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

Abstract

The present application relates to a point cloud denoising method, device, detection radar and storage medium. The method comprises: obtaining a point cloud set to be processed from an original point cloud set; for each point cloud point to be processed in the point cloud set to be processed, obtaining the pulse width of the electromagnetic signal corresponding to the point cloud point to be processed to reach the first intensity, and obtaining the pulse width time correction value according to the pulse width of the electromagnetic signal; for each point cloud point to be processed in the point cloud set to be processed, obtaining the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed, and obtaining the rising edge time correction value according to the rising edge time difference of the electromagnetic signal; calculating the difference between the pulse width time correction value and the rising edge time correction value; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, then removing the certain point cloud point to be processed as an abnormal point from the original point cloud set. The use of this method can improve the detection accuracy and authenticity of the detection radar.

Description

Technical Field

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

[0002] Point cloud is a collection of massive points that express the surface characteristics of an object in the same spatial reference system, which can be obtained through LiDAR. LiDAR emits detection light in multiple directions, and calculates the distance between the object in the current direction and itself by the time difference from the emission to the return of the detection light, thereby achieving a certain degree of detection of the environment.

[0003] When the edge of an object closer to the laser radar blocks a target farther from the laser radar, since the detection light emitted by the laser radar is a single beam of detection light with a certain direction and a certain cross-sectional area, part of the single beam of detection light is irradiated on the object closer to the laser radar, and the other part is irradiated on the object farther from the laser radar, resulting in the distance value obtained by the single beam of detection light being neither the distance value of the object closer to the laser radar nor the distance value of the object farther from the laser radar, but a distance value between the two. In fact, there is no object between the two objects, but there is an abnormal point cloud representing the distance value, which is the drag point.

[0004] The dragging point will obviously affect the lidar's judgment of the environment ahead, making the empty area between the two objects look like a slope, thus affecting the accuracy and authenticity of the lidar detection. Summary of the invention

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

[0006] A point cloud denoising method, comprising:

[0007] Obtain a point cloud set to be processed from an original point cloud set;

[0008] For each point cloud point to be processed in the point cloud set to be processed, a pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching a first intensity is obtained, and a pulse width time correction value of each point cloud point to be processed is obtained according to the pulse width of the electromagnetic signal;

[0009] For each point cloud point to be processed in the point cloud set to be processed, the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed is obtained, and according to the rising edge time difference of the electromagnetic signal, the rising edge time correction value of each point cloud point to be processed is obtained; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength;

[0010] The difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed is calculated; if the difference between the pulse width time correction value and the rising edge time correction value of a point cloud point to be processed is greater than a preset threshold, the point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0011] In one embodiment, obtaining a point cloud set to be processed from an original point cloud set includes:

[0012] Clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set;

[0013] Clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set;

[0014] Performing a union operation on at least one first point cloud set and at least one second point cloud set to obtain an initial point cloud set;

[0015] The initial point cloud set in the original point cloud set is removed to obtain the point cloud set to be processed.

[0016] In one embodiment, clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set includes:

[0017] Calculate the distance between the point cloud points that are adjacent in order;

[0018] Acquire point cloud points whose distance is less than a distance threshold to form at least a first original point cloud set;

[0019] The number of point cloud points in at least one first original point cloud set is counted, and a first original point cloud set whose number is greater than a first number threshold is obtained as a first point cloud set.

[0020] In one embodiment, clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set includes:

[0021] Calculate and obtain the angle between the straight lines formed by sequentially adjacent point cloud points;

[0022] Acquire point cloud points whose angles are smaller than an angle threshold to form at least one second original point cloud set;

[0023] The number of point cloud points in the second original point cloud set is counted, and a second original point cloud set whose number is greater than a second number threshold is obtained as a second point cloud set.

[0024] In one embodiment, obtaining the pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal includes:

[0025] Determine the pulse width time correction value corresponding to the pulse width of the electromagnetic signal according to the pulse width of the electromagnetic signal and the preset corresponding relationship, and obtain the pulse width time correction value of each point cloud point to be processed;

[0026] According to the rising edge time difference of the electromagnetic signal, the rising edge time correction value of each point cloud point to be processed is obtained, including:

[0027] According to the rising edge time difference of the electromagnetic signal and the preset corresponding relationship, the rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined to obtain the rising edge time correction value of each point cloud point to be processed; wherein the preset corresponding relationship includes the corresponding relationship between the target parameter and the time correction value, and the target parameter includes the pulse width and the rising edge time difference.

[0028] In one embodiment, the method further comprises:

[0029] At least two test electromagnetic signals are transmitted and received in descending order of signal strength, and a preset corresponding relationship is generated according to the at least two test electromagnetic signals.

[0030] In one embodiment, generating a preset correspondence relationship according to at least two test electromagnetic signals includes:

[0031] acquiring a target parameter of each of the at least two test electromagnetic signals;

[0032] Obtaining a time difference between a moment when the signal strength of the test electromagnetic signal first reaches a first strength and a moment when the signal strength of the test electromagnetic signal with a maximum preset strength first reaches the first strength, as a time correction value corresponding to the test electromagnetic signal;

[0033] The target parameter of the test electromagnetic signal is associated with the corresponding time correction value to obtain a preset corresponding relationship.

[0034] A point cloud denoising device, comprising:

[0035] A first acquisition module is used to acquire a point cloud set to be processed from an original point cloud set;

[0036] The second acquisition module is used to obtain, for each point cloud point to be processed in the point cloud set to be processed, a pulse width at which the electromagnetic signal corresponding to the point cloud point to be processed reaches a first intensity, and obtain a pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal;

[0037] The third acquisition module is used to obtain the rising edge time difference of the electromagnetic signal corresponding to each point cloud point to be processed in the point cloud set to be processed, and obtain the rising edge time correction value of each point cloud point to be processed according to the rising edge time difference of the electromagnetic signal; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength;

[0038] The point cloud denoising module is used to calculate the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed; if the difference between the pulse width time correction value and the rising edge time correction value of a point cloud point to be processed is greater than a preset threshold, the point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0039] A detection radar comprises a signal transmitter, a signal receiver, a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0040] Obtain a point cloud set to be processed from an original point cloud set;

[0041] For each point cloud point to be processed in the point cloud set to be processed, a pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching a first intensity is obtained, and a pulse width time correction value of each point cloud point to be processed is obtained according to the pulse width of the electromagnetic signal;

[0042] For each point cloud point to be processed in the point cloud set to be processed, the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed is obtained, and according to the rising edge time difference of the electromagnetic signal, the rising edge time correction value of each point cloud point to be processed is obtained; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength;

[0043] The difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed is calculated; if the difference between the pulse width time correction value and the rising edge time correction value of a point cloud point to be processed is greater than a preset threshold, the point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0044] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0045] Obtain a point cloud set to be processed from an original point cloud set;

[0046] For each point cloud point to be processed in the point cloud set to be processed, a pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching a first intensity is obtained, and a pulse width time correction value of each point cloud point to be processed is obtained according to the pulse width of the electromagnetic signal;

[0047] For each point cloud point to be processed in the point cloud set to be processed, the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed is obtained, and according to the rising edge time difference of the electromagnetic signal, the rising edge time correction value of each point cloud point to be processed is obtained; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength;

[0048] The difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed is calculated; if the difference between the pulse width time correction value and the rising edge time correction value of a point cloud point to be processed is greater than a preset threshold, the point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0049] The above-mentioned point cloud denoising method, device, detection radar and storage medium obtain a point cloud set to be processed from the original point cloud set, and for each point cloud point to be processed in the point cloud set, obtain the pulse width of the electromagnetic signal corresponding to the point cloud point to be processed to reach the first intensity, and obtain the pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal; for each point cloud point to be processed in the point cloud set to be processed, obtain the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed, and obtain the rising edge time correction value of each point cloud point to be processed according to the rising edge time difference of the electromagnetic signal; thereby calculating the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed; and when the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, the certain point cloud point to be processed is removed from the original point cloud set as an abnormal point. By utilizing the electromagnetic signal corresponding to the same point cloud point, the pulse width time correction value determined by the pulse width and the rising edge time correction value determined by the rising edge time difference should be equal or meet a certain error range to determine whether the above-mentioned same point cloud point to be processed is a drag point, and then the drag point is removed to achieve point cloud denoising and improve the detection accuracy and authenticity of the detection radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A diagram showing an application environment of a point cloud denoising method in one embodiment;

[0051] Figure 2 Schematic diagram of a process of a point cloud denoising method in one embodiment;

[0052] Figure 3 is a waveform diagram of an electromagnetic signal in one embodiment;

[0053] Figure 4 is a waveform diagram of an electromagnetic signal in another embodiment;

[0054] Figure 5 A schematic diagram of a process for obtaining a point cloud set to be processed in one embodiment;

[0055] Figure 6 A schematic diagram of a process of obtaining a first point cloud set according to an embodiment;

[0056] Figure 7 A schematic diagram of a process of obtaining a second point cloud set according to an embodiment;

[0057] Figure 8 A schematic diagram of the distribution of point cloud points on an object in one embodiment;

[0058] Fig. 9 A schematic diagram of a flow chart for obtaining a preset corresponding relationship in an embodiment;

[0059] Fig.10 A diagram showing a corresponding relationship between a rising edge time difference and a time correction value in an embodiment;

[0060] Fig.11 is a structural block diagram of a point cloud denoising device in one embodiment;

[0061] Fig.12 1 is a diagram of the internal structure of a detection radar in one embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0063] The point cloud denoising method provided in this application can be applied to Figure 1In the application environment shown. Among them, the detection radar 100 emits an electromagnetic signal outward, a part of the electromagnetic signal is irradiated on the object A which is closer to the detection radar 100, and the other part is irradiated on the object B which is farther from the detection radar 100. The distance detected by the detection signal is neither the distance value between the detection radar 100 and the object A, nor the distance value between the detection radar 100 and the object B, but the distance value between the detection radar 100 and a certain point between the object A and the object B, namely the drag point, and there is no association or other objects between the object A and the object B. The essence of point cloud denoising in this application is to remove the above-mentioned drag points in the point cloud points. The point cloud denoising method provided in this application can be implemented directly through the detection radar 100, and can also be implemented through a terminal that communicates with the detection radar 100. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers and portable wearable devices.

[0064] In one embodiment, Figure 2 As shown, in this embodiment, a point cloud denoising method is provided, and the method is applied to Figure 1 The detection radar in the example is used to illustrate, including the following steps:

[0065] S210: Obtain a point cloud set to be processed from the original point cloud set.

[0066] Optionally, the detection radar may be a laser radar, a millimeter wave radar, a microwave radar, or the like used to achieve ranging. In this embodiment, the type of the detection radar is not specifically limited.

[0067] Optionally, the detection radar can emit multi-threaded electromagnetic signals or periodically rotate to emit electromagnetic signals, which hit the object to form a large number of point cloud points, and accordingly obtain the coordinate position of each point cloud point, thereby obtaining an original point cloud set including the coordinate positions of a large number of point cloud points, and filtering the point cloud points in the original point cloud set to obtain the point cloud set to be processed.

[0068] Optionally, the detection radar may filter the point cloud points in the original point cloud set according to the distance between each point cloud point in the original point cloud set and the detection radar to obtain a point cloud set to be processed. For example, the point cloud points in the original point cloud set that are farthest and smallest from the detection radar are removed. The detection radar may also use a clustering algorithm to cluster the point cloud points in the original point cloud set to obtain a clustering result, and remove the clustering results in which the number of point cloud points is less than a preset number in the original point cloud set to obtain a point cloud set to be processed. In this embodiment, the specific method for obtaining the point cloud set to be processed from the original point cloud set is not limited.

[0069] S220, for each point cloud point to be processed in the point cloud set to be processed, obtain the pulse width of the electromagnetic signal corresponding to the point cloud point to be processed that reaches the first intensity, and obtain the pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal.

[0070] During the scanning process, each single beam of electromagnetic signal emitted by the detection radar forms a corresponding cloud point when it hits the object, and is received by the detection radar for photoelectric conversion after being reflected by the object. Photoelectric conversion converts the received electromagnetic signal into an electrical signal, and uses the voltage value to represent the signal strength. Figure 3 As shown, the received electromagnetic signals include pulse rising and falling edges, where curve a is the intensity waveform of the electromagnetic signal received by the detection radar to form a certain point cloud point under ideal conditions. Under ideal conditions, there is no loss in the signal strength of the returned electromagnetic signal, and curve a is in a stable state. Curves b and c are the intensity waveforms of the electromagnetic signal received by the detection radar to form a point cloud point under actual conditions. Under actual conditions, due to the loss caused by the detection radar's own noise, curves b and c increase first and then decrease. Among them, the stronger the signal strength of the emitted electromagnetic signal, the closer the corresponding intensity waveform is to curve a.

[0071] The pulse width of the electromagnetic signal corresponding to the point cloud point to be processed is the time length when the signal strength of the electromagnetic signal obtained by the detection radar reaches the first strength P1, that is, the pulse width. The time difference between the moment when the signal strength of the electromagnetic signal in an ideal situation first reaches the first strength P1 and the moment when the signal strength of the electromagnetic signal in an actual situation first reaches the first strength P1 is the time correction value.

[0072] Optionally, the detection radar uses the corresponding relationship between the pulse width and the time correction value to obtain the time correction value corresponding to the calculated pulse width of the electromagnetic signal, that is, the pulse width time correction value.

[0073] S230, for each point cloud point to be processed in the point cloud set to be processed, obtain the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed, and obtain the rising edge time correction value of each point cloud point to be processed according to the rising edge time difference of the electromagnetic signal.

[0074] The rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when it first reaches the third strength, and the third strength is greater than the second strength.

[0075] Please continue reading Figure 3 The time difference between the moment when the electromagnetic signal strength reaches the second strength P2 for the first time and the moment when the electromagnetic signal strength reaches the third strength P3 for the first time is the rising edge time difference. The third strength P3 may be equal to the first strength P1.

[0076] Optionally, the detection radar uses the corresponding relationship between the rising edge time difference and the time correction value to obtain the time correction value corresponding to the calculated rising edge time difference of the electromagnetic signal, that is, the rising edge time correction value.

[0077] S240, calculating the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, then the certain point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0078] like Figure 4 As shown, curve o represents the waveform of the electromagnetic signal with greater signal strength. Figure 1 In the figure, it can be a waveform diagram of the electromagnetic signal of a certain point cloud point on object A which is closer to the detection radar, curve p represents a waveform diagram of the electromagnetic signal with a smaller signal strength, it can be a waveform diagram of the electromagnetic signal of a certain point cloud point on object B which is farther from the detection radar, and curve q represents a waveform diagram of the electromagnetic signal of the drag point between object A and object B. Among them, curve q is formed by the waveform superposition of the part of the electromagnetic signal (i.e., early wave) that hits object A that returns earlier and the part of the electromagnetic signal (i.e., late wave) that hits object B that returns later. The superposition of the front and back waves will cause the late wave to significantly affect the second half of the early wave, while the impact on the first half is very light. Therefore, the correct time correction value cannot be determined by relying solely on the corresponding relationship between the pulse width and the time correction value. However, there is no signal strength waveform diagram for the superposition of the front and back waves, that is, the waveform diagram of the electromagnetic signal corresponding to the normal point cloud point, the time correction corresponding to the pulse width and the time correction value corresponding to the rising edge time difference should be equal or meet a certain error range. This application uses this feature to determine whether the point cloud point to be processed is an abnormal point, i.e., a drag point.

[0079] Specifically, the detection radar calculates the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed, and compares the difference with the preset threshold value, and then determines whether the point cloud point to be processed is an abnormal point, that is, a drag point, based on the comparison result. If so, the detection radar removes the point cloud point to be processed from the original point cloud set; if not, the detection radar retains the point cloud point. Among them, if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than the preset threshold, the detection radar determines that the certain point cloud point to be processed is the above-mentioned abnormal point; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is not greater than the preset threshold, the detection radar determines that the certain point cloud point to be processed is not the above-mentioned abnormal point.

[0080] In this embodiment, the detection radar obtains a set of point cloud to be processed from the original point cloud set, obtains the pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching the first intensity for each point cloud point to be processed in the point cloud set, and obtains the pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal; obtains the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed for each point cloud point to be processed in the point cloud set, and obtains the rising edge time correction value of each point cloud point to be processed according to the rising edge time difference of the electromagnetic signal; thereby calculating the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed; and when the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, the certain point cloud point to be processed is removed from the original point cloud set as an abnormal point. By utilizing the electromagnetic signal corresponding to the same point cloud point, the pulse width time correction value determined by the pulse width and the rising edge time correction value determined by the rising edge time difference should be equal or meet a certain error range to determine whether the above-mentioned same point cloud point to be processed is a drag point, and then the drag point is removed to achieve point cloud denoising and improve the detection accuracy and authenticity of the detection radar.

[0081] In one embodiment, to reduce the amount of data processing for point cloud denoising, Figure 5 As shown, the above S210 includes:

[0082] S510 , clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set.

[0083] In actual application scenarios, the distances between point cloud points formed on the same object are relatively close. Optionally, the detection radar can cluster the point cloud points according to the Euclidean distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set. For example, the detection radar uses the k-means clustering algorithm to cluster the point cloud points in the original point cloud set, that is, first randomly determine k point cloud points in the point cloud points as cluster center points, compare the Euclidean distances between the point cloud points and each cluster center point, and classify the point cloud points and the cluster center with the closest Euclidean distance into one category, and then recalculate the category center points of the clusters based on the point cloud points of the classified categories, and continuously classify and determine the center points until the center points no longer change, so as to obtain the final clustering result, that is, to obtain at least one first point cloud set. Among them, the same first point cloud set includes point cloud points formed on the same object.

[0084] S520 , clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set.

[0085] In actual application scenarios, the angles between the point cloud points formed on the same object relative to the location of the detection radar are also relatively close. Optionally, the angles between the point cloud points in the original point cloud set can be the angles between the point cloud points relative to the location of the detection radar. The detection radar can cluster the point cloud points according to the angles between the point cloud points in the original point cloud set relative to the location of the detection radar to obtain at least one second point cloud set. For example, the detection radar obtains the angles between the point cloud points in the point cloud set relative to the location of the detection radar, and classifies the point cloud points with angles less than a preset angle into one category to obtain the final clustering result, that is, to obtain at least one second point cloud set. Among them, the same second point cloud set includes point cloud points formed on the same object.

[0086] S530: Perform a union operation on at least one first point cloud set and at least one second point cloud set to obtain an initial point cloud set.

[0087] S540: Remove the initial point cloud set from the original point cloud set to obtain a point cloud set to be processed.

[0088] Optionally, the detection radar performs a union operation on at least one first point cloud set and at least one second point cloud set obtained by the above clustering to obtain an initial point cloud set including normal point cloud points. For example, the original point cloud set includes point cloud points 1 to 1000, and clustering is performed according to distance to obtain three first point cloud sets A1 to A3, which are A1{point cloud points 1 to 400}, A2{point cloud points 470 to 800}, and A3{point cloud points 850 to 1000} respectively. Clustering is performed according to angle to obtain three second point cloud sets B1 to B3, which are B1{point cloud points 1 to 405}, B2{point cloud points 472 to 792}, and B3{point cloud points 851 to 1000} respectively. The detection radar further performs a union operation on the three first point cloud sets A1 to A3 and the three second point cloud sets B1 to B3 obtained by the above clustering, thereby obtaining an initial point cloud set C{point cloud points 1 to 405, point cloud points 470 to 800, point cloud points 850 to 1000} including normal point cloud points. The detection radar further removes the initial point cloud set in the original point cloud set to obtain the point cloud set to be processed. Based on the above content, the point cloud set to be processed D{point cloud points 406-469, point cloud points 801-849} is finally obtained.

[0089] In an optional embodiment, in order to improve the accuracy of clustering based on distance, Figure 6 As shown, the above S510 includes:

[0090] S610: Calculate the distances between sequentially adjacent point cloud points.

[0091] Specifically, the detection radar calculates the distance between adjacent point cloud points in the acquisition order according to the acquisition order of each point cloud point in the original point cloud set. For example, the detection radar acquires point cloud points 1 to 1000 in sequence to form the above original point cloud set, and then calculates the distance between the next point cloud point and the previous point cloud point in the acquisition order, that is, the distance L1 between point cloud point 2 and point cloud point 1 is acquired, the distance L2 between point cloud point 3 and point cloud point 2 is acquired, the distance L3 between point cloud point 4 and point cloud point 3 is acquired... until the distance L999 between point cloud point 1000 and point cloud point 999 is acquired.

[0092] S620: Acquire point cloud points whose distance is less than a distance threshold to form at least one first original point cloud set.

[0093] Specifically, the detection radar obtains a distance less than the distance threshold in the calculated distance, and correspondingly obtains the point cloud points of the distance to form at least one first original point cloud set. For example, among the above distances L1 to L999, the distances L1 to L399, the distances L400 to L419, the distances L420 to L709, and the distances L710 to L999 are all greater than the distance threshold L, and the first original point cloud set E1 {point cloud points 1 to 400}, the first original point cloud set E2 {point cloud points 401 to 420}, the first original point cloud set E3 {point cloud points 420 to 710}, and the first original point cloud set E4 {point cloud points 711 to 1000} are correspondingly determined.

[0094] S630: Count the number of point cloud points in at least one first original point cloud set, and obtain a first original point cloud set whose number is greater than a first number threshold as a first point cloud set.

[0095] Specifically, the detection radar counts the number of point cloud points in each first original point cloud set, and obtains the first original point cloud set whose number is greater than the first number threshold as the first point cloud set. For example, the number of point cloud points in the first original point cloud set E1 {point cloud points 1 to 400} is 400, the number of point cloud points in the first original point cloud set E2 {point cloud points 401 to 420} is 20, the number of point cloud points in the first original point cloud set E3 {point cloud points 421 to 710} is 290, and the number of point cloud points in the first original point cloud set E4 {point cloud points 711 to 1000} is 290. The first number threshold e=100, and the detection radar determines that the first original point cloud sets E1, E3, and E4 are the first point cloud sets respectively. The detection radar determines that the first original point cloud set with a smaller number of point cloud points may include abnormal points, that is, the first original point cloud set E2 {point cloud points 401 to 420} may include abnormal points.

[0096] In an optional embodiment, in order to improve the accuracy of clustering according to angles, Figure 7As shown, the above S520 includes:

[0097] 710. Calculate and obtain the angle between straight lines formed by sequentially adjacent point cloud points.

[0098] Specifically, the detection radar calculates the angle between the straight lines formed by the adjacent point cloud points in the acquisition order according to the acquisition order of each point cloud point in the original point cloud set. Figure 8 As shown, P i-1 , P i , P i+1 , and P i+2 To obtain four consecutive point cloud points in the above original point cloud set, the detection radar obtains the point cloud point P i-1 and point cloud point P i The formed straight line S1 and point cloud point P i and point cloud point P i+1 The angle θ between the straight lines S2 formed i The detection radar obtains point cloud points 1 to 1000 in sequence to form the above original point cloud set, and the corresponding angle θ is obtained 2 ~θ 999 .

[0099] 720. Obtain point cloud points whose angles are smaller than an angle threshold to form at least one second original point cloud set.

[0100] Specifically, the detection radar obtains an angle less than the angle threshold from the calculated angles. The angle less than the angle threshold indicates that the point cloud points forming the angle are collinear, i.e., on the same plane (i.e., the same object). The point cloud points corresponding to the angle are obtained to form at least one second original point cloud set. 2 ~θ 999 Correspondingly, the second original point cloud set F1 {point cloud points 1 to 185}, the second original point cloud set F2 {point cloud points 185 to 385}, the second original point cloud set F3 {point cloud points 386 to 426}, the second original point cloud set F4 {point cloud points 426 to 723}, and the second original point cloud set F5 {point cloud points 724 to 1000} are determined.

[0101] 730 . Count the number of point cloud points in the second original point cloud set, and obtain a second original point cloud set whose number is greater than a second number threshold as a second point cloud set.

[0102] Specifically, the detection radar counts the number of point cloud points in each second original point cloud set, and obtains the second original point cloud set whose number is greater than the second number threshold as the second point cloud set. For example, the number of point cloud points in the second original point cloud set F1 {point cloud points 1 to 185} is 185, the number of point cloud points in the second original point cloud set F2 {point cloud points 185 to 385} is 201, the number of point cloud points in the second original point cloud set F3 {point cloud points 386 to 426} is 41, the number of point cloud points in the first original point cloud set F4 {point cloud points 426 to 723} is 298, and the number of point cloud points in the first original point cloud set F5 {point cloud points 724 to 1000} is 277. The second number threshold f can be equal to the first number and e, and the detection radar determines the first original point cloud sets F1, F2, E4, and E5 as the second point cloud sets respectively. Similarly, the detection radar determines that the second original point cloud set with a smaller number of point cloud points may include abnormal points, that is, the second original point cloud set F3 {point cloud points 386-426} may include abnormal points.

[0103] In this embodiment, the detection radar clusters the point cloud points according to the distance between the point cloud points in the original point cloud set to obtain at least one first point cloud set, and clusters the point cloud points according to the angle between the point cloud points in the original point cloud set to obtain at least one second point cloud set, and then performs a merge operation on the at least one first point cloud set and the at least one second point cloud set to obtain an initial point cloud set, and then removes the initial point cloud set in the original point cloud set to obtain a point cloud set to be processed. The above-mentioned clustering method using distance and angle can accurately perform classification, and by utilizing the dense distribution of point cloud points on real objects, the small distance between them, and the characteristics of collinearity, the normal point cloud points located on the real object can be accurately screened out from the original point cloud set to obtain the initial point cloud set, and the initial point cloud set including normal points in the original point cloud set can be removed to obtain the point cloud set to be processed that needs to be further judged as abnormal points, thereby reducing the data processing amount of point cloud denoising and improving the denoising efficiency.

[0104] In one embodiment, to further simplify the denoising process, the above 220 includes:

[0105] The pulse width time correction value corresponding to the pulse width of the electromagnetic signal is determined according to the pulse width of the electromagnetic signal and a preset corresponding relationship, and the pulse width time correction value of each point cloud point to be processed is obtained.

[0106] Accordingly, the above S230 includes:

[0107] The rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined according to the rising edge time difference of the electromagnetic signal and a preset corresponding relationship, and the rising edge time correction value of each point cloud point to be processed is obtained.

[0108] The preset corresponding relationship includes the corresponding relationship between the target parameter and the time correction value, and the target parameter includes the pulse width and the rising edge time difference.

[0109] Optionally, before performing the above-mentioned point cloud denoising method, the detection radar pre-transmits and receives electromagnetic signals of different signal strengths, and obtains the time correction value, pulse width and rising edge time difference corresponding to the electromagnetic signals of different signal strengths, and then constructs the corresponding relationship between the time correction value, pulse width and rising edge time difference of the electromagnetic signals of different signal strengths to obtain the above-mentioned preset corresponding relationship.

[0110] In an optional embodiment, the detection radar transmits at least two test electromagnetic signals in descending order of signal strength, and generates a preset corresponding relationship based on the at least two test electromagnetic signals.

[0111] like Fig. 9 As shown, generating a preset corresponding relationship includes the following steps:

[0112] S910, obtaining a target parameter of each test electromagnetic signal of at least two test electromagnetic signals.

[0113] Specifically, the detection radar can transmit at least two test electromagnetic signals to a flat surface and receive the test electromagnetic signals reflected by the flat surface. The signal strength loss caused by the test electromagnetic signals during transmission and reflection can be ignored. The detection radar obtains the pulse width and rising edge time difference of each of the at least two test electromagnetic signals received.

[0114] S920. Obtain a time difference between the moment when the signal strength of the test electromagnetic signal first reaches a first strength and the moment when the signal strength of the test electromagnetic signal with the maximum preset strength first reaches the first strength, as a time correction value corresponding to the test electromagnetic signal.

[0115] S930: Associating the target parameter of the test electromagnetic signal with the corresponding time correction value to obtain a preset corresponding relationship.

[0116] The maximum preset strength test electromagnetic signal is the electromagnetic signal with the maximum signal strength that can be emitted by the detection radar. The greater the signal strength of the electromagnetic signal, the smaller the influence of the detection radar's own noise on the signal strength. Therefore, the waveform corresponding to the maximum preset strength electromagnetic signal can be approximated as follows: Figure 3 Curve a in .

[0117] Specifically, the detection radar calculates the time difference between the moment when each test electromagnetic signal first reaches the first intensity and the moment when the signal intensity of the test electromagnetic signal with the maximum preset intensity first reaches the first intensity as the time correction value corresponding to the test electromagnetic signal, and associates the pulse width and rising edge time difference of each electromagnetic signal obtained with the corresponding time correction value, thereby obtaining the above-mentioned preset corresponding relationship. Fig.10 That is, the corresponding relationship between the rising edge time difference obtained by plotting and drawing and the corresponding time correction value. Similarly, the corresponding relationship between the pulse width and the corresponding time correction value can also be obtained, which together constitute the above preset corresponding relationship.

[0118] In this embodiment, the detection radar pre-establishes a preset correspondence between the target parameters and the time correction value, so that the pulse width time correction value corresponding to the pulse width and the rising edge time correction value corresponding to the rising edge time difference can be directly determined by using the preset correspondence. This simplifies the process of determining the corresponding time correction value according to different parameters, further simplifies the denoising process, and improves the denoising efficiency.

[0119] It should be understood that although Figure 2-9 The steps in the flowchart are shown in sequence as indicated by the arrows, but the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of the steps, and the steps can be executed in other orders. Moreover, Figure 2-9 At least part of the steps in the flowchart may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the steps or stages in other steps.

[0120] In one embodiment, Fig.11 As shown, a point cloud denoising device is provided, comprising: a first acquisition module 1101, a second acquisition module 1102, a third acquisition module 1103 and a point cloud denoising module 1104, wherein:

[0121] The first acquisition module 1101 is used to acquire a point cloud set to be processed from an original point cloud set;

[0122] The second acquisition module 1102 is used to obtain the pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching the first intensity for each point cloud point to be processed in the point cloud set to be processed, and obtain the pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal;

[0123] The third acquisition module 1103 is used to obtain the rising edge time difference of the electromagnetic signal corresponding to each point cloud point to be processed in the point cloud set to be processed, and obtain the rising edge time correction value of each point cloud point to be processed according to the rising edge time difference of the electromagnetic signal; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when it first reaches the third strength, and the third strength is greater than the second strength;

[0124] The point cloud denoising module 1104 is used to calculate the difference between the pulse width time correction value and the rising edge time correction value of each point cloud point to be processed; if the difference between the pulse width time correction value and the rising edge time correction value of a point cloud point to be processed is greater than a preset threshold, the point cloud point to be processed is removed from the original point cloud set as an abnormal point.

[0125] In one embodiment, the first acquisition module 1101 is specifically used for:

[0126] Clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set; clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set; performing a union operation on the at least one first point cloud set and the at least one second point cloud set to obtain an initial point cloud set; removing the initial point cloud set in the original point cloud set to obtain a point cloud set to be processed.

[0127] In one embodiment, the first acquisition module 1101 is specifically used for:

[0128] Calculate and obtain the distance between sequentially adjacent point cloud points; obtain point cloud points whose distance is less than a distance threshold to form at least one first original point cloud set; count the number of point cloud points in at least one first original point cloud set, and obtain a first original point cloud set whose number is greater than a first number threshold as the first point cloud set.

[0129] In one embodiment, the first acquisition module 1101 is specifically used for:

[0130] Calculate and obtain the angle between the straight lines formed by sequentially adjacent point cloud points; obtain point cloud points whose angles are less than an angle threshold to form at least one second original point cloud set; count the number of point cloud points in the second original point cloud set, and obtain a second original point cloud set whose number is greater than a second number threshold as the second point cloud set.

[0131] In one embodiment, the second acquisition module 1102 is specifically used for:

[0132] According to the pulse width of the electromagnetic signal and the preset corresponding relationship, the pulse width time correction value corresponding to the pulse width of the electromagnetic signal is determined to obtain the pulse width time correction value of each point cloud point to be processed; according to the rising edge time difference of the electromagnetic signal and the preset corresponding relationship, the rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined to obtain the rising edge time correction value of each point cloud point to be processed; wherein the preset corresponding relationship includes the corresponding relationship between the target parameters and the time correction value, and the target parameters include the pulse width and the rising edge time difference.

[0133] In one embodiment, the above device further includes a signal testing module, which is used to:

[0134] At least two test electromagnetic signals are transmitted and received in descending order of signal strength, and a preset corresponding relationship is generated according to the at least two test electromagnetic signals.

[0135] In one embodiment, the signal testing module is specifically used for:

[0136] Obtain target parameters of each test electromagnetic signal of at least two test electromagnetic signals; obtain the time difference between the moment when the signal strength of the test electromagnetic signal first reaches a first intensity and the moment when the signal strength of the test electromagnetic signal with a maximum preset intensity first reaches the first intensity, as a time correction value corresponding to the test electromagnetic signal; associate the target parameters of the test electromagnetic signal with the corresponding time correction value to obtain a preset corresponding relationship.

[0137] For the specific definition of the point cloud denoising device, please refer to the definition of the point cloud denoising method above, which will not be repeated here. Each module in the above-mentioned point cloud denoising device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0138] In one embodiment, a detection radar is provided, and the internal structure diagram of the detection radar can be as follows: Fig.12As shown. The detection radar includes a signal transmitter, a signal receiver, a processor, a memory and a network interface connected through a system bus. Among them, the signal transmitter of the detection radar is used to transmit electromagnetic signals, and the signal receiver is used to receive electromagnetic signals. The processor of the detection radar is used 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 operation of the operating system and the computer program in the non-volatile storage medium. The database of the detection radar is used to store the coordinate position data of the point cloud points. The network interface of the detection radar is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a point cloud denoising method is implemented.

[0139] Those skilled in the art will understand that Fig.12 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0141] A set of unprocessed point clouds is obtained from the original point cloud set; for each unprocessed point cloud point in the set of unprocessed point cloud points, a pulse width of the electromagnetic signal corresponding to the unprocessed point cloud point reaching a first intensity is obtained, and a pulse width time correction value of each unprocessed point cloud point is obtained according to the pulse width of the electromagnetic signal; for each unprocessed point cloud point in the set of unprocessed point cloud points, a rising edge time difference of the electromagnetic signal corresponding to the unprocessed point cloud point is obtained, and a rising edge time correction value of each unprocessed point cloud point is obtained according to the rising edge time difference of the electromagnetic signal; wherein the rising edge time difference is the time difference between the moment when the signal intensity of the electromagnetic signal corresponding to the unprocessed point cloud point first reaches the second intensity and the moment when it first reaches the third intensity, and the third intensity is greater than the second intensity; the difference between the pulse width time correction value and the rising edge time correction value of each unprocessed point cloud point is calculated; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point is greater than a preset threshold value, the certain point cloud point is removed from the original point cloud set as an abnormal point.

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

[0143] Clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set; clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set; performing a union operation on the at least one first point cloud set and the at least one second point cloud set to obtain an initial point cloud set; removing the initial point cloud set in the original point cloud set to obtain a point cloud set to be processed.

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

[0145] Calculate and obtain the distance between sequentially adjacent point cloud points; obtain point cloud points whose distance is less than a distance threshold to form at least one first original point cloud set; count the number of point cloud points in at least one first original point cloud set, and obtain a first original point cloud set whose number is greater than a first number threshold as the first point cloud set.

[0146] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0147] Calculate and obtain the angle between the straight lines formed by sequentially adjacent point cloud points; obtain point cloud points whose angles are less than an angle threshold to form at least one second original point cloud set; count the number of point cloud points in the second original point cloud set, and obtain a second original point cloud set whose number is greater than a second number threshold as the second point cloud set.

[0148] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0149] According to the pulse width of the electromagnetic signal and the preset corresponding relationship, the pulse width time correction value corresponding to the pulse width of the electromagnetic signal is determined to obtain the pulse width time correction value of each point cloud point to be processed; according to the rising edge time difference of the electromagnetic signal and the preset corresponding relationship, the rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined to obtain the rising edge time correction value of each point cloud point to be processed; wherein the preset corresponding relationship includes the corresponding relationship between the target parameters and the time correction value, and the target parameters include the pulse width and the rising edge time difference.

[0150] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0151] At least two test electromagnetic signals are transmitted and received in descending order of signal strength, and a preset corresponding relationship is generated according to the at least two test electromagnetic signals.

[0152] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0153] Obtain target parameters of each test electromagnetic signal of at least two test electromagnetic signals; obtain the time difference between the moment when the signal strength of the test electromagnetic signal first reaches a first intensity and the moment when the signal strength of the test electromagnetic signal with a maximum preset intensity first reaches the first intensity, as a time correction value corresponding to the test electromagnetic signal; associate the target parameters of the test electromagnetic signal with the corresponding time correction value to obtain a preset corresponding relationship.

[0154] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0155] A set of unprocessed point clouds is obtained from the original point cloud set; for each unprocessed point cloud point in the set of unprocessed point cloud points, a pulse width of the electromagnetic signal corresponding to the unprocessed point cloud point reaching a first intensity is obtained, and a pulse width time correction value of each unprocessed point cloud point is obtained according to the pulse width of the electromagnetic signal; for each unprocessed point cloud point in the set of unprocessed point cloud points, a rising edge time difference of the electromagnetic signal corresponding to the unprocessed point cloud point is obtained, and a rising edge time correction value of each unprocessed point cloud point is obtained according to the rising edge time difference of the electromagnetic signal; wherein the rising edge time difference is the time difference between the moment when the signal intensity of the electromagnetic signal corresponding to the unprocessed point cloud point first reaches the second intensity and the moment when it first reaches the third intensity, and the third intensity is greater than the second intensity; the difference between the pulse width time correction value and the rising edge time correction value of each unprocessed point cloud point is calculated; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point is greater than a preset threshold value, the certain point cloud point is removed from the original point cloud set as an abnormal point.

[0156] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0157] Clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set; clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set; performing a union operation on the at least one first point cloud set and the at least one second point cloud set to obtain an initial point cloud set; removing the initial point cloud set in the original point cloud set to obtain a point cloud set to be processed.

[0158] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0159] Calculate and obtain the distance between sequentially adjacent point cloud points; obtain point cloud points whose distance is less than a distance threshold to form at least one first original point cloud set; count the number of point cloud points in at least one first original point cloud set, and obtain a first original point cloud set whose number is greater than a first number threshold as the first point cloud set.

[0160] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0161] Calculate and obtain the angle between the straight lines formed by sequentially adjacent point cloud points; obtain point cloud points whose angles are less than an angle threshold to form at least one second original point cloud set; count the number of point cloud points in the second original point cloud set, and obtain a second original point cloud set whose number is greater than a second number threshold as the second point cloud set.

[0162] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0163] According to the pulse width of the electromagnetic signal and the preset corresponding relationship, the pulse width time correction value corresponding to the pulse width of the electromagnetic signal is determined to obtain the pulse width time correction value of each point cloud point to be processed; according to the rising edge time difference of the electromagnetic signal and the preset corresponding relationship, the rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined to obtain the rising edge time correction value of each point cloud point to be processed; wherein the preset corresponding relationship includes the corresponding relationship between the target parameters and the time correction value, and the target parameters include the pulse width and the rising edge time difference.

[0164] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0165] At least two test electromagnetic signals are transmitted and received in descending order of signal strength, and a preset corresponding relationship is generated according to the at least two test electromagnetic signals.

[0166] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented:

[0167] Obtain target parameters of each test electromagnetic signal of at least two test electromagnetic signals; obtain the time difference between the moment when the signal strength of the test electromagnetic signal first reaches a first intensity and the moment when the signal strength of the test electromagnetic signal with a maximum preset intensity first reaches the first intensity, as a time correction value corresponding to the test electromagnetic signal; associate the target parameters of the test electromagnetic signal with the corresponding time correction value to obtain a preset corresponding relationship.

[0168] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this 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, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0169] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0170] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A point cloud denoising method, characterized in that: The method comprises: Obtain a point cloud set to be processed from an original point cloud set; For each point cloud point to be processed in the point cloud set to be processed, a pulse width of the electromagnetic signal corresponding to the point cloud point to be processed reaching a first intensity is obtained, and according to the pulse width of the electromagnetic signal, a pulse width time correction value of each point cloud point to be processed is obtained; For each point cloud point to be processed in the point cloud set to be processed, the rising edge time difference of the electromagnetic signal corresponding to the point cloud point to be processed is correspondingly obtained, and according to the rising edge time difference of the electromagnetic signal, the rising edge time correction value of each point cloud point to be processed is obtained; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the point cloud point to be processed first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength; Calculate the difference between the pulse width time correction value and the rising edge time correction value of each of the point cloud points to be processed; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, remove the certain point cloud point to be processed from the original point cloud set as an abnormal point.

2. The method according to claim 1, characterized in that The step of obtaining a to-be-processed point cloud set from an original point cloud set includes: Clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set; Clustering the point cloud points according to angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set; Performing a union operation on the at least one first point cloud set and the at least one second point cloud set to obtain an initial point cloud set; The initial point cloud set in the original point cloud set is removed to obtain the point cloud set to be processed.

3. The method according to claim 2, characterized in that The step of clustering the point cloud points according to the distances between the point cloud points in the original point cloud set to obtain at least one first point cloud set includes: Calculate the distance between the point cloud points that are adjacent in order; Acquire point cloud points whose distance is less than a distance threshold to form at least a first original point cloud set; The number of point cloud points in the at least one first original point cloud set is counted, and a first original point cloud set whose number is greater than a first number threshold is obtained as the first point cloud set.

4. The method according to claim 2, characterized in that: The step of clustering the point cloud points according to the angles between the point cloud points in the original point cloud set to obtain at least one second point cloud set includes: Calculate and obtain the angle between the straight lines formed by sequentially adjacent point cloud points; Acquire point cloud points whose angles are smaller than the angle threshold to form at least one second original point cloud set; The number of point cloud points in the second original point cloud set is counted, and a second original point cloud set whose number is greater than a second number threshold is obtained as the second point cloud set.

5. The method according to any one of claims 1 to 4, characterized in that: The step of obtaining the pulse width time correction value of each of the point cloud points to be processed according to the pulse width of the electromagnetic signal includes: Determine a pulse width time correction value corresponding to the pulse width of the electromagnetic signal according to the pulse width of the electromagnetic signal and a preset corresponding relationship, and obtain the pulse width time correction value of each point cloud point to be processed; The step of obtaining a rising edge time correction value of each of the point cloud points to be processed according to the rising edge time difference of the electromagnetic signal comprises: According to the rising edge time difference of the electromagnetic signal and the preset corresponding relationship, the rising edge time correction value corresponding to the rising edge time difference of the electromagnetic signal is determined to obtain the rising edge time correction value of each point cloud point to be processed; wherein the preset corresponding relationship includes the corresponding relationship between the target parameter and the time correction value, and the target parameter includes the pulse width and the rising edge time difference.

6. The method according to claim 5, characterized in that The method further comprises: At least two test electromagnetic signals are transmitted and received in descending order of signal strength, and the preset corresponding relationship is generated according to the at least two test electromagnetic signals.

7. The method according to claim 6, characterized in that Generating the preset corresponding relationship according to the at least two test electromagnetic signals includes: Acquiring a target parameter of each of the at least two test electromagnetic signals; Obtaining a time difference between a moment when the signal strength of the test electromagnetic signal first reaches the first strength and a moment when the signal strength of the test electromagnetic signal with the maximum preset strength first reaches the first strength as a time correction value corresponding to the test electromagnetic signal; The target parameter of the test electromagnetic signal is associated with the corresponding time correction value to obtain the preset corresponding relationship.

8. A point cloud denoising device, characterized in that: The device comprises: A first acquisition module is used to acquire a point cloud set to be processed from an original point cloud set; A second acquisition module is used to obtain, for each point cloud point to be processed in the point cloud set to be processed, a pulse width of the electromagnetic signal corresponding to the point cloud point to be processed that reaches a first intensity, and obtain a pulse width time correction value of each point cloud point to be processed according to the pulse width of the electromagnetic signal; A third acquisition module is used to obtain, for each of the to-be-processed point cloud points in the to-be-processed point cloud set, a rising edge time difference of the electromagnetic signal corresponding to the to-be-processed point cloud point, and obtain a rising edge time correction value of each of the to-be-processed point cloud points according to the rising edge time difference of the electromagnetic signal; wherein the rising edge time difference is the time difference between the moment when the signal strength of the electromagnetic signal corresponding to the to-be-processed point cloud point first reaches the second strength and the moment when the signal strength first reaches the third strength, and the third strength is greater than the second strength; The point cloud denoising module is used to calculate the difference between the pulse width time correction value and the rising edge time correction value of each of the point cloud points to be processed; if the difference between the pulse width time correction value and the rising edge time correction value of a certain point cloud point to be processed is greater than a preset threshold, the certain point cloud point to be processed is removed from the original point cloud set as an abnormal point.

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, the steps of the method according to any one of claims 1 to 7 are implemented.

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

Images