Point cloud denoising method and device suitable for closed environment

Abstract

The invention provides a point cloud denoising method and device suitable for a closed environment, and the method comprises the steps: obtaining a first frame of point cloud data and a second frame of point cloud data which are successively collected by a radar; for the first frame of point cloud data and the second frame of point cloud data, dividing the point cloud center into a plurality of sectors according to a preset degree; performing clustering processing on the point cloud data in each sector in the plurality of sectors to obtain a plurality of point cloud clusters of the first frame of point cloud data and a plurality of point cloud clusters of the second frame of point cloud data; aiming at the point number of each point cloud cluster, calculating the point cloud attribute of each point cloud cluster; and based on the point cloud clusters and the point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, determining whether each point cloud cluster is noise or not, and processing the point cloud clusters which are determined to be noise. The point cloud denoising precision is improved, and the point cloud denoising time is shortened.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a point cloud denoising method, device, electronic device and storage medium suitable for a closed environment. Background Art

[0002] In a closed environment, such as a mine environment, there are tiny objects such as dust and smoke. When collecting data through radar, the noise caused by these tiny objects such as dust and smoke will affect the quality and effectiveness of the collected data.

[0003] In the prior art, there are two main methods for removing noise data caused by noise in radar point clouds. One method is to project the point cloud into a depth map, and then determine whether there is a noise point based on the centroid position to remove the point cloud noise; the other is to upsample and downsample the point cloud to obtain shape features and then remove the point cloud noise. The problem with the first method is that the detection accuracy is not high, and the problem with the second method is that it takes a long time. Summary of the invention

[0004] The present application aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the first objective of the present application is to propose a point cloud denoising method suitable for closed environments to improve the accuracy of point cloud denoising and reduce the time of point cloud denoising.

[0006] The second objective of the present application is to propose a point cloud denoising device suitable for closed environments.

[0007] The third objective of the present application is to provide an electronic device.

[0008] A fourth objective of the present application is to provide a computer-readable storage medium.

[0009] A fifth object of the present application is to provide a computer program product.

[0010] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a point cloud denoising method suitable for a closed environment, comprising:

[0011] Obtain the first frame of point cloud data and the second frame of point cloud data collected by the radar;

[0012] For the first frame of point cloud data and the second frame of point cloud data, dividing the point cloud center into a plurality of sectors according to a preset degree; and performing clustering processing on the point cloud data in each of the plurality of sectors to obtain a plurality of point cloud clusters of the first frame of point cloud data and a plurality of point cloud clusters of the second frame of point cloud data;

[0013] According to the number of points in each point cloud cluster, calculating the point cloud attributes of each point cloud cluster;

[0014] Based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, it is determined whether each of the point cloud clusters is noise, and the point cloud clusters determined to be noise are processed.

[0015] In some implementations, the step of dividing the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data includes:

[0016] For the first frame of point cloud data, the center of the point cloud is divided into multiple sectors according to the size of 10 degrees horizontally and 5 degrees vertically;

[0017] For the second frame of point cloud data, the center of the point cloud is divided into multiple sectors with a size of 10 degrees horizontally and 5 degrees vertically.

[0018] In some implementations, the calculating the point cloud attributes of each point cloud cluster according to the number of points of each point cloud cluster includes:

[0019] When the number of points in the point cloud cluster is 2, the distance between two points in the point cloud cluster and the average distance of all points are calculated;

[0020] When the number of points in the point cloud cluster is 3, the area of ​​the triangle formed by the three points of the point cloud cluster and the average distance of all points are calculated;

[0021] When the number of points in the point cloud cluster is greater than 3, the unit volume of the three-dimensional region composed of all the points in the point cloud cluster and the average distance of all the points are calculated.

[0022] In some implementations, calculating the unit volume of a three-dimensional region composed of all points of the point cloud cluster includes:

[0023] The volume of the three-dimensional convex hull composed of all points in the point cloud cluster is obtained by the Quickhull algorithm;

[0024] The volume of the three-dimensional convex hull is divided by the number of points in the point cloud cluster to obtain the unit volume of the point cloud cluster.

[0025] In some implementations, determining whether each point cloud cluster is noise based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, and processing the point cloud clusters determined to be noise includes:

[0026] Performing the following discrimination processing on the point cloud clusters in the second frame of point cloud data one by one until all the point cloud clusters in the second frame of point cloud data are processed;

[0027] Acquire the number of points of the current point cloud cluster in the second frame of point cloud data, and refer to the point cloud cluster in the first frame of point cloud data corresponding to the spatial position of the current point cloud cluster as the first point cloud cluster;

[0028] If the number of points in the first point cloud cluster is 0, the current point cloud cluster is determined as noise and a point cloud cluster deletion operation is performed;

[0029] If the number of points in the current point cloud cluster is 1, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0030] If the number of points in the current point cloud cluster is 2, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0031] If the number of points in the current point cloud cluster is 2, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0032] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0033] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is 3, the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, and the absolute value of the difference between the area of ​​a triangle of the first point cloud cluster and the area of ​​a triangle of the current point cloud cluster is greater than a second threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0034] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is less than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0035] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the unit volume of the first point cloud cluster and the unit volume of the current point cloud cluster is greater than a third threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed;

[0036] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is less than or equal to 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed.

[0037] In some implementations, before dividing the point cloud center into a plurality of sectors according to a preset degree, the method includes:

[0038] Points in the first frame of point cloud data and the second frame of point cloud data that correspond to points outside the radar operating range are removed.

[0039] To achieve the above-mentioned purpose, the second aspect of the present application proposes a point cloud denoising device suitable for a closed environment, comprising:

[0040] A data acquisition module, used to acquire the first frame of point cloud data and the second frame of point cloud data collected by the radar;

[0041] A data processing module, for dividing the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data; and performing clustering processing on the point cloud data in each of the plurality of sectors to obtain a plurality of point cloud clusters of the first frame of point cloud data and a plurality of point cloud clusters of the second frame of point cloud data;

[0042] A feature acquisition module, used for calculating the point cloud attributes of each point cloud cluster according to the number of points in each point cloud cluster;

[0043] The noise discrimination module is used to determine whether each point cloud cluster is noise based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame point cloud data and the first frame point cloud data, and process the point cloud clusters determined to be noise.

[0044] In some implementations, the data processing module is used to divide the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data:

[0045] For the first frame of point cloud data, the center of the point cloud is divided into multiple sectors according to the size of 10 degrees horizontally and 5 degrees vertically;

[0046] For the second frame of point cloud data, the center of the point cloud is divided into multiple sectors with a size of 10 degrees horizontally and 5 degrees vertically.

[0047] In some implementations, when calculating the point cloud attributes of each point cloud cluster according to the number of points in each point cloud cluster, the feature acquisition module is used to:

[0048] When the number of points in the point cloud cluster is 2, the distance between two points in the point cloud cluster and the average distance of all points are calculated;

[0049] When the number of points in the point cloud cluster is 3, the area of ​​the triangle formed by the three points of the point cloud cluster and the average distance of all points are calculated;

[0050] When the number of points in the point cloud cluster is greater than 3, the unit volume of the three-dimensional region composed of all the points in the point cloud cluster and the average distance of all the points are calculated.

[0051] In some implementations, the feature acquisition module, when calculating the unit volume of the three-dimensional area composed of all points of the point cloud cluster, is used to:

[0052] The volume of the three-dimensional convex hull composed of all points in the point cloud cluster is obtained by the Quickhull algorithm;

[0053] The volume of the three-dimensional convex hull is divided by the number of points in the point cloud cluster to obtain the unit volume of the point cloud cluster.

[0054] In some implementations, the noise discrimination module is specifically configured to:

[0055] Performing the following discrimination processing on the point cloud clusters in the second frame of point cloud data one by one until all the point cloud clusters in the second frame of point cloud data are processed;

[0056] Acquire the number of points of the current point cloud cluster in the second frame of point cloud data, and refer to the point cloud cluster in the first frame of point cloud data corresponding to the spatial position of the current point cloud cluster as the first point cloud cluster;

[0057] If the number of points in the first point cloud cluster is 0, the current point cloud cluster is determined as noise and a point cloud cluster deletion operation is performed;

[0058] If the number of points in the current point cloud cluster is 1, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0059] If the number of points in the current point cloud cluster is 2, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0060] If the number of points in the current point cloud cluster is 2, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed;

[0061] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0062] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is 3, the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, and the absolute value of the difference between the area of ​​a triangle of the first point cloud cluster and the area of ​​a triangle of the current point cloud cluster is greater than a second threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0063] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is less than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed;

[0064] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the unit volume of the first point cloud cluster and the unit volume of the current point cloud cluster is greater than a third threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed;

[0065] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is less than or equal to 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed.

[0066] In some implementations, before dividing the point cloud center into a plurality of sectors according to a preset degree, the data processing module is further configured to:

[0067] Points in the first frame of point cloud data and the second frame of point cloud data that correspond to points outside the radar operating range are removed.

[0068] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect.

[0069] To achieve the above-mentioned purpose, the fourth aspect of the present application proposes a computer-readable storage medium, in which computer-readable storage medium is stored computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0070] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0071] The point cloud denoising method, device, electronic device and storage medium suitable for closed environments provided in the present application identify and delete noise points in the point cloud data by determining the number of point cloud cluster points and point cloud attributes of the relative position of the two frames of point cloud data collected by the radar. Compared with the prior art, the accuracy of point cloud denoising is improved and the time of point cloud denoising is reduced. It can be used for noise detection of radar point clouds in closed environments, such as noise caused by fine objects such as dust and smoke, to effectively improve the validity of the data.

[0072] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0074] Figure 1 A schematic diagram of a flow chart of a point cloud denoising method applicable to a closed environment provided in an embodiment of the present application;

[0075] Figure 2 A block diagram of a point cloud denoising device suitable for a closed environment provided in an embodiment of the present application;

[0076] Figure 3 A block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0077] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0078] The following describes a point cloud denoising method, device and apparatus applicable to a closed environment according to an embodiment of the present application with reference to the accompanying drawings.

[0079] Figure 1A flowchart of a point cloud denoising method suitable for a closed environment provided in an embodiment of the present application.

[0080] It should be noted that the executor of the point cloud denoising method suitable for a closed environment of the embodiment of the present application is the point cloud denoising device suitable for a closed environment of the embodiment of the present application. The point cloud denoising device suitable for a closed environment can be configured in an electronic device so that the electronic device can perform the point cloud denoising function suitable for a closed environment.

[0081] like Figure 1 As shown, the point cloud denoising method suitable for closed environments includes the following steps:

[0082] Step S101, obtaining a first frame of point cloud data and a second frame of point cloud data collected successively by a radar.

[0083] It can be understood that the point cloud denoising method of the present application identifies noise points in the point cloud data based on changes in different frames of point cloud data collected. Therefore, it is necessary to obtain two frames of point cloud data collected before and after the radar.

[0084] Step S102, for the first frame point cloud data and the second frame point cloud data, divide the point cloud center into multiple sectors according to a preset degree; and cluster the point cloud data in each of the multiple sectors to obtain multiple point cloud clusters of the first frame point cloud data and multiple point cloud clusters of the second frame point cloud data.

[0085] In some embodiments, before dividing the point cloud center into a plurality of sectors according to a preset degree, the method includes: removing points in the first frame of point cloud data and the second frame of point cloud data that correspond to points outside the radar working range.

[0086] For example, the radar is installed on the equipment, and the distance between the radar and the target to be detected is greater than 0.5. All data points collected within 0.5m from the radar in the collected point cloud data are deleted, because these points are outside the working range of the radar and are definitely noise.

[0087] As an implementation method, for the first frame point cloud data and the second frame point cloud data, the point cloud center is divided into multiple sectors according to a preset degree; including: for the first frame point cloud data, the point cloud center is divided into multiple sectors according to a size of 10 degrees horizontally and 5 degrees vertically; for the second frame point cloud data, the point cloud center is divided into multiple sectors according to a size of 10 degrees horizontally and 5 degrees vertically.

[0088] It should be noted that in order to ensure that the number of point cloud clusters is within an appropriate range, the resolution of the sector segmentation can be adjusted according to the specific radar.

[0089] Exemplarily, the Kmeans clustering algorithm is selected for clustering processing.

[0090] It can be understood that the first frame point cloud data and the second frame point cloud data are processed in the same way, each frame point cloud data is divided into a number of sectors, and then the point cloud data falling in the sectors are clustered into a number of point cloud clusters.

[0091] Step S103, calculating the point cloud attributes of each point cloud cluster according to the number of points in each point cloud cluster.

[0092] As an implementation method, the point cloud attributes of each point cloud cluster are calculated according to the number of points in each point cloud cluster, including: when the number of points in the point cloud cluster is 2, the distance between two points of the point cloud cluster and the average distance of all points are calculated; when the number of points in the point cloud cluster is 3, the area of ​​the triangle formed by the three points of the point cloud cluster and the average distance of all points are calculated; when the number of points in the point cloud cluster is greater than 3, the unit volume of the three-dimensional area formed by all points of the point cloud cluster and the average distance of all points are calculated.

[0093] As an implementation method, a method for calculating the unit volume of a three-dimensional area composed of all points of the point cloud cluster includes: obtaining the volume of a three-dimensional convex hull composed of all points of the point cloud cluster through a Quickhull algorithm; dividing the volume of the three-dimensional convex hull by the number of points of the point cloud cluster to obtain the unit volume of the point cloud cluster.

[0094] Exemplarily, when calculating point cloud attributes, each point cloud cluster is processed as follows:

[0095] (1) If the point cloud cluster has only one point, no processing is performed;

[0096] (2) If the point cloud cluster has only two points, then calculate the distance between the two points and the average distance.

[0097] (3) If the point cloud cluster has only three points, calculate the area and average distance of the triangle formed by these three points.

[0098] (4) If the point cloud cluster has more than three points, first calculate the volume of the three-dimensional region composed of these points. You can use the points in the point cloud to construct a three-dimensional convex hull, and then divide it by the number of points in the point cloud in this area, and record it as the unit volume of the point cloud cluster.

[0099] Step S104, based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, determine whether each point cloud cluster is noise, and process the point cloud cluster determined to be noise.

[0100] As an implementation method, based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, a method for determining whether each point cloud cluster is noise and processing the point cloud cluster determined to be noise is provided; comprising:

[0101] The point cloud clusters in the second frame of point cloud data are subjected to the following discrimination processing one by one until all the point cloud clusters in the second frame of point cloud data are processed;

[0102] The number of points of the current point cloud cluster in the second frame of point cloud data is obtained, and the point cloud cluster corresponding to the spatial position of the current point cloud cluster in the first frame of point cloud data is called the first point cloud cluster;

[0103] If the number of points in the first point cloud cluster is 0, the current point cloud cluster is determined as noise and the point cloud cluster is deleted;

[0104] If the number of points in the current point cloud cluster is 1, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined as noise and the point cloud cluster is deleted;

[0105] If the number of points in the current point cloud cluster is 2, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined as noise and the point cloud cluster is deleted;

[0106] If the number of points in the current point cloud cluster is 2, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than the first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster is deleted;

[0107] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than the first threshold, then both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster deletion operation is performed;

[0108] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is 3, the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, and the absolute value of the difference between the area of ​​the triangle of the first point cloud cluster and the area of ​​the triangle of the current point cloud cluster is greater than a second threshold, then both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster deletion operation is performed;

[0109] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is less than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster is deleted;

[0110] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the unit volume of the first point cloud cluster and the unit volume of the current point cloud cluster is greater than a third threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed;

[0111] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is less than or equal to 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed.

[0112] Exemplarily, noise discrimination is performed on each point cloud cluster in the second frame of point cloud data one by one, and the processing is performed, and the discrimination is as follows:

[0113] For the current point cloud cluster currently identified, find the point cloud cluster at the corresponding position in the first frame of point cloud data (hereinafter referred to as the first frame) according to the spatial position relationship. If the corresponding point cloud cluster cannot be found, the current point cloud cluster is considered to be noise, and the current point cloud cluster is removed, and the next point cloud cluster identification is started directly, otherwise continue to the next step;

[0114] If the number of points in the current point cloud cluster is 1, and the number of points in the corresponding point cloud cluster of the first frame exceeds 3, it means that the current point cloud cluster and the point cloud cluster corresponding to the first frame are both noise and should be deleted; if the number of points in the corresponding point cloud cluster of the first frame does not exceed two, it will be retained.

[0115] If the number of points in the current point cloud cluster is 2, and the number of points in the corresponding point cloud cluster of the first frame exceeds three, both are deleted; otherwise, if the average distances between the two are significantly different, both are deleted.

[0116] If the number of points in the current point cloud cluster is 3, and the number of points in the corresponding first frame point cloud cluster is greater than three, the change in average distance is calculated. If the change is large, all are deleted. If the number of points in the corresponding point cloud cluster is also three, the difference between the triangle area and the average distance is calculated. If the difference is large, all are deleted; if the number of points in the corresponding first frame is less than three, the change in average distance is calculated. If the change is large, all are deleted;

[0117] If the number of points in the current point cloud cluster is greater than 3 and the number of points in the point cloud cluster corresponding to the first frame is greater than 3, the difference in unit volume is compared. If the unit volume changes greatly, the current point cloud cluster is deleted; if the number of points in the point cloud cluster corresponding to the first frame is less than or equal to three, the average distance is compared. If the average distance changes greatly, the current point cloud cluster is deleted.

[0118] The point cloud denoising method suitable for a closed environment in the embodiment of the present application identifies and deletes noise points in the point cloud data by determining the number of point cloud cluster points and point cloud attributes of the two frames of point cloud data collected by the radar. Compared with the prior art, the accuracy of point cloud denoising is improved and the time of point cloud denoising is reduced. It can be used for noise detection of radar point clouds in closed environments, such as noise caused by fine objects such as dust and smoke, to effectively improve the validity of the data.

[0119] In order to implement the above embodiment, the present application also proposes a point cloud denoising device suitable for a closed environment. Figure 2 A block diagram of a point cloud denoising device suitable for a closed environment provided in an embodiment of the present application. Figure 2 As shown, the point cloud denoising device suitable for a closed environment may include: a data acquisition module 201 , a data processing module 202 , a feature acquisition module 203 and a noise discrimination module 204 .

[0120] The data acquisition module 201 is used to acquire the first frame of point cloud data and the second frame of point cloud data collected by the radar;

[0121] The data processing module 202 is used to divide the point cloud center into a plurality of sectors according to a preset degree for the first frame point cloud data and the second frame point cloud data; and perform clustering processing on the point cloud data in each sector of the plurality of sectors to obtain a plurality of point cloud clusters of the first frame point cloud data and a plurality of point cloud clusters of the second frame point cloud data;

[0122] The feature acquisition module 203 is used to calculate the point cloud attributes of each point cloud cluster according to the number of points in each point cloud cluster;

[0123] The noise identification module 204 is used to determine whether each point cloud cluster is noise based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame point cloud data and the first frame point cloud data, and process the point cloud clusters determined to be noise.

[0124] Further, in a possible implementation of the embodiment of the present application, the data processing module 202 divides the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data;

[0125] For the first frame of point cloud data, the point cloud center is divided into multiple sectors with a size of 10 degrees horizontally and 5 degrees vertically;

[0126] For the second frame of point cloud data, the point cloud center is divided into multiple sectors with a size of 10 degrees horizontally and 5 degrees vertically.

[0127] Furthermore, in a possible implementation of the embodiment of the present application, the feature acquisition module 203 is used to:

[0128] When the number of points in a point cloud cluster is 2, calculate the distance between two points in the point cloud cluster and the average distance of all points;

[0129] When the number of points in the point cloud cluster is 3, calculate the area of ​​the triangle formed by the three points of the point cloud cluster and the average distance of all points;

[0130] When the number of points in a point cloud cluster is greater than 3, the unit volume of the three-dimensional area composed of all the points in the point cloud cluster and the average distance of all the points are calculated.

[0131] Furthermore, in a possible implementation of the embodiment of the present application, the feature acquisition module 203 is used to calculate the unit volume of the three-dimensional area composed of all points of the point cloud cluster:

[0132] The volume of the three-dimensional convex hull composed of all points in the point cloud cluster is obtained by the Quickhull algorithm;

[0133] The volume of the three-dimensional convex hull is divided by the number of points in the point cloud cluster to obtain the unit volume of the point cloud cluster.

[0134] Furthermore, in a possible implementation of the embodiment of the present application, the noise determination module 204 is specifically configured to:

[0135] The point cloud clusters in the second frame of point cloud data are subjected to the following discrimination processing one by one until all the point cloud clusters in the second frame of point cloud data are processed;

[0136] The number of points of the current point cloud cluster in the second frame of point cloud data is obtained, and the point cloud cluster corresponding to the spatial position of the current point cloud cluster in the first frame of point cloud data is called the first point cloud cluster;

[0137] If the number of points in the first point cloud cluster is 0, the current point cloud cluster is determined as noise and the point cloud cluster is deleted;

[0138] If the number of points in the current point cloud cluster is 1, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined as noise and the point cloud cluster is deleted;

[0139] If the number of points in the current point cloud cluster is 2, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined as noise and the point cloud cluster is deleted;

[0140] If the number of points in the current point cloud cluster is 2, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than the first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster is deleted;

[0141] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than the first threshold, then both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster deletion operation is performed;

[0142] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is 3, the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, and the absolute value of the difference between the area of ​​the triangle of the first point cloud cluster and the area of ​​the triangle of the current point cloud cluster is greater than a second threshold, then both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster deletion operation is performed;

[0143] If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is less than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and the point cloud cluster is deleted;

[0144] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the unit volume of the first point cloud cluster and the unit volume of the current point cloud cluster is greater than a third threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed;

[0145] If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is less than or equal to 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed.

[0146] Furthermore, in a possible implementation of the embodiment of the present application, before dividing the point cloud center into a plurality of sectors according to a preset degree, the data processing module 202 is further configured to:

[0147] The points outside the radar working range in the first frame of point cloud data and the second frame of point cloud data are removed.

[0148] It should be noted that the aforementioned explanation of the embodiment of the point cloud denoising method applicable to a closed environment is also applicable to the point cloud denoising device applicable to a closed environment of this embodiment, and will not be repeated here.

[0149] In order to implement the above embodiment, the present application also proposes an electronic device. Figure 3 , Figure 3 is a block diagram of an electronic device provided in an embodiment of the present application. Figure 3 As shown, the electronic device 300 includes: a processor 301, and a memory 302 communicatively connected to the processor 301; the memory 302 stores computer-executable instructions; the processor 301 executes the computer-executable instructions stored in the memory to implement the method provided in the aforementioned embodiment.

[0150] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0151] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0152] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0153] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0154] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0155] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0156] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0157] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0158] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0159] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0160] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A point cloud denoising method suitable for a closed environment, characterized in that: The following steps are involved: Obtain the first frame of point cloud data and the second frame of point cloud data collected by the radar in sequence; For the first frame of point cloud data and the second frame of point cloud data, dividing the point cloud center into a plurality of sectors according to a preset degree; and performing clustering processing on the point cloud data in each of the plurality of sectors to obtain a plurality of point cloud clusters of the first frame of point cloud data and a plurality of point cloud clusters of the second frame of point cloud data; According to the number of points in each point cloud cluster, calculating the point cloud attributes of each point cloud cluster; Based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, it is determined whether each of the point cloud clusters is noise, and the point cloud clusters determined to be noise are processed.

2. The method according to claim 1, characterized in that The method of dividing the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data includes: For the first frame of point cloud data, the center of the point cloud is divided into multiple sectors according to the size of 10 degrees horizontally and 5 degrees vertically; For the second frame of point cloud data, the center of the point cloud is divided into multiple sectors with a size of 10 degrees horizontally and 5 degrees vertically.

3. The method according to claim 1, characterized in that The step of calculating the point cloud attributes of each point cloud cluster according to the number of points of each point cloud cluster comprises: When the number of points in the point cloud cluster is 2, the distance between two points in the point cloud cluster and the average distance of all points are calculated; When the number of points in the point cloud cluster is 3, the area of ​​the triangle formed by the three points of the point cloud cluster and the average distance of all points are calculated; When the number of points in the point cloud cluster is greater than 3, the unit volume of the three-dimensional region composed of all the points in the point cloud cluster and the average distance of all the points are calculated.

4. The method according to claim 3, characterized in that The step of calculating the unit volume of the three-dimensional region composed of all points of the point cloud cluster includes: The volume of the three-dimensional convex hull composed of all points in the point cloud cluster is obtained by the Quickhull algorithm; The volume of the three-dimensional convex hull is divided by the number of points in the point cloud cluster to obtain the unit volume of the point cloud cluster.

5. The method according to claim 3, characterized in that: The method of determining whether each point cloud cluster is noise based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame of point cloud data and the first frame of point cloud data, and processing the point cloud clusters determined to be noise, comprises: Performing the following discrimination processing on the point cloud clusters in the second frame of point cloud data one by one until all the point cloud clusters in the second frame of point cloud data are processed; Acquire the number of points of the current point cloud cluster in the second frame of point cloud data, and refer to the point cloud cluster in the first frame of point cloud data corresponding to the spatial position of the current point cloud cluster as the first point cloud cluster; If the number of points in the first point cloud cluster is 0, the current point cloud cluster is determined as noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 1, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 2, and the number of points in the first point cloud cluster is greater than 3, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 2, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, both the current point cloud cluster and the first point cloud cluster are determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is 3, the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, and the absolute value of the difference between the area of ​​a triangle of the first point cloud cluster and the area of ​​a triangle of the current point cloud cluster is greater than a second threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is 3, the number of points in the first point cloud cluster is less than 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster and the first point cloud cluster are both determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is greater than 3, and the absolute value of the difference between the unit volume of the first point cloud cluster and the unit volume of the current point cloud cluster is greater than a third threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed; If the number of points in the current point cloud cluster is greater than 3, the number of points in the first point cloud cluster is less than or equal to 3, and the absolute value of the difference between the average distance of the first point cloud cluster and the average distance of the current point cloud cluster is greater than a first threshold, the current point cloud cluster is determined to be noise and a point cloud cluster deletion operation is performed.

6. The method according to claim 1, characterized in that Before dividing the point cloud center into a plurality of sectors according to a preset degree, the method includes: Points in the first frame of point cloud data and the second frame of point cloud data that correspond to points outside the radar operating range are removed.

7. A point cloud denoising device suitable for a closed environment, characterized in that: include: A data acquisition module, used to acquire the first frame of point cloud data and the second frame of point cloud data collected by the radar; A data processing module, for dividing the point cloud center into a plurality of sectors according to a preset degree for the first frame of point cloud data and the second frame of point cloud data; and performing clustering processing on the point cloud data in each of the plurality of sectors to obtain a plurality of point cloud clusters of the first frame of point cloud data and a plurality of point cloud clusters of the second frame of point cloud data; A feature acquisition module, used for calculating the point cloud attributes of each point cloud cluster according to the number of points in each point cloud cluster; The noise discrimination module is used to determine whether each point cloud cluster is noise based on the point cloud clusters and point cloud attributes corresponding to the spatial positions in the second frame point cloud data and the first frame point cloud data, and process the point cloud clusters determined to be noise.

8. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when being executed by a processor.

Images