A point cloud processing method, apparatus and electronic device

Abstract

This application discloses a point cloud processing method, apparatus, and electronic device, relating to the field of data processing technology, to solve the problem of low efficiency in existing point cloud encoding. The method includes: acquiring radar point cloud data; generating a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; determining a point cloud number range parameter for dividing the ground point cloud according to the Z-axis distribution histogram; and dividing ground point cloud strips from the radar point cloud data according to the point cloud number range parameter. The embodiments of this application can ensure the reasonable division of ground point cloud data from radar point cloud data, which is beneficial to improving point cloud encoding efficiency.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a point cloud processing method, apparatus and electronic device. Background Technology

[0002] A 3D point cloud is composed of a large amount of positional information and corresponding attribute information (such as color, reflectivity, normal vector, etc.), which can flexibly represent the spatial structure and surface properties of 3D objects or scenes.

[0003] Currently, massive point cloud data faces significant challenges in encoding and transmission, resulting in low point cloud encoding efficiency. Summary of the Invention

[0004] This application provides a point cloud processing method, apparatus, and electronic device to solve the problem of low efficiency in existing point cloud encoding.

[0005] In a first aspect, embodiments of this application provide a point cloud processing method, including:

[0006] Acquire radar point cloud data;

[0007] Based on the Z-axis coordinate data of the radar point cloud data, a Z-axis distribution histogram of the radar point cloud data is generated, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds;

[0008] Based on the Z-axis distribution histogram, determine the range of point cloud quantity parameters used to divide the ground point cloud;

[0009] Based on the point cloud quantity range parameter, ground point cloud strips are divided from the radar point cloud data.

[0010] Optionally, determining the point cloud quantity range parameter for dividing the ground point cloud based on the Z-axis distribution histogram includes:

[0011] Based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, a third Z-axis interval is determined, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval;

[0012] Based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval, the point cloud quantity range parameter used to divide the ground point cloud is determined.

[0013] Optionally, the first peak value is the highest peak value, and the second peak value is the second highest peak value;

[0014] The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

[0015] Optionally, the first low peak value is the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram.

[0016] Optionally, determining the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval includes:

[0017] The minimum number of point clouds to divide the ground point cloud is determined by m times the first low peak value, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0018] The second peak value is determined as the maximum number of point clouds used to divide the ground point cloud.

[0019] Optionally, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold;

[0020] The step of dividing ground point cloud stripes from the radar point cloud data according to the point cloud quantity range parameter includes:

[0021] The Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold is determined as the first target Z-axis interval.

[0022] The point cloud data located in the Z-axis region of the first target in the radar point cloud data is divided into ground point cloud strips.

[0023] Optionally, after dividing the radar point cloud data located in the Z-axis interval of the first target into ground point cloud strips, the method further includes:

[0024] The first target Z-axis interval is divided into N sub-intervals, and the ground point cloud strip is divided into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

[0025] Optionally, after generating the Z-axis distribution histogram of the radar point cloud data, the method further includes:

[0026] Based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, a second target Z-axis interval is determined for dividing the second type of point cloud, wherein the second type of point cloud is a point cloud located near the radar.

[0027] The point cloud data located in the Z-axis region of the second target in the radar point cloud data is divided into second type point cloud strips.

[0028] Optionally, the method further includes:

[0029] The radar point cloud data, excluding the ground point cloud and the second type of point cloud, is divided into third type of point cloud strips.

[0030] Secondly, embodiments of this application also provide a point cloud processing apparatus, comprising:

[0031] The acquisition module is used to acquire radar point cloud data;

[0032] The generation module is used to generate a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data of the radar point cloud data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds;

[0033] The first determining module is used to determine the range parameter of the number of points in the ground point cloud for dividing the ground point cloud based on the Z-axis distribution histogram.

[0034] The first segmentation module is used to segment ground point cloud stripes from the radar point cloud data according to the point cloud quantity range parameter.

[0035] Optionally, the first determining module includes:

[0036] The first determining submodule is used to determine a third Z-axis interval based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval;

[0037] The second determining submodule is used to determine the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0038] Optionally, the first peak value is the highest peak value, and the second peak value is the second highest peak value;

[0039] The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

[0040] Optionally, the first low peak value is the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram.

[0041] Optionally, the second determining submodule includes:

[0042] The first determining unit is used to determine m times the first low peak value as the minimum point cloud quantity threshold for dividing the ground point cloud, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0043] The second determining unit is used to determine the second peak value as the maximum point cloud quantity threshold for dividing the ground point cloud.

[0044] Optionally, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold;

[0045] The first partitioning module includes:

[0046] The third determining submodule is used to determine the Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold as the first target Z-axis interval.

[0047] The first segmentation submodule is used to divide the radar point cloud data located in the Z-axis interval of the first target into ground point cloud strips.

[0048] Optionally, the first partitioning module further includes:

[0049] The second partitioning submodule is used to divide the first target Z-axis interval into N sub-intervals, and to divide the ground point cloud strip into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

[0050] Optionally, the point cloud processing device further includes:

[0051] The second determining module is used to determine the second target Z-axis interval for dividing the second type of point cloud based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, wherein the second type of point cloud is a point cloud located near the radar.

[0052] The second division module is used to divide the point cloud data located in the Z-axis interval of the second target in the radar point cloud data into a second type of point cloud strip.

[0053] Optionally, the point cloud processing device further includes:

[0054] The third segmentation module is used to segment the radar point cloud data, excluding the ground point cloud and the second type of point cloud, into third type of point cloud strips.

[0055] Thirdly, embodiments of this application also provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the point cloud processing method described above.

[0056] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the point cloud processing method described above.

[0057] In this embodiment, radar point cloud data is acquired; based on the Z-axis coordinate data of the radar point cloud data, a Z-axis distribution histogram of the radar point cloud data is generated, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; according to the Z-axis distribution histogram, a point cloud number range parameter for dividing the ground point cloud is determined; and according to the point cloud number range parameter, ground point cloud strips are divided from the radar point cloud data. Thus, by using the Z-axis distribution histogram of the radar point cloud data, combined with the analysis of the geometric characteristics of the radar point cloud data itself, the point cloud number range parameter for dividing the ground point cloud is determined, thereby ensuring the reasonable division of ground point cloud data from the radar point cloud data and improving point cloud encoding efficiency. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of the point cloud processing method provided in the embodiments of this application;

[0060] Figure 2 This is a histogram of the Z-axis distribution of radar point cloud data provided in the embodiments of this application;

[0061] Figure 3 This is a schematic diagram of point cloud striping provided in an embodiment of this application;

[0062] Figure 4 This is a structural diagram of the point cloud processing device provided in the embodiments of this application;

[0063] Figure 5 This is a structural diagram of the electronic device provided in the embodiments of this application. Detailed Implementation

[0064] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0065] See Figure 1 , Figure 1 This is a flowchart of the point cloud processing method provided in the embodiments of this application, such as... Figure 1 As shown, it includes the following steps:

[0066] Step 101: Obtain radar point cloud data.

[0067] The aforementioned radar point cloud data can refer to three-dimensional point cloud data collected using radar equipment. This three-dimensional point cloud data consists of a large amount of three-dimensional coordinate position information and corresponding attribute information, such as color, reflectivity, and normal vector, which can flexibly characterize the spatial structure and surface attributes of three-dimensional objects or scenes.

[0068] For example, a lidar device can be mounted on a dedicated point cloud acquisition vehicle. The vehicle can then drive on the road section where point cloud data needs to be collected, and the radar can be activated to scan the road section while driving to collect radar point cloud data for that section.

[0069] The acquisition of radar point cloud data described above can be achieved by reading one frame of radar point cloud data at a time from the already collected radar point cloud data and performing subsequent striping processing. In other words, radar point cloud data can be collected and transmitted frame by frame. For example, after collecting each frame of radar point cloud data, that frame of point cloud data is transmitted to the point cloud processing device, so as to perform the point cloud processing method in the embodiments of this application on the radar point cloud data frame by frame.

[0070] Step 102: Based on the Z-axis coordinate data of the radar point cloud data, generate a Z-axis distribution histogram of the radar point cloud data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds.

[0071] The radar point cloud data includes X-axis coordinate data, Y-axis coordinate data, and Z-axis coordinate data of the point cloud. The X-axis coordinate data and Y-axis coordinate data represent the coordinate data of the point cloud in the horizontal direction, while the Z-axis coordinate data represents the coordinate data of the point cloud in the vertical direction.

[0072] In this embodiment of the application, given that the ground point cloud has relatively consistent height coordinates and a large number of data, the distribution characteristics of the Z-axis coordinate data of the radar point cloud data are selected to analyze the radar point cloud data, and then the ground point cloud strips are reasonably divided from it.

[0073] Therefore, in this step, a Z-axis distribution histogram of the radar point cloud data can be generated based on the Z-axis coordinate data of the radar point cloud data. That is, the radar point cloud data can be divided into Z-axis coordinate intervals according to the Z-axis coordinate data of each point, and the number of point clouds in the radar point cloud data located in different Z-axis coordinate intervals can be statistically analyzed, thereby forming a Z-axis distribution histogram that represents the relationship between the Z-axis coordinate interval and the number of point clouds.

[0074] For example, after reading in a frame of point cloud data to be processed, the geometric distribution characteristics of the Z-axis of that frame of point cloud can be analyzed, and the number of points within different Z-axis coordinate ranges can be counted. Let the minimum Z-axis coordinate of the entire frame of point cloud be Min. z The maximum value is Max. z Then the Z-axis coordinate range of the point cloud in this frame is Range. z =Max z -Min z Let the width of each block in the histogram be hist. z Calculate the Z-axis coordinates of the point cloud in this frame in each histogram. z The number of point clouds within the range, i.e. the number of points, can ultimately yield the Z-axis distribution histogram of the point cloud for that frame.

[0075] The Z-axis distribution histogram generated from a frame of radar point cloud data can be as follows: Figure 2 As shown, the horizontal axis of this histogram represents the Z-axis coordinate of the point cloud, and the vertical axis represents the number of points.

[0076] Step 103: Determine the range parameter of the number of points in the ground point cloud based on the Z-axis distribution histogram.

[0077] In this step, a reasonable range of point cloud quantity parameters for dividing the ground point cloud can be determined based on the Z-axis distribution histogram and the geometric distribution characteristics of the radar point cloud data. The point cloud quantity range parameter is a parameter that represents the quantity range of the ground point cloud. For example, it is determined that the quantity of the ground point cloud should be between a and b, where a and b are two quantity values.

[0078] Specifically, the geometric distribution characteristics of radar point cloud data are as follows: since radar point cloud data is obtained by scanning the road surface, the number of ground point clouds in the radar point cloud data is usually relatively large. Therefore, by analyzing the Z-axis distribution histogram, the Z-axis intervals corresponding to a large number of point clouds usually exhibit intervals with relatively high peak values, such as the second highest peak value or the third highest peak value. This Z-axis interval is roughly the Z-axis coordinate interval of the ground point cloud, and the point cloud quantity range parameter used to divide the ground point cloud can be determined based on the number of point clouds corresponding to this Z-axis interval. Alternatively, the point cloud quantity range parameter used to divide the ground point cloud can be further determined based on the point cloud quantity of another Z-axis interval in the Z-axis distribution histogram that has a similar number of point clouds to the one corresponding to the Z-axis interval. Of course, other reasonable methods can also be used based on the Z-axis distribution histogram to determine the point cloud quantity range parameter used to divide the ground point cloud.

[0079] For example, for radar point cloud data that includes the largest number of point clouds near the radar and the second largest number of ground point clouds, it is possible to base it on... Figure 2 The middle and secondary peaks hist peak2 The number of nearby point clouds determines the range of point cloud numbers used to divide the ground point cloud, such as approximately between 4000 and 1200.

[0080] Optionally, step 103 includes:

[0081] Based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, a third Z-axis interval is determined, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval;

[0082] Based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval, the point cloud quantity range parameter used to divide the ground point cloud is determined.

[0083] Based on the geometric distribution characteristics of lidar point cloud data, there are usually two peak values ​​in the Z-axis distribution histogram. These two peak values ​​represent the point cloud distribution near the radar and the distribution of the ground point cloud, respectively. Therefore, in one embodiment, the point cloud quantity range parameter used to divide the ground point cloud can be reasonably determined based on these two peak values.

[0084] The aforementioned first and second peak values ​​can refer to two peak values ​​in the Z-axis distribution histogram, such as the highest and second highest peak values, respectively, which are the peak values ​​with the most or second most point cloud points. Figure 2 The point cloud Z-axis distribution histogram contains two peak values, denoted as hist and hist. peak1 and hist peak2 , and hist peak1 histpeak2 The first peak value can be hist peak1 The second peak value can be hist peak2 .

[0085] In this embodiment, the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram can be determined separately. For example, the Z-axis coordinate interval where the first peak value is located can be determined as the first Z-axis interval, and the Z-axis coordinate interval where the second peak value is located can be determined as the second Z-axis interval. Figure 2 In the middle, the highest peak hist peak1 The corresponding Z-axis interval is near 0. For example, if the starting point of this Z-axis interval is -1 and the ending point is 1, the second highest peak value is hist. peak2 The corresponding Z-axis interval is around -2400, for example, the starting point of this Z-axis interval is -2450 and the ending point is -2350.

[0086] A third Z-axis interval can be determined based on the first Z-axis interval and the second Z-axis interval. For example, the interval between the first Z-axis interval and the second Z-axis interval can be determined as the third Z-axis interval. Alternatively, this interval can be further fine-tuned.

[0087] In one optional implementation, the first peak value is the highest peak value, and the second peak value is the second highest peak value;

[0088] The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

[0089] In one implementation, the highest and second-highest peak values ​​in the Z-axis distribution histogram can be determined, such as... Figure 2 hist peak1 and hist peak2 Furthermore, a third Z-axis interval can be determined by using the endpoints of the Z-axis intervals of these two peak values ​​as boundaries. The third Z-axis interval may include the Z-axis intervals of these two peak values ​​or may not include the Z-axis intervals of these two peak values, in order to find a suitable low peak value between the highest peak value and the second highest peak value, thereby facilitating the determination of a suitable range of point cloud quantity parameters for dividing the ground point cloud.

[0090] Specifically, a third Z-axis interval, including the first and second Z-axis intervals, can be determined using the outermost endpoints of the first and second Z-axis intervals as boundaries. For example, when the first Z-axis interval corresponding to the highest peak is located to the right of the second Z-axis interval corresponding to the second highest peak, the left endpoint (i.e., the start point) of the second Z-axis interval can be used as the left endpoint (i.e., the start point) of the third Z-axis interval, and the right endpoint (i.e., the end point) of the first Z-axis interval can be used as the right endpoint (i.e., the end point) of the third Z-axis interval. Similarly, when the first Z-axis interval corresponding to the highest peak is located to the left of the second Z-axis interval corresponding to the second highest peak, the left endpoint (i.e., the start point) of the first Z-axis interval can be used as the left endpoint (i.e., the start point) of the third Z-axis interval, and the right endpoint (i.e., the end point) of the second Z-axis interval can be used as the right endpoint (i.e., the end point) of the third Z-axis interval.

[0091] like Figure 2 In the middle, the second highest peak hist peak2 The corresponding Z-axis interval starting point (e.g., -2450) to the highest peak value hist peak1 The corresponding Z-axis interval endpoint (e.g., 50) is used as the third Z-axis interval, i.e., the third Z-axis interval is [-2450, 50].

[0092] In this way, by combining the intervals corresponding to the highest and second-highest peak values, a reasonable third Z-axis interval can be quickly determined.

[0093] Then, the first low peak value in the third Z-axis interval of the Z-axis distribution histogram can be determined, such as the lowest peak value, the second lowest peak value, etc.

[0094] Optionally, the first low peak value can be the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram. For example... Figure 2 As shown, two peak histograms can be found. peak1 and hist peak2 The statistical minimum between them is denoted as hist. min You can then use the histogram. peak1 The corresponding Z-axis interval to hist peak2 Within the corresponding Z-axis interval (e.g., between -2450 and 50), find the lowest peak value with the fewest points; this value is the histogram. min .

[0095] Finally, based on the first low peak value and the second high peak value, the point cloud quantity range parameter for dividing the ground point cloud can be determined. For example, a value that is several times the first low peak value or higher than a certain value of the first low peak value can be used as the starting value of the point cloud quantity range for dividing the ground point cloud, that is, the minimum ground point cloud quantity. The second high peak value or a value that is similar to the second high peak value can be used as the ending value of the point cloud quantity range for dividing the ground point cloud, that is, the maximum ground point cloud quantity.

[0096] Thus, based on this implementation method, the range of point cloud quantity parameters used for dividing ground point clouds can be accurately and reasonably determined, ensuring that the number of ground point cloud strips divided is reasonable.

[0097] Further, determining the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval includes:

[0098] The minimum number of point clouds to divide the ground point cloud is determined by m times the first low peak value, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0099] The second peak value is determined as the maximum number of point clouds used to divide the ground point cloud.

[0100] In one embodiment, the lowest peak value of the third Z-axis interval can be hist. min m times the minimum point cloud quantity threshold of the ground point cloud, with the second peak value as the second highest peak value hist peak2 The maximum number of ground point clouds is defined as the threshold value for the number of ground point clouds in the radar point cloud data. This means that the number of ground point clouds in the radar point cloud data should be no less than the minimum number of point clouds threshold value and no more than the maximum number of point clouds threshold value.

[0101] Where m can be a reasonable multiple value set according to actual needs. The larger m is set, the smaller the Z-axis interval corresponding to the ground point cloud strip to be divided, that is, the stricter the requirements for the division of ground point cloud strips.

[0102] This allows for flexible determination of a reasonable threshold for the number of ground point clouds, ensuring that the ground point cloud strips are accurately and reasonably divided based on this threshold.

[0103] Step 104: Based on the point cloud quantity range parameter, divide the ground point cloud strips from the radar point cloud data.

[0104] In this step, the radar point cloud data can be divided into stripes to obtain multiple point cloud stripes, which facilitates the subsequent encoding and transmission of the radar point cloud data according to the divided point cloud stripes, thereby improving encoding efficiency and flexibility.

[0105] Specifically, based on the point cloud quantity range parameter, the Z-axis coordinate interval corresponding to the point cloud quantity range parameter can be determined from the Z-axis distribution histogram. The point cloud in the radar point cloud data whose Z-axis coordinate falls into the Z-axis coordinate interval is the ground point cloud. Thus, the ground point cloud can be divided from the radar point cloud data to obtain a ground point cloud strip including the ground point cloud.

[0106] In existing technologies, for massive point cloud data, the common approach is to divide the entire point cloud into multiple point cloud strips for transmission to meet bandwidth requirements. However, existing point cloud striping methods cannot achieve good and reasonable striping for sparse radar point cloud data. In this embodiment, by using the Z-axis distribution histogram of radar point cloud data and combining it with the analysis of the geometric characteristics of the radar point cloud data itself, the range parameters for the number of point clouds used to divide the ground point cloud are determined. This ensures that ground point cloud data can be reasonably divided from the radar point cloud data, achieving reasonable and accurate segmentation of the radar point cloud data.

[0107] Optionally, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold;

[0108] Step 104 may include:

[0109] The Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold is determined as the first target Z-axis interval.

[0110] The point cloud data located in the Z-axis region of the first target in the radar point cloud data is divided into ground point cloud strips.

[0111] In one embodiment, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold, so that the minimum point cloud quantity threshold and the maximum point cloud quantity threshold can constitute the point cloud quantity range.

[0112] Specifically, in this embodiment, the Z-axis interval into which the ground point cloud falls can be determined based on the minimum point cloud quantity threshold and the maximum point cloud quantity threshold. In other words, the Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud quantity threshold and less than or equal to the maximum point cloud quantity threshold can be determined. This Z-axis interval is the first target Z-axis interval into which the ground point cloud falls in the radar point cloud data.

[0113] In this way, it can be determined that the point cloud data located in the Z-axis interval of the first target in the radar point cloud data is the ground point cloud, and these ground point cloud data can be divided into ground point cloud strips.

[0114] like Figure 2 As shown, assuming the minimum point cloud quantity threshold is determined to be 4000, and the maximum point cloud quantity threshold is the second highest peak value hist peak2 If the number of points in the radar point cloud is 12000, then the Z-axis range of points with a number greater than or equal to 4000 and less than or equal to 12000 is approximately between -2500 and -2300. Thus, the point cloud data with Z-axis coordinates between -2500 and -2300 in the radar point cloud data can be separately divided into ground point cloud strips.

[0115] In this way, based on the determined minimum and maximum point cloud quantity thresholds, the Z-axis interval where the ground point cloud is located can be accurately and quickly determined, thereby accurately dividing the ground point cloud strips.

[0116] It should be noted that when determining the range of point cloud quantity parameters used to divide the ground point cloud, the second highest peak value (hist) can also be recorded based on the Z-axis distribution histogram of the radar point cloud data obtained in step 102. peak2 The corresponding Z-axis midpoint serves as the reference starting point for the ground position coordinates in the radar point cloud data, i.e., the starting Z-axis coordinate of the ground point cloud, and can be denoted as ground. z ground z It can only be used as a reference value for dividing ground point clouds, and not as a threshold for actual division.

[0117] Optionally, after dividing the radar point cloud data located in the Z-axis interval of the first target into ground point cloud strips, the method further includes:

[0118] The first target Z-axis interval is divided into N sub-intervals, and the ground point cloud strip is divided into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

[0119] In one implementation, after the ground point cloud strips are divided, they can be further divided into multiple ground point cloud sub-strips to further improve the encoding and transmission efficiency of point cloud data.

[0120] Specifically, the ground point cloud strips can be divided into equal or unequal intervals along the Z-axis, and the number of sub-strips can be preset. For example, the number of ground point cloud sub-strips to be divided can be preset directly, or the total number of strips to be divided from the radar point cloud data can be preset. Thus, when dividing the ground point cloud strips, the number of ground point cloud sub-strips to be divided can be determined by subtracting the number of other types of point cloud strips from the preset total number of strips.

[0121] If it is necessary to divide the ground point cloud into N sub-strips, the first target Z-axis interval can be divided into N sub-intervals, for example, into N sub-intervals at equal intervals. Then, the ground point cloud strip is divided into N ground point cloud sub-strips according to the N sub-intervals. That is, the ground point cloud falling into each sub-interval is a ground point cloud sub-strip.

[0122] In this way, ground point cloud strips with a large number of points can be further divided into multiple ground point cloud sub-strips, which helps to meet the bandwidth limitations of massive point cloud data transmission and improve coding efficiency and flexibility.

[0123] Optionally, after step 102, the method further includes:

[0124] Based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, a second target Z-axis interval is determined for dividing the second type of point cloud, wherein the second type of point cloud is a point cloud located near the radar.

[0125] The point cloud data located in the Z-axis region of the second target in the radar point cloud data is divided into second type point cloud strips.

[0126] In one embodiment, the radar point cloud data can also be divided into radar-near point cloud strips based on the Z-axis geometric distribution characteristics of the radar point cloud data. That is, the number of point clouds near the radar is usually the largest. Therefore, the Z-axis distribution histogram will show the highest peak value, which represents the number of point clouds near the radar.

[0127] Specifically, it can be based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, such as... Figure 2 The highest peak hist peak1 The corresponding Z-axis interval, i.e., the interval where the Z-axis coordinate is greater than or equal to -1 and less than or equal to 1 (i.e., [-1, 1]), or the interval where the Z-axis coordinate is greater than or equal to -1 and less than or equal to 0 (i.e., [-1, 0]), is used to determine the Z-axis interval for dividing the second type of point cloud, i.e., the point cloud located near the radar, which is the second target Z-axis interval. For example, the Z-axis interval corresponding to the highest peak can be directly determined as the second target Z-axis interval, or the Z-axis interval corresponding to the highest peak can be widened and then determined as the second target Z-axis interval.

[0128] Then, it can be determined that the point cloud data in the radar point cloud data whose Z-axis coordinates fall within the Z-axis interval of the second target is the radar vicinity point cloud, thereby dividing this part of the radar point cloud data into radar vicinity point cloud strips. Figure 2 As can be clearly observed from the Z-axis distribution histogram shown, this type of data accounts for a large proportion and their Z-axis coordinates are basically the same. Among them, the ground point cloud stripes are type I point cloud stripes, and the point cloud stripes near the radar are type II point cloud stripes.

[0129] In this way, based on the analysis of the Z-axis distribution histogram of the radar point cloud data, the radar point cloud data can be reasonably divided into ground point cloud strips and radar-near point cloud strips.

[0130] Optionally, the method further includes:

[0131] The radar point cloud data, excluding the ground point cloud and the second type of point cloud, is divided into third type of point cloud strips.

[0132] In one embodiment, the radar point cloud data, excluding the ground point cloud and the second type of point cloud, can be further divided into a third type of point cloud strip, which is the remaining point cloud strip after removing the ground point cloud strip and the point cloud strip near the radar.

[0133] In this way, radar point cloud data can be divided into three point cloud strips: the point cloud strip near the radar, the ground point cloud strip, and the remaining point cloud strip, and the point cloud strip can be reasonably divided.

[0134] The following is combined with Figure 3 The following is an example illustrating the point cloud striping method for radar point cloud data in the embodiments of this application:

[0135] The radar point cloud data can first be divided into point cloud strips near the radar, ground point cloud strips, and remaining point cloud strips using the method described above. Then, the ground point cloud strips are further divided into multiple ground point cloud sub-strips. Specifically, assuming the user-defined number of point cloud strips is N′, after removing the radar location point cloud strip and the remaining point cloud strip, the ground point cloud strips are further divided into N′-2 sub-strips at equal intervals along the Z-axis, ultimately generating a total of N′ point cloud strips. Figure 3 As shown, these are the point cloud strips near the radar, the remaining point cloud strips, ground point cloud sub-strip 1, ground point cloud sub-strip 2, ..., ground point cloud sub-strip N′-2, respectively.

[0136] This application proposes a striping method based on an adaptive Z-axis distribution histogram by utilizing the geometric distribution characteristics of LiDAR point cloud data. This method can obtain multiple independently encoded and decoded local point cloud structures (i.e., structures with multiple point cloud strips), supporting parallel processing, enhancing system fault tolerance, and improving coding efficiency. Furthermore, the point cloud striping methods provided in this application result in point cloud strips with different geometric characteristics, and each sub-strip can adopt different quantization and encoding methods according to its characteristics, thereby enabling more flexible processing of point cloud data.

[0137] The point cloud processing method of this application embodiment acquires radar point cloud data; generates a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; determines a range parameter for the number of point clouds used to divide the ground point cloud based on the Z-axis distribution histogram; and divides ground point cloud strips from the radar point cloud data based on the point cloud number range parameter. Thus, by using the Z-axis distribution histogram of the radar point cloud data, combined with the analysis of the geometric characteristics of the radar point cloud data itself, the range parameter for the number of point clouds used to divide the ground point cloud is determined, thereby ensuring the reasonable division of ground point cloud data from the radar point cloud data and improving point cloud encoding efficiency.

[0138] This application also provides a point cloud processing apparatus. See [link to previous document]. Figure 4 , Figure 4 This is a structural diagram of the point cloud processing apparatus provided in the embodiments of this application. Since the principle of the point cloud processing apparatus in solving the problem is similar to that of the point cloud processing method in the embodiments of this application, the implementation of this point cloud processing apparatus can refer to the implementation of the method, and the repeated parts will not be described again.

[0139] like Figure 4 As shown, the point cloud processing device 400 includes:

[0140] Acquisition module 401 is used to acquire radar point cloud data;

[0141] The generation module 402 is used to generate a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data of the radar point cloud data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds;

[0142] The first determining module 403 is used to determine the range parameter of the number of points in the ground point cloud based on the Z-axis distribution histogram.

[0143] The first segmentation module 404 is used to segment ground point cloud stripes from the radar point cloud data according to the point cloud quantity range parameter.

[0144] Optionally, the first determining module 403 includes:

[0145] The first determining submodule is used to determine a third Z-axis interval based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval;

[0146] The second determining submodule is used to determine the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0147] Optionally, the first peak value is the highest peak value, and the second peak value is the second highest peak value;

[0148] The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

[0149] Optionally, the first low peak value is the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram.

[0150] Optionally, the second determining submodule includes:

[0151] The first determining unit is used to determine m times the first low peak value as the minimum point cloud quantity threshold for dividing the ground point cloud, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0152] The second determining unit is used to determine the second peak value as the maximum point cloud quantity threshold for dividing the ground point cloud.

[0153] Optionally, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold;

[0154] The first partitioning module 404 includes:

[0155] The third determining submodule is used to determine the Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold as the first target Z-axis interval.

[0156] The first segmentation submodule is used to divide the radar point cloud data located in the Z-axis interval of the first target into ground point cloud strips.

[0157] Optionally, the first partitioning module 404 further includes:

[0158] The second partitioning submodule is used to divide the first target Z-axis interval into N sub-intervals, and to divide the ground point cloud strip into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

[0159] Optionally, the point cloud processing device 400 also includes:

[0160] The second determining module is used to determine the second target Z-axis interval for dividing the second type of point cloud based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, wherein the second type of point cloud is a point cloud located near the radar.

[0161] The second division module is used to divide the point cloud data located in the Z-axis interval of the second target in the radar point cloud data into a second type of point cloud strip.

[0162] Optionally, the point cloud processing device 400 also includes:

[0163] The third segmentation module is used to segment the radar point cloud data, excluding the ground point cloud and the second type of point cloud, into third type of point cloud strips.

[0164] The point cloud processing device 400 provided in this application embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0165] The point cloud processing apparatus 400 of this application embodiment acquires radar point cloud data; generates a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; determines a range parameter for the number of point clouds used to divide the ground point cloud based on the Z-axis distribution histogram; and divides ground point cloud strips from the radar point cloud data based on the point cloud number range parameter. Thus, by combining the Z-axis distribution histogram of the radar point cloud data with the analysis of the geometric characteristics of the radar point cloud data itself, the range parameter for the number of point clouds used to divide the ground point cloud is determined, thereby ensuring that ground point cloud data is reasonably divided from the radar point cloud data and improving point cloud coding efficiency.

[0166] This application also provides an electronic device. Since the principle by which the electronic device solves the problem is similar to the point cloud processing method in this application, the implementation of this electronic device can be found in the implementation of the method, and repeated details will not be described again. Figure 5 As shown, the electronic device according to an embodiment of this application includes:

[0167] Processor 500 is used to read the program from memory 520 and execute the following procedures:

[0168] Acquire radar point cloud data;

[0169] Based on the Z-axis coordinate data of the radar point cloud data, a Z-axis distribution histogram of the radar point cloud data is generated, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds;

[0170] Based on the Z-axis distribution histogram, determine the range of point cloud quantity parameters used to divide the ground point cloud;

[0171] Based on the point cloud quantity range parameter, ground point cloud strips are divided from the radar point cloud data.

[0172] Among them, Figure 5 In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (processor 500) and memory (memory 520). The bus architecture can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. A bus interface provides the interface. Processor 500 is responsible for managing the bus architecture and general processing, and memory 520 can store data used by processor 500 during operation.

[0173] Optionally, the processor 500 is also used to read the program from the memory 520 and perform the following steps:

[0174] Based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, a third Z-axis interval is determined, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval;

[0175] Based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval, the point cloud quantity range parameter used to divide the ground point cloud is determined.

[0176] Optionally, the first peak value is the highest peak value, and the second peak value is the second highest peak value;

[0177] The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

[0178] Optionally, the first low peak value is the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram.

[0179] Optionally, the processor 500 is also used to read the program from the memory 520 and perform the following steps:

[0180] The minimum number of point clouds to divide the ground point cloud is determined by m times the first low peak value, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval.

[0181] The second peak value is determined as the maximum number of point clouds used to divide the ground point cloud.

[0182] Optionally, the point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold;

[0183] Processor 500 is also used to read programs from memory 520 and perform the following steps:

[0184] The Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold is determined as the first target Z-axis interval.

[0185] The point cloud data located in the Z-axis region of the first target in the radar point cloud data is divided into ground point cloud strips.

[0186] Optionally, the processor 500 is also used to read the program from the memory 520 and perform the following steps:

[0187] The first target Z-axis interval is divided into N sub-intervals, and the ground point cloud strip is divided into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

[0188] Optionally, the processor 500 is also used to read the program from the memory 520 and perform the following steps:

[0189] Based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, a second target Z-axis interval is determined for dividing the second type of point cloud, wherein the second type of point cloud is a point cloud located near the radar.

[0190] The point cloud data located in the Z-axis region of the second target in the radar point cloud data is divided into second type point cloud strips.

[0191] Optionally, the processor 500 is also used to read the program from the memory 520 and perform the following steps:

[0192] The radar point cloud data, excluding the ground point cloud and the second type of point cloud, is divided into third type of point cloud strips.

[0193] The electronic device provided in this application embodiment can execute the above method embodiment, and its implementation principle and technical effect are similar, so it will not be described again here.

[0194] Furthermore, the computer-readable storage medium of this application embodiment is used to store a computer program, which can be executed by a processor. Figure 1 Each step in the method embodiment shown.

[0195] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0196] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can be physically included separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0197] The integrated units implemented as software functional units described above can be stored in a computer-readable storage medium. These software functional units, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute some steps of the transmission and reception methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0198] The above description is the preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principles described in this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A point cloud processing method, characterized in that, include: Acquire radar point cloud data; Based on the Z-axis coordinate data of the radar point cloud data, a Z-axis distribution histogram of the radar point cloud data is generated, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; Based on the Z-axis distribution histogram, determine the range of point cloud quantity parameters used to divide the ground point cloud; Based on the point cloud quantity range parameter, ground point cloud strips are divided from the radar point cloud data; The step of determining the point cloud quantity range parameter for dividing the ground point cloud based on the Z-axis distribution histogram includes: determining a third Z-axis interval based on the first Z-axis interval corresponding to the first peak value and the second Z-axis interval corresponding to the second peak value in the Z-axis distribution histogram, wherein the second peak value is smaller than the first peak value, and the Z-axis interval is a Z-axis coordinate interval; and determining the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second peak value in the third Z-axis interval of the Z-axis distribution histogram. The step of determining the point cloud quantity range parameter for dividing the ground point cloud based on the first low peak value and the second high peak value of the Z-axis distribution histogram in the third Z-axis interval includes: determining m times the first low peak value as the minimum point cloud quantity threshold for dividing the ground point cloud, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval; and determining the second high peak value as the maximum point cloud quantity threshold for dividing the ground point cloud.

2. The method according to claim 1, characterized in that, The first peak value is the highest peak value, and the second peak value is the second highest peak value; The two endpoints of the third Z-axis interval are the endpoints of the first Z-axis interval and the endpoints of the second Z-axis interval, respectively.

3. The method according to claim 1, characterized in that, The first low peak value is the minimum number of point clouds in the third Z-axis interval of the Z-axis distribution histogram.

4. The method according to any one of claims 1 to 3, characterized in that, The point cloud quantity range parameter includes a minimum point cloud quantity threshold and a maximum point cloud quantity threshold; The step of dividing ground point cloud stripes from the radar point cloud data according to the point cloud quantity range parameter includes: The Z-axis interval in the Z-axis distribution histogram where the number of point clouds is greater than or equal to the minimum point cloud number threshold and less than or equal to the maximum point cloud number threshold is determined as the first target Z-axis interval. The point cloud data located in the Z-axis region of the first target in the radar point cloud data is divided into ground point cloud strips.

5. The method according to claim 4, characterized in that, After dividing the radar point cloud data located in the Z-axis interval of the first target into ground point cloud strips, the method further includes: The first target Z-axis interval is divided into N sub-intervals, and the ground point cloud strip is divided into N ground point cloud sub-strips according to the N sub-intervals, wherein one ground point cloud sub-strip corresponds to one sub-interval, and N is an integer greater than 1.

6. The method according to claim 1, characterized in that, After generating the Z-axis distribution histogram of the radar point cloud data, the method further includes: Based on the Z-axis interval corresponding to the highest peak value in the Z-axis distribution histogram, a second target Z-axis interval is determined for dividing the second type of point cloud, wherein the second type of point cloud is a point cloud located near the radar. The point cloud data located in the Z-axis region of the second target in the radar point cloud data is divided into second type point cloud strips.

7. The method according to claim 6, characterized in that, The method further includes: The radar point cloud data, excluding the ground point cloud and the second type of point cloud, is divided into third type of point cloud strips.

8. A point cloud processing device, characterized in that, include: The acquisition module is used to acquire radar point cloud data; The generation module is used to generate a Z-axis distribution histogram of the radar point cloud data based on the Z-axis coordinate data of the radar point cloud data, wherein the Z-axis distribution histogram represents the relationship between the Z-axis coordinate interval and the number of point clouds; The first determining module is used to determine the range parameter of the number of points in the ground point cloud for dividing the ground point cloud based on the Z-axis distribution histogram. The first segmentation module is used to segment ground point cloud stripes from the radar point cloud data according to the point cloud quantity range parameter. The first determining module includes: a first determining submodule, configured to determine a third Z-axis interval based on a first Z-axis interval corresponding to a first peak value and a second Z-axis interval corresponding to a second peak value in the Z-axis distribution histogram, wherein the second peak value is less than the first peak value, and the Z-axis interval is a Z-axis coordinate interval; and a second determining submodule, configured to determine a point cloud quantity range parameter for dividing the ground point cloud based on a first low peak value and a second peak value in the third Z-axis interval of the Z-axis distribution histogram. The second determining submodule includes: a first determining unit, used to determine m times the first low peak value as the minimum point cloud quantity threshold for dividing the ground point cloud, where m is a preset value, and the first low peak value is the lowest peak value of the Z-axis distribution histogram in the third Z-axis interval; and a second determining unit, used to determine the second high peak value as the maximum point cloud quantity threshold for dividing the ground point cloud.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor; The processor is used to read the program in the memory to implement the steps in the point cloud processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, which, when executed by a processor, implements the steps of the point cloud processing method as described in any one of claims 1 to 7.

Images