Point cloud denoising method and device, detection radar and storage medium

Abstract

The application relates to an electric cloud denoising method and device, a detection radar and a storage medium. The method comprises the following steps: acquiring a distance value between a target point cloud point and a detection radar; based on the distance value, a dynamic search area is acquired; wherein the dynamic search area changes with the distance value; according to the distance value, a corresponding quantity threshold value is acquired; wherein the distance value and the quantity threshold value have a preset corresponding relationship; the number of point cloud points in the dynamic search area is compared with the corresponding quantity threshold value; if the number of point cloud points in the dynamic search area is not greater than the corresponding quantity threshold value, the target point cloud point is removed as a noise point. According to the method, different dynamic search radii can be determined according to different distance values, the dynamic search radius is dynamically adjusted, the actual distribution of point cloud points at different positions is adapted, the point cloud points far away from the detection radar are prevented from being mistaken as noise points and removed, and the denoising accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic information, in particular to a point cloud denoising method and device, a detection radar and a storage medium. BACKGROUND

[0002] Laser radar is a radar system that detects the position, speed and other characteristic quantities of a target by emitting a laser beam. It is increasingly used in target positioning, obstacle detection and classification. Therefore, high-quality laser point cloud data is particularly important.

[0003] However, adverse weather such as rain and snow can cause a large amount of noise in the detection process of the laser radar, especially in the field of autonomous driving. In order to ensure the accuracy of environmental detection, it is necessary to denoise the point cloud points obtained by detection. The traditional point cloud denoising method is to form a fixed-size filtering region centered on a certain point cloud point, and then determine whether the point cloud point is a noise point according to the distribution of the point cloud points in the filtering region, and then decide whether to remove it.

[0004] However, due to the resolution of the laser radar itself, the distribution of point cloud points far from the laser radar is relatively sparse. When the above traditional method is used for denoising, the sparse point cloud points far away will be misjudged as noise points and removed, thereby reducing the accuracy of laser radar detection. SUMMARY

[0005] Therefore, it is necessary to provide a point cloud denoising method, device, detection radar and storage medium to solve the above technical problems.

[0006] A point cloud denoising method comprises:

[0007] obtaining a distance value between a target point cloud point and a detection radar;

[0008] based on the distance value, obtaining a dynamic search region; wherein the dynamic search region changes with the distance value;

[0009] obtaining a corresponding number threshold according to the distance value; wherein the distance value and the number threshold have a preset corresponding relationship;

[0010] comparing the number of point cloud points in the dynamic search region with the corresponding number threshold;

[0011] if the number of point cloud points in the dynamic search region is not greater than the corresponding number threshold, the target point cloud point is removed as a noise point.

[0012] In one embodiment, based on the distance value, the dynamic search region is obtained, comprising:

[0013] obtaining the horizontal angle resolution of the detection radar and the distance variation constant of the detection radar in the vertical direction;

[0014] The search radius is obtained according to the distance value, the horizontal angle resolution, and the distance change constant;

[0015] The dynamic search region is determined according to the target point cloud point and the search radius.

[0016] In one of the embodiments, the search radius is obtained according to the distance value, the horizontal angle resolution, and the distance change constant, comprising:

[0017] The product of the distance value, the horizontal angle resolution, and the distance change constant is calculated to obtain an initial search radius;

[0018] It is judged whether the initial search radius is less than a radius threshold value;

[0019] If yes, the radius threshold value is determined as the search radius;

[0020] If no, the initial search radius is determined as the search radius.

[0021] In one of the embodiments, the dynamic search region is determined according to the target point cloud point and the search radius, comprising:

[0022] The target point cloud point is taken as the center and the search radius is taken as the radius to form the dynamic search region.

[0023] In one of the embodiments, the corresponding quantity threshold value is obtained according to the distance value, comprising:

[0024] The distance ratio between the distance value and the maximum measurement distance of the detection radar is calculated;

[0025] The quantity threshold value corresponding to the distance ratio is determined according to the corresponding relationship between the preset quantity threshold value and the distance ratio range.

[0026] In one of the embodiments, the quantity threshold value corresponding to the distance ratio is determined according to the corresponding relationship between the preset quantity threshold value and the distance ratio range, comprising:

[0027] The distance ratio is matched with the distance ratio range to determine the target distance ratio range to which the distance ratio belongs;

[0028] According to the corresponding relationship between the preset quantity threshold value and the distance ratio range, the quantity threshold value corresponding to the target distance ratio range is determined as the quantity threshold value corresponding to the distance ratio.

[0029] In one of the embodiments, the above method further comprises:

[0030] The Euclidean distance between the other point cloud point and the target point cloud point is obtained;

[0031] If the Euclidean distance is less than or equal to the search radius, the other point cloud points are regarded as point cloud points in the dynamic search area.

[0032] The point cloud points in the dynamic search area are counted to obtain the number of point cloud points in the dynamic search area.

[0033] A point cloud denoising device comprises:

[0034] A distance obtaining module is configured to obtain a distance value between a target point cloud point and a detection radar.

[0035] An area calculating module is configured to calculate a dynamic search area based on the distance value, wherein the dynamic search area changes with the distance value.

[0036] A number obtaining module is configured to obtain a corresponding number threshold based on the distance value, wherein the distance value and the number threshold have a preset corresponding relationship.

[0037] A threshold comparing module is configured to compare the number of point cloud points in the dynamic search area with the corresponding number threshold.

[0038] A noise removing module is configured to remove the target point cloud point as a noise point in the case that the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold.

[0039] A detection radar comprises a signal transmitter, a signal receiver, a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0040] A distance value between a target point cloud point and the detection radar is obtained.

[0041] A dynamic search area is obtained based on the distance value, wherein the dynamic search area changes with the distance value.

[0042] A corresponding number threshold is obtained based on the distance value, wherein the distance value and the number threshold have a preset corresponding relationship.

[0043] The number of point cloud points in the dynamic search area is compared with the corresponding number threshold.

[0044] If the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold, the target point cloud point is removed as a noise point.

[0045] A computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0046] A distance value between a target point cloud point and a detection radar is obtained.

[0047] Based on the distance value, a dynamic search area is obtained; wherein the dynamic search area changes with the distance value;

[0048] According to the distance value, a corresponding quantity threshold is obtained; wherein the distance value and the quantity threshold have a preset corresponding relationship;

[0049] The number of point cloud points in the dynamic search area is compared with the corresponding quantity threshold;

[0050] If the number of point cloud points in the dynamic search area is not greater than the corresponding quantity threshold, the target point cloud point is removed as a noise point.

[0051] The above point cloud denoising method, device, detection radar and storage medium, by obtaining the distance value between the target point cloud point and the detection radar, based on the distance value, obtaining the dynamic search area which changes with the distance value, and according to the preset corresponding relationship between the distance value and the quantity threshold, obtaining the quantity threshold of the corresponding point cloud point, and then comparing the number of point cloud points in the dynamic search area with the corresponding quantity threshold, in the case that the number of point cloud points in the dynamic search area is not greater than the corresponding quantity threshold, the target point cloud point is removed as a noise point. According to different distance values, different dynamic search radii are determined, the dynamic search radius is dynamically adjusted, and the actual distribution of point cloud points at different positions is adapted, so as to avoid the point cloud points far away from the detection radar being mistaken for noise points and removed, thereby improving the accuracy of denoising. Especially when applied in the field of automatic driving, the accuracy of environmental detection can be improved, and the safety hidden danger can be reduced. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 It is the application environment diagram of the point cloud denoising method in one embodiment;

[0053] Figure 2 It is the flowchart of the point cloud denoising method in one embodiment;

[0054] Figure 3 It is the flowchart of determining the dynamic search area in one embodiment;

[0055] Figure 4 It is the flowchart of determining the search radius in one embodiment;

[0056] Figure 5 It is the flowchart of determining the quantity threshold in one embodiment;

[0057] Figure 6 It is the flowchart of determining the quantity threshold in another embodiment;

[0058] Figure 7 It is the flowchart of obtaining the number of point cloud points in the dynamic search area in one embodiment;

[0059] Figure 8 This is a structural block diagram of a point cloud denoising device in one embodiment;

[0060] Figure 9 This is a diagram of the internal structure of a detection radar in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] The point cloud denoising method provided in this application can be applied to, for example... Figure 1 In the application environment shown, the detection radar 100 emits electromagnetic signals. These signals are reflected back to the radar 100 after being reflected by an object. The locations where the electromagnetic signals strike the object form point cloud points. By emitting multiple electromagnetic signals or periodically rotating to emit electromagnetic signals, a large number of point cloud points can be obtained. The detection radar 100 acquires the distance values ​​between the target point cloud points and the radar 100 from this large number of point cloud points, and calculates a dynamic search radius based on these distance values. The dynamic search area changes with the distance values. The detection radar 100 obtains the quantity threshold corresponding to the distance value according to a preset correspondence between the distance value and a quantity threshold, and compares the number of point cloud points in the dynamic search area with the corresponding quantity threshold. If the number of point cloud points in the dynamic search area is not greater than the corresponding quantity threshold, the target point cloud point is removed as noise. The above method can be implemented directly by the detection radar 100, or by a terminal communicating with the detection radar 100. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices.

[0063] In one embodiment, such as Figure 2 As shown, a point cloud denoising method is provided, which is applied to... Figure 1 Taking the detection radar in the middle as an example, the following steps are included:

[0064] S210, Obtain the distance between the target point cloud and the detection radar.

[0065] Optionally, the detection radar can be a radar used for ranging, such as lidar, millimeter-wave radar, or microwave radar. In this embodiment, the type of detection radar is not specifically limited.

[0066] Optionally, the detection radar can form a large number of point cloud points on the object by emitting a multi-thread electromagnetic signal or periodically rotating the emitted electromagnetic signal, and obtain the coordinate position of each point cloud point accordingly. The manner of obtaining the coordinate position of each point cloud point is the same as that of the prior art, and will not be described here. The detection radar can obtain the distance value r of the target point cloud point from itself according to the coordinate position of the point cloud point p .

[0067] wherein the target point cloud point is any point cloud point in the large number of point cloud points, and the coordinate position of the target point cloud point P is (x i , y i , z i ),

[0068] S220, obtaining a dynamic search area according to the distance value.

[0069] wherein the dynamic search area is a spatial area determined according to the target point cloud point, and the dynamic search area changes with the distance value.

[0070] Optionally, the area of the dynamic search area can change with the distance value, or the shape of the dynamic search area can change with the distance value, or both the area and the shape of the dynamic search area can change with the distance value. In this embodiment, the manner in which the dynamic search area changes with the distance value is not specifically limited.

[0071] Optionally, in the case where the area of the dynamic search area changes with the distance value, the detection radar can determine the area of the dynamic search area corresponding to the obtained distance value according to a preset corresponding relationship between the distance value and the area, and form the dynamic search area of the area according to the target point cloud point. For example, the dynamic search area of the area including the target point cloud point can be formed, or the dynamic search area of the area can be formed with the target point cloud point as the center point.

[0072] S230, obtaining a corresponding quantity threshold according to the distance value.

[0073] wherein the quantity threshold is a point cloud point quantity threshold, and the distance value and the quantity threshold have a preset corresponding relationship. Optionally, the preset corresponding relationship includes a direct corresponding relationship between the distance threshold and the point cloud point quantity threshold, or includes a corresponding relationship between the dynamic search area corresponding to the distance threshold and the point cloud point quantity threshold.

[0074] Optionally, after the detection radar determines the dynamic search area according to the obtained distance threshold, the detection radar can determine the corresponding point cloud point quantity threshold according to the corresponding relationship between the dynamic search area and the point cloud point quantity threshold. The point cloud point data threshold here is the maximum number of point cloud points that can be accommodated in the dynamic search area.

[0075] S240, compare the number of point cloud points in the dynamic search area with the corresponding number threshold.

[0076] S250, if the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold, remove the target point cloud point as a noise point.

[0077] Specifically, the detection radar obtains the number of point cloud points in the dynamic search area, compares the number of point cloud points with the corresponding number threshold, and determines whether the target point cloud point is removed as a noise point according to the comparison result. If the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold, the target point cloud point is removed as a noise point. Otherwise, if the number of point cloud points in the dynamic search area is greater than the corresponding number threshold, it is determined that the target point cloud point is not a noise point, and the target point cloud point is retained.

[0078] In the embodiment, the detection radar obtains a distance value between the target point cloud point and the detection radar, obtains a dynamic search area that changes with the distance value based on the distance value, and obtains a corresponding number threshold of point cloud points according to a preset corresponding relationship between the distance value and the number threshold. Then, the number of point cloud points in the dynamic search area is compared with the corresponding number threshold, and the target point cloud point is removed as a noise point when the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold. In this way, different dynamic search radii are determined according to different distance values, and the dynamic search radius is dynamically adjusted to adapt to the actual distribution of point cloud points at different positions, so that point cloud points far from the detection radar are not mistakenly removed as noise points, thereby improving the accuracy of noise removal. In particular, when applied to the field of autonomous driving, the accuracy of environmental detection can be improved, and safety hazards can be reduced.

[0079] In one embodiment, to improve the applicability of the dynamic search area to different detection radars, as shown in Figure 3 S220 includes:

[0080] S310, obtain the horizontal angle resolution of the detection radar and the distance variation constant of the detection radar in the vertical direction.

[0081] S320, obtain the search radius according to the distance value, the horizontal angle resolution, and the distance variation constant.

[0082] Specifically, the horizontal angle resolution and the distance variation constant in the vertical direction of different detection radars are different. When determining the dynamic search area, the detection radar can obtain its own horizontal angle resolution a and distance variation constant b in the vertical direction, and obtain the search radius according to the distance value r of the target point cloud and the detection radar p, the horizontal angle resolution a and the distance variation constant to obtain a search radius DR for determining the dynamic search area.

[0083] S330, determining the dynamic search area according to the target point cloud point and the search radius.

[0084] After the search radius DR is determined, the detection radar further determines the dynamic search area according to the target point cloud point and the search radius DR. Optionally, the detection radar forms the dynamic search area corresponding to the target point cloud with the target point cloud point as the center and the search radius DR as the radius. In this embodiment, the dynamic search area formed is a sphere with the target point cloud point as the center and the search radius DR as the radius. As shown in Figure 1 , the target point cloud P1 corresponds to the dynamic search area A formed with P1 as the center and search radius DR1 as the radius, and the target point cloud P2 corresponds to the dynamic search area B formed with P2 as the center and search radius DR2 as the radius.

[0085] In an alternative embodiment, as shown in Figure 4 , the above S320 comprises:

[0086] S410, calculating the product of the distance value, the horizontal angle resolution and the distance variation constant to obtain an initial search radius.

[0087] Specifically, the detection radar obtains the initial search radius DR0 by calculating the product of the distance value r p , the horizontal angle resolution a and the distance variation constant β, that is, DR0 = r p * a * β.

[0088] S420, judging whether the initial search radius is smaller than a radius threshold.

[0089] S430, if yes, determining the radius threshold as the search radius.

[0090] S440, if no, determining the initial search radius as the search radius.

[0091] The radius threshold r is the minimum set value of the search radius DR. To further improve the accuracy of radar detection, the detection radar compares the size of the initial search radius DR0 and the radius threshold r, if DR0 < r, the radius threshold r is determined as the search radius DR to avoid inaccurate detection caused by too small search radius DR, and if DR0 ≥ r, the initial search radius DR0 is determined as the search radius DR.

[0092] In this embodiment, the detection radar obtains the horizontal angle resolution and the distance variation constant in the vertical direction of itself, and calculates the search radius of the dynamic search area according to the distance value of the target point cloud point, the horizontal angle resolution and the distance variation constant in the vertical direction of itself, and then determines the dynamic search area according to the target point cloud point and the search radius. Specifically, the dynamic search area is formed with the target point cloud point as the center and the search radius as the radius. Since the search radius of the dynamic search area is obtained according to the horizontal angle resolution and the distance variation constant in the vertical direction of the detection radar itself, the determined dynamic search area is suitable for the corresponding detection radar, thereby improving the applicability of the dynamic search area to different detection radars.

[0093] In one embodiment, the preset corresponding relationship between the distance value and the quantity threshold value includes a corresponding relationship between a ratio of the distance value and a maximum measurement distance of the detection radar and the quantity threshold value of the point cloud point, as shown in Figure 5 The S230 includes the following steps, as shown in

[0094] S510, calculating a distance ratio between the distance value and the maximum measurement distance of the detection radar.

[0095] Specifically, the detection radar obtains a distance ratio between the distance value r p and the maximum measurement distance D of the detection radar. For example, as shown in Figure 1 , the ratio of the distance value r1 of the target point cloud P1 and D of the detection radar is r1 / D, and the ratio of the distance value r2 of the target point cloud P2 and D of the detection radar is r2 / D.

[0096] S520, determining the quantity threshold value corresponding to the distance ratio according to the preset corresponding relationship between the quantity threshold value and the distance ratio range.

[0097] The preset corresponding relationship between the quantity threshold value and the distance ratio range includes that the quantity threshold value k1 corresponds to the distance ratio range T1, the quantity threshold value k2 corresponds to the distance ratio range T2, and the quantity threshold value kn corresponds to the distance ratio range Tn. For example, the quantity threshold value 7 corresponds to the distance ratio range (0, 1 / 3], the quantity threshold value 5 corresponds to the distance ratio range (1 / 3, 2 / 3], and the quantity threshold value 3 corresponds to the distance ratio range (2 / 3, 1].

[0098] In an optional embodiment, the S520 specifically includes the following steps, as shown in Figure 6

[0099] S610, matching the distance ratio with the distance ratio range to determine the target distance ratio range to which the distance ratio belongs.

[0100] ​S620, determining the quantity threshold corresponding to the target distance ratio range as the quantity threshold corresponding to the distance ratio according to the correspondence between the preset quantity threshold and distance ratio range and the target distance ratio range.

[0101] Specifically, the detection radar matches the distance ratio obtained by the detection radar with the distance ratio range in the correspondence between the preset quantity threshold and distance ratio range, and determines the target distance ratio range to which the distance ratio belongs. For example, please continue to refer to Figure 1 , if the ratio of the distance value r1 of the target point cloud P1 and D of the detection radar is r1 / D=5 / 6, the detection radar matches the distance ratio 5 / 6 with the above distance ratio range (0, 1 / 3], (1 / 3, 2 / 3] and (2 / 3, 1], and determines that the target distance ratio range to which the distance ratio 5 / 6 belongs is (2 / 3, 1], and further determines the quantity threshold 3 corresponding to the target distance ratio range (2 / 3, 1] as the quantity threshold corresponding to the distance ratio 5 / 6 according to the correspondence between the preset quantity threshold and distance ratio range.

[0102] In the embodiment, the detection radar calculates the distance ratio between the distance value and the maximum measurement distance of the detection radar, so as to determine the quantity threshold corresponding to the distance ratio according to the correspondence between the preset quantity threshold and distance ratio range, so as to be suitable for detection radars with different ranging capabilities, and further improve the applicability of the above point cloud denoising method to different detection radars.

[0103] In one embodiment, in order to simplify the process of obtaining the number of point cloud points in the dynamic search area, as Figure 7 shown, the above method further comprises:

[0104] S710, obtaining the Euclidean distance between the other point cloud points and the target point cloud point.

[0105] Among them, the other point cloud points are point cloud points in the large number of point cloud points obtained by the detection radar, except the target point cloud point. The detection radar selects any point cloud point in the large number of point cloud points obtained as the target point cloud point, and calculates the Euclidean distance between the other point cloud points and the target point cloud point according to the coordinates of the other point cloud points and the coordinates of the target point cloud point.

[0106] S720, if the Euclidean distance is less than or equal to the search radius, the other point cloud points are regarded as point cloud points in the dynamic search area.

[0107] S730, counting the point cloud points in the dynamic search area to obtain the number of point cloud points in the dynamic search area.

[0108] Specifically, the detection radar compares the Euclidean distance between the target point cloud point obtained by calculation and other point cloud points with the size of the search radius, and determines whether the target point cloud point is a point cloud point in the dynamic search area according to the comparison result. Wherein, if the Euclidean distance is less than or equal to the search radius, the other point cloud points are regarded as the point cloud points in the dynamic search area, and then the point cloud points in the dynamic search area are counted to obtain the number of point cloud points in the dynamic search area. As shown in Figure 1 The Euclidean distances between the other point cloud points p1-p5 relative to the target point cloud point P1 and the target point cloud point P1 are less than the search radius DR1, so the other point cloud points p1-p5 are regarded as the point cloud points in the dynamic search area A, and the number of point cloud points in the dynamic search area A is counted to obtain 5.

[0109] In this embodiment, the detection radar obtains the Euclidean distance between the other point cloud points and the target point cloud point, and if the Euclidean distance is less than or equal to the search radius, the other point cloud points are regarded as the point cloud points in the dynamic search area, and then the point cloud points in the dynamic search area are counted to obtain the number of point cloud points in the dynamic search area. The way of calculating the number of point cloud points in the dynamic search area by calculating the Euclidean distance between the target point cloud point and the other point cloud points involves simple calculation, which further simplifies the process of obtaining the number of point cloud points in the dynamic search area and improves the overall denoising efficiency.

[0110] It should be understood that, although Figures 2-7 Each step in the flowchart of Figures 2-7 At least part of the steps in

[0111] In one embodiment, as shown in Figure 8 A point cloud denoising device is provided, comprising: a distance obtaining module 801, a region calculating module 802, a number obtaining module 803, a threshold comparing module 804, and a noise removing module 805, wherein:

[0112] The distance obtaining module 801 is configured to obtain a distance value between a target point cloud point and a detection radar;

[0113] The region calculating module 802 is configured to calculate a dynamic search area based on the distance value; wherein the dynamic search area changes with the distance value;

[0114] The quantity obtaining module 803 is configured to obtain a corresponding quantity threshold according to the distance value, wherein the distance value and the quantity threshold have a preset corresponding relationship.

[0115] The threshold comparison module 804 is configured to compare the quantity of the point cloud points in the dynamic search area with the corresponding quantity threshold.

[0116] The noise removing module 805 is configured to remove the target point cloud point as a noise point in a case where the quantity of the point cloud points in the dynamic search area is not greater than the corresponding quantity threshold.

[0117] In one of the embodiments, the area calculation module 802 is specifically configured to:

[0118] obtain a horizontal angle resolution of the detection radar and a distance variation constant of the detection radar in a vertical direction, obtain a search radius according to the distance value, the horizontal angle resolution and the distance variation constant, and determine the dynamic search area according to the target point cloud point and the search radius.

[0119] In one of the embodiments, the area calculation module 802 is specifically configured to:

[0120] calculate a product of the distance value, the horizontal angle resolution and the distance variation constant to obtain an initial search radius, determine whether the initial search radius is less than a radius threshold, determine the radius threshold as the search radius if yes, and determine the initial search radius as the search radius if no.

[0121] In one of the embodiments, the area calculation module 802 is specifically configured to:

[0122] form the dynamic search area with the target point cloud point as the center and the search radius as the radius.

[0123] In one of the embodiments, the quantity obtaining module 803 is specifically configured to:

[0124] calculate a distance ratio between the distance value and a maximum measurement distance of the detection radar, and determine the quantity threshold corresponding to the distance ratio according to a preset corresponding relationship between the quantity threshold and a distance ratio range.

[0125] In one of the embodiments, the quantity obtaining module 803 is specifically configured to:

[0126] match the distance ratio with the distance ratio range to determine a target distance ratio range to which the distance ratio belongs, and determine the quantity threshold corresponding to the target distance ratio range as the quantity threshold corresponding to the distance ratio according to a preset corresponding relationship between the quantity threshold and the distance ratio range.

[0127] In one of the embodiments, the apparatus further includes a quantity determination module specifically configured to:

[0128] obtaining the Euclidean distance between the other point cloud point and the target point cloud point; if the Euclidean distance is less than or equal to the search radius, the other point cloud point is regarded as a point cloud point in the dynamic search region; and counting the point cloud points in the dynamic search region to obtain the number of point cloud points in the dynamic search region.

[0129] The specific limitations of the point cloud denoising device can refer to the limitations of the point cloud denoising method in the foregoing, and will not be described here. Each module in the above point cloud denoising device can be realized by software, hardware, and a combination thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the detection radar in hardware form, or can be stored in the memory in the detection radar in software form, so as to be called and executed by the processor to perform the operations corresponding to each of the above modules.

[0130] In one embodiment, a detection radar is provided, and an internal structure diagram of the detection radar can be as shown in Figure 9 The detection radar includes a signal transmitter, a signal receiver, a processor, a memory, and a network interface connected through a system bus. The signal transmitter of the detection radar is configured to transmit electromagnetic signals, and the signal receiver is configured to receive electromagnetic signals. The processor of the detection radar is configured to provide computing and control capabilities. The memory of the detection radar includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the detection radar is configured to store coordinate position data of point cloud points. The network interface of the detection radar is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a point cloud denoising method.

[0131] Those skilled in the art can understand that Figure 9 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the detection radar to which the scheme of the present application is applied. The specific detection radar can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0132] In one embodiment, a detection radar is provided, and the detection radar includes a signal transmitter, a signal receiver, and a memory and a processor, and the memory stores a computer program. The processor executes the computer program to implement the following steps:

[0133] Obtaining a distance value between the target point cloud point and the detection radar; obtaining a dynamic search region based on the distance value; wherein the dynamic search region changes with the distance value; obtaining a corresponding quantity threshold according to the distance value; wherein the distance value and the quantity threshold have a preset corresponding relationship; comparing the quantity of point cloud points in the dynamic search region with the corresponding quantity threshold; if the quantity of point cloud points in the dynamic search region is not greater than the corresponding quantity threshold, the target point cloud point is removed as a noise point.

[0134] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0135] Obtaining a horizontal angle resolution of the detection radar and a distance variation constant of the detection radar in the vertical direction; obtaining a search radius according to the distance value, the horizontal angle resolution and the distance variation constant; determining the dynamic search region according to the target point cloud point and the search radius.

[0136] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0137] Calculating the product of the distance value, the horizontal angle resolution and the distance variation constant to obtain an initial search radius; determining whether the initial search radius is less than a radius threshold; if yes, determining the radius threshold as the search radius; if no, determining the initial search radius as the search radius.

[0138] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0139] Forming the dynamic search region with the target point cloud point as the center and the search radius as the radius.

[0140] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0141] Calculating a distance ratio between the distance value and a maximum measurement distance of the detection radar; determining a quantity threshold corresponding to the distance ratio according to a preset corresponding relationship between the quantity threshold and the distance ratio range.

[0142] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0143] Matching the distance ratio with the distance ratio range to determine a target distance ratio range to which the distance ratio belongs; determining the quantity threshold corresponding to the distance ratio as the quantity threshold corresponding to the target distance ratio range according to the corresponding relationship between the quantity threshold and the distance ratio range.

[0144] In one embodiment, the processor, when executing the computer program, also implements the following steps:

[0145] Obtaining an Euclidean distance between the other point cloud point and the target point cloud point; if the Euclidean distance is less than or equal to the search radius, the other point cloud point is regarded as a point cloud point in the dynamic search area; and counting the point cloud points in the dynamic search area to obtain a number of point cloud points in the dynamic search area.

[0146] In one embodiment, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the following steps:

[0147] Obtaining a distance value between the target point cloud point and the detection radar; and obtaining a dynamic search area based on the distance value, wherein the dynamic search area changes with the distance value; obtaining a corresponding number threshold based on the distance value, wherein the distance value and the number threshold have a preset corresponding relationship; and comparing the number of point cloud points in the dynamic search area with the corresponding number threshold, wherein if the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold, the target point cloud point is removed as a noise point.

[0148] In one embodiment, the computer program is executed by the processor to further implement the following steps:

[0149] Obtaining a horizontal angle resolution of the detection radar and a distance variation constant of the detection radar in a vertical direction; obtaining a search radius based on the distance value, the horizontal angle resolution and the distance variation constant; and determining the dynamic search area based on the target point cloud point and the search radius.

[0150] In one embodiment, the computer program is executed by the processor to further implement the following steps:

[0151] Calculating a product of the distance value, the horizontal angle resolution and the distance variation constant to obtain an initial search radius; determining whether the initial search radius is less than a radius threshold; if yes, determining the radius threshold as the search radius; and if no, determining the initial search radius as the search radius.

[0152] In one embodiment, the computer program is executed by the processor to further implement the following steps:

[0153] Forming the dynamic search area with the target point cloud point as the center and the search radius as the radius.

[0154] In one embodiment, the computer program is executed by the processor to further implement the following steps:

[0155] Calculating a distance ratio between the distance value and a maximum measurement distance of the detection radar; and determining a number threshold corresponding to the distance ratio based on a preset corresponding relationship between the number threshold and a distance ratio range.

[0156] In one embodiment, the computer program is executed by the processor to further implement the following steps:

[0157] The distance ratio is matched with the distance ratio range to determine a target distance ratio range to which the distance ratio belongs; and a quantity threshold corresponding to the target distance ratio range is determined as the quantity threshold corresponding to the distance ratio according to a preset correspondence between quantity thresholds and distance ratio ranges.

[0158] In one embodiment, the computer program, when executed by the processor, further implements the following steps:

[0159] The Euclidean distance between the target point cloud point and other point cloud points is obtained; if the Euclidean distance is less than or equal to the search radius, the other point cloud points are regarded as point cloud points in the dynamic search area; and the point cloud points in the dynamic search area are counted to obtain the number of point cloud points in the dynamic search area.

[0160] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, the processes of the above-mentioned embodiments can be included. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0161] Each technical feature of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of each technical feature in the above embodiments are not described, however, as long as the combination of these technical features does not exist contradictory, it should be considered as the scope of the present application.

[0162] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of the patent of the present application should be subject to the appended claims.

Claims

1. A point cloud denoising method, characterized in that, The method includes: Obtain the distance values ​​between the target point cloud and the detection radar; Based on the distance value, a dynamic search region is obtained; wherein, the dynamic search region changes as the distance value changes; Calculate the distance ratio between the distance value and the maximum measurement range of the detection radar; The quantity threshold corresponding to the distance ratio is determined according to the preset correspondence between the quantity threshold and the distance ratio range; wherein, the distance value and the quantity threshold have a preset correspondence; the quantity threshold is the quantity threshold of point cloud points; the larger the distance ratio, the smaller the quantity threshold; Compare the number of point cloud points within the dynamic search area with the corresponding number threshold; If the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold, then the target point cloud point is removed as a noise point.

2. The method according to claim 1, characterized in that, The step of obtaining the dynamic search region based on the distance value includes: Obtain the horizontal angular resolution of the detection radar and the distance variation constant of the detection radar in the vertical direction; The search radius is obtained based on the distance value, the horizontal angular resolution, and the distance variation constant. The dynamic search area is determined based on the target point cloud and the search radius.

3. The method according to claim 2, characterized in that, The step of obtaining the search radius based on the distance value, the horizontal angular resolution, and the distance variation constant includes: The initial search radius is obtained by calculating the product of the distance value, the horizontal angular resolution, and the distance variation constant. Determine whether the initial search radius is less than the radius threshold; If so, then the radius threshold is determined as the search radius; If not, then the initial search radius is determined to be the search radius.

4. The method according to claim 2, characterized in that, Determining the dynamic search area based on the target point cloud points and the search radius includes: The dynamic search area is formed with the target point cloud as the center and the search radius as the radius.

5. The method according to claim 1, characterized in that, The step of determining the quantity threshold corresponding to the distance ratio based on the preset correspondence between the quantity threshold and the distance ratio range includes: Match the distance ratio with the range of distance ratios to determine the target distance ratio range to which the distance ratio belongs; Based on the preset correspondence between the quantity threshold and the distance ratio range, the quantity threshold corresponding to the target distance ratio range is determined as the quantity threshold corresponding to the distance ratio.

6. The method according to claim 2, characterized in that, The method further includes: Obtain the Euclidean distance between other point cloud points and the target point cloud point; If the Euclidean distance is less than or equal to the search radius, then the other point cloud points are taken as point cloud points within the dynamic search area; The number of point cloud points within the dynamic search area is obtained by counting the point cloud points within the dynamic search area.

7. The method according to claim 1, characterized in that, The method further includes: If the number of point cloud points in the dynamic search area is greater than the corresponding number threshold, then the target point cloud point is determined not to be a noise point, and the target point cloud point is retained.

8. A point cloud noise reduction device, characterized in that, The device includes: The distance acquisition module is used to acquire the distance value between the target point cloud and the detection radar; A region calculation module is used to calculate a dynamic search region based on the distance value; wherein the dynamic search region changes as the distance value changes; The quantity acquisition module is used to calculate the distance ratio between the distance value and the maximum measurement distance of the detection radar, and to determine the quantity threshold corresponding to the distance ratio based on the preset correspondence between the quantity threshold and the distance ratio range; wherein, the distance value and the quantity threshold have a preset correspondence; the quantity threshold is the quantity threshold of point cloud points; the larger the distance ratio, the smaller the quantity threshold; The threshold comparison module is used to compare the number of point cloud points in the dynamic search area with the corresponding number threshold. The noise removal module is used to remove the target point cloud points as noise points when the number of point cloud points in the dynamic search area is not greater than the corresponding number threshold.

9. A detection radar, comprising a signal transmitter, a signal receiver, a memory, and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images