point cloud serial numbers of the laser scanning points in the area are determined, the scanning angles corresponding to the point cloud serial numbers of the laser scanning points in the area are determined according to the association relation, and the positions of the laser scanning points in the area in the laser point cloud frame are determined based on the determined scanning angles and the scanning distances of the laser scanning points, which are included by the second compressed information.

8. The method of claim 7, wherein the determining the position of each laser scanning point located on the target straight line in the laser point cloud frame based on the straight line information of the target straight line and the scanning angle of each laser scanning point comprises:

9. A compression device of laser point cloud is characterized by comprising:

10. A decompression device of laser point cloud is characterized by comprising:

the information decompression unit is used for determining the point cloud serial numbers of the laser scanning points on the target straight line according to the point cloud serial number maximum value and the point cloud serial number minimum value of the laser scanning points on the target straight line, determining the scanning angles corresponding to the point cloud serial numbers of the laser scanning points on the target straight line according to the incidence relation between the point cloud serial numbers of the laser scanning points and the scanning angles, and determining the positions of the laser scanning points on the target straight line in the laser point cloud frame based on the straight line information of the target straight line and the scanning angles of the laser scanning points; and

11. A computer-readable storage medium having a computer program stored therein, the computer program characterized by: the computer program, when executed by a processor, implements the method of any of claims 1-8.

12. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program operable on the processor, the computer program, when executed by the processor, implementing the method of any of claims 1-8.