CN116087909A - Radar point cloud data downsampling processing method and related equipment - Google Patents
Radar point cloud data downsampling processing method and related equipment Download PDFInfo
- Publication number
- CN116087909A CN116087909A CN202211456375.XA CN202211456375A CN116087909A CN 116087909 A CN116087909 A CN 116087909A CN 202211456375 A CN202211456375 A CN 202211456375A CN 116087909 A CN116087909 A CN 116087909A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- sampling
- downsampling
- coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a radar point cloud data downsampling processing method and related equipment. The method comprises the following steps: acquiring down-sampling boundary area point cloud data of a target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient; and carrying out height segmentation on the down-sampling boundary region point cloud data according to a first preset height threshold value to obtain first region point cloud data and second region point cloud data, and carrying out target identification on the first region point cloud data and the second region point cloud data to obtain target object information. According to the radar point cloud data downsampling processing method, downsampling operations are carried out in the point cloud space by using different scales, so that the number of points in the point cloud can be effectively reduced to ensure the efficiency of subsequent processing, and the point cloud characteristics can be maintained to ensure the accuracy of subsequent clustering and other operations.
Description
Technical Field
The present disclosure relates to the field of signal processing, and more particularly, to a radar point cloud data downsampling method and related devices.
Background
The number of original point clouds of the laser radar is huge, and in order to improve the processing efficiency, downsampling (also called downsampling) is often performed on the original point clouds to reduce the amount of point cloud data, thereby improving the efficiency of subsequent point cloud processing. Common point cloud downsampling methods include uniform downsampling, random downsampling, voxel downsampling, and the like. While the various methods implement different principles and effects, the processing of point cloud space is equal. For example, voxel downsampling uniformly partitions the point cloud space into tightly arranged grids, and downsamples and outputs the center of gravity of the point cloud in each grid. Voxel downsampling, in contrast to the way the geometric center of the grid is taken as the downsampling output, is performed according to the local spatial distribution of the point cloud, but the processing of the point cloud space during the overall grid generation process is uniform, i.e. the grid size and spatial distribution are uniform.
This uniform spatial processing scheme results in a single downsampling scale that is difficult to determine: if the grid scale is too small, the number of points left after downsampling is large, and the downsampling effect is not obvious; if the grid scale is too large, point cloud features are easy to filter, processing results such as subsequent spatial clustering are affected, and the risk of target missed detection is increased. This problem is particularly pronounced for large spatial ranges and small clustered objects.
Disclosure of Invention
In the summary, a series of concepts in a simplified form are introduced, which will be further described in detail in the detailed description. The summary of the invention is not intended to define the key features and essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.
In order to provide down-sampling point cloud data more meeting the identification requirement, in a first aspect, the invention provides a method for down-sampling radar point cloud data, which comprises the following steps:
acquiring down-sampling boundary area point cloud data of a target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient;
the method comprises the steps of carrying out height segmentation on point cloud data of a down-sampling boundary area according to a first preset height threshold value to obtain first area point cloud data and second area point cloud data, wherein the first area point cloud data is point cloud data of an area with the height lower than or equal to the first preset height threshold value, a first down-sampling coefficient is correspondingly arranged in the first area, the second area point cloud data is point cloud data with the height higher than the first preset height threshold value, a second down-sampling coefficient is correspondingly arranged in the second area point cloud data, the perceived internal down-sampling coefficient is larger than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient;
and carrying out target identification on the first regional point cloud data and the second regional point cloud data to acquire target object information.
Optionally, the target object information includes object type information of a target object;
the method further comprises the following steps:
and determining a third downsampling coefficient of a corresponding region of the target object based on the object type information.
Optionally, the determining the third downsampling coefficient of the corresponding region of the target object based on the object type information includes:
setting a third downsampling coefficient corresponding to the target object as the first downsampling coefficient when the height of the target object is smaller than or equal to the first preset height threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as the perceived internal downsampling coefficient when the height of the target object is greater than the first preset height threshold and the length is greater than the length threshold or the height of the target object is greater than the first preset height threshold or the width is greater than the width threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in a downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the length is smaller than or equal to the length threshold;
and/or the number of the groups of groups,
and setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in the downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the width is smaller than the width threshold.
Optionally, the method further comprises:
when there is an overlapping region in the corresponding regions of the plurality of target objects, a downsampling process is performed on the overlapping region by selecting a smaller downsampling coefficient in the corresponding region of the overlapping target object.
Optionally, the method further comprises:
under the condition that a smaller target object is overlapped behind a larger target object, acquiring the movement rule of the smaller target object;
and carrying out downsampling processing in the overlapped area based on the movement rule and the smaller downsampling coefficient.
Optionally, the method further comprises:
acquiring the moving speed of the moving object when the target object is the moving object;
and determining a fourth downsampling coefficient of the area corresponding to the moving object based on the moving speed.
Optionally, the method further comprises:
acquiring the moving direction of the moving object when the target object is the moving object;
the perceived internal downsampling coefficient is set to a downsampling coefficient of the moving object corresponding region when the moving direction of the object does not overlap with the moving direction of the target vehicle.
In a second aspect, the present invention further provides a radar point cloud data downsampling processing device, including:
the acquisition unit is used for acquiring the down-sampling boundary area point cloud data of the target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on the original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient;
the segmentation unit is used for carrying out height segmentation on the point cloud data of the down-sampling boundary region according to a first preset height threshold value according to a first down-sampling coefficient and a second down-sampling coefficient to obtain first region point cloud data and second region point cloud data, wherein the first region point cloud data is point cloud data of which the height is lower than or equal to that of a region corresponding to the first preset height threshold value, the second region point cloud data is point cloud data of which the height is higher than that of the region corresponding to the first preset height threshold value, the perceived internal down-sampling coefficient is larger than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient;
and the identification unit is used for carrying out target identification on the first regional point cloud data and the second regional point cloud data so as to acquire target object information.
In a third aspect, an electronic device, comprising: the method comprises the steps of a radar point cloud data downsampling processing method according to any one of the first aspect, a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor is configured to execute the computer program stored in the memory.
In a fourth aspect, the present invention also proposes a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the radar point cloud data downsampling processing method of any one of the first aspects.
In summary, the radar point cloud data downsampling processing method of the embodiment of the application includes: acquiring down-sampling boundary area point cloud data of a target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient; according to a first preset height threshold, carrying out height segmentation on point cloud data of a down-sampling boundary region according to a first down-sampling coefficient and a second down-sampling coefficient to obtain first region point cloud data and second region point cloud data, wherein the first region point cloud data is point cloud data of which the height is lower than or equal to that of a region corresponding to the first preset height threshold, the second region point cloud data is point cloud data of which the height is higher than that of the region corresponding to the first preset height threshold, the perceived internal down-sampling coefficient is larger than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient; and carrying out target identification on the first area point cloud data and the second area point cloud data to acquire target object information. According to the radar point cloud data downsampling processing method, the original point cloud data are subjected to boundary segmentation based on the preset downsampling boundary width and the downsampling boundary downsampling coefficient, primary downsampling processing is carried out to obtain the point cloud data of the downsampling boundary area, secondary downsampling is carried out on the point cloud data of the downsampling boundary area according to the first downsampling coefficient and the second downsampling according to the first preset height threshold, and downsampling operation is carried out in the point cloud space by using different scales, so that the number of points in the point cloud can be effectively reduced to ensure the efficiency of subsequent processing, and point cloud characteristics can be maintained to ensure the accuracy of subsequent clustering and other operations.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the specification. Also, like reference numerals are used to designate like parts throughout the figures. In the drawings:
fig. 1 is a schematic flow chart of a radar point cloud data downsampling processing method according to an embodiment of the present application;
fig. 2 is a schematic range diagram of radar point cloud data acquired by a vehicle according to an embodiment of the present application;
fig. 3 is a schematic diagram of a positional relationship between a vehicle and a target object according to an embodiment of the present application;
fig. 4 is a schematic flow chart of another radar point cloud data downsampling processing method according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a radar point cloud data downsampling processing device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device for radar point cloud data downsampling processing according to an embodiment of the present application.
Detailed Description
According to the radar point cloud data downsampling processing method, the original point cloud data is divided into the boundary outer area, the downsampling boundary area and the inner area based on the preset downsampling boundary and the boundary width, the downsampling boundary point cloud data is further divided into the boundary low-level area and the boundary non-low-level area according to the first preset height threshold value, downsampling operation is carried out in different separated point cloud spaces by using different scales, the number of point cloud points can be effectively reduced to ensure the efficiency of subsequent processing, and the accuracy of the point cloud characteristics to ensure the operation such as subsequent clustering can be maintained.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims of this application and in the above-described figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application.
It should be noted that, for convenience of description, it is assumed that the downsampling coefficient and the downsampling rate are positively correlated, that is, the larger the downsampling coefficient value is, the larger the downsampling rate is, the more points are filtered out, and the fewer points are reserved; the smaller the downsampling coefficient value, the smaller the downsampling rate, the fewer points are filtered out, and the more points remain. This assumption is not a limitation on the technology. Setting a down-sampling boundary, wherein the point clouds outside the down-sampling boundary are sparse due to distance from the far point cloud, and do not perform down-sampling operation; points within the downsampled boundary perform a downsampling operation.
Referring to fig. 1, a schematic flow chart of a radar point cloud data downsampling processing method provided in an embodiment of the present application may specifically include:
s110, acquiring down-sampling boundary area point cloud data of a target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on original point cloud data acquired by the target vehicle based on a preset down-sampling boundary and a boundary width;
by way of example, the target vehicle can acquire original point cloud data around the driving road through a radar sensor mounted on the vehicle, the original point cloud data has huge data volume, and the original point cloud data can be used for obstacle recognition only by carrying out downsampling processing on the original point cloud data, and the method firstly carries out downsampling on the original point cloud data according to a preset downsampling boundary and a boundary width delta dPerforming boundary segmentation processing, as shown in fig. 2, to obtain a schematic view of the range of the point cloud data area acquired by a plurality of radar sensors of the vehicle, performing boundary segmentation on the original point cloud data by presetting a downsampling boundary and a boundary width Δd, wherein the area with the distance smaller than or equal to Δd from the downsampling boundary is called a boundary area, and the point cloud in the boundary area is denoted as P 1 The internal point cloud excluding the boundary region is denoted as P 2 And the external point cloud outside the boundary area is sparse due to the long distance, and downsampling is not performed. The boundary downsampling factor is smaller than the internal downsampling factor, i.e. the data density of the downsampled boundary point cloud data is greater than the data density of the internal perception point cloud. The downsampling boundary is a target area cut out according to the vehicle type, road, project requirement, etc. in the sensing range of the sensor. In order to timely sense the obstacle entering the sensing area, the point cloud space near the boundary of the sensing area is used as a key sensing area and a smaller sampling grid is arranged. Let the boundary sensing region distance threshold be Δd (Δd=n·a may be set) 0 N is a positive integer, a 0 For the initial downsampling factor), the area inward from the downsampling boundary by a distance Δd is the boundary area.
S120, carrying out height segmentation on the point cloud data of the down-sampling boundary region according to a first preset height threshold value to obtain first region point cloud data and second region point cloud data, wherein the first region point cloud data is point cloud data of which the height is lower than or equal to that of a region corresponding to the first preset height threshold value, the first region is correspondingly provided with a first down-sampling coefficient, the second region point cloud data is point cloud data of which the height is higher than that of the region corresponding to the first preset height threshold value, the second region point cloud data is correspondingly provided with a second down-sampling coefficient, the perceived internal down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient;
for example, there may be small objects such as cones, roadblocks, etc. with small heights in the down-sampled boundary region point cloud data, and there may be obstacles such as vehicles with large heights. Small objects such as cones, roadblocks and the like are generally shorter in heightAbove P 1 The point cloud space can be continuously segmented through the first preset height in the sub point cloud space. Selecting a first preset height threshold h 0 (can take the value h 0 =0.5m) and point clouds with a height larger than or equal to a first preset height form a sub-point cloud P 3 (second regional point cloud data), point clouds smaller than or equal to a first preset threshold value form a sub-point cloud P 4 (first regional point cloud data). P (P) 4 Mainly used for identifying short and short targets, the downsampling grid of the downsampling grid can be set to be smaller in size so as to reserve more points for subsequent clustering and other processes, and P is recorded 3 、P 4 Downsampling grid parameter in space is a 3 、a 4 Wherein n is 1 、n 2 Is a positive integer, n 1 For the second downsampling coefficient, n 2 For the first downsampling factor, the range of values may be defined as n 1 ∈[2,4]，n 2 >n 1 ，
For the second downsampling factor,/>
The first downsampling factor z is the height value of the point. a, a 0 To downsample the initial grid parameters for P 2 The downsampling of the sub-point cloud can be generally carried out at a value of between 0.3cm and 0.5 cm.
S130, carrying out target recognition on the first regional point cloud data and the second regional point cloud data to acquire target object information.
By way of example, the point cloud data for identifying the obstacle, which is acquired by the target vehicle at the current position, can be obtained by combining the first area point cloud data and the second area point cloud data, and the obstacle near the target vehicle can be quickly and accurately identified by identifying the point cloud data after the down-sampling processing through a clustering algorithm or a deep learning algorithm, so that the target vehicle is guided to perform intelligent driving.
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, the original point cloud data is subjected to point cloud boundary segmentation based on the preset downsampling boundary and the boundary width to obtain downsampling boundary area point cloud data, the downsampling boundary area point cloud data is segmented into a boundary low-low sub-area and a boundary non-low sub-area according to the first preset height threshold value, the first downsampling coefficient and the second downsampling coefficient are calculated through the preset default downsampling parameters, downsampling operations are performed in different point cloud spaces by using different scales, the number of point cloud points can be effectively reduced to ensure the efficiency of subsequent processing, and the point cloud characteristics can be maintained to ensure the accuracy of subsequent clustering and other operations.
In some examples, the target object information includes object type information of the target object;
the method further comprises the following steps:
and determining a third downsampling coefficient of a corresponding region of the target object based on the object type information.
By way of example, after the boundary segmentation and the height segmentation, the type information of the object can be obtained by identifying the first area point cloud data and the second area point cloud data, the object can be a small fixed object, a large fixed object, a small moving object, a large moving object and the like, and after the object is initially identified, the downsampling coefficient is adjusted according to the type of the target object, and the third downsampling coefficient of the corresponding area of different objects is redetermined so as to meet the identification requirement of different types of objects, and the identification speed is effectively improved.
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, after an object is initially identified, downsampling coefficients of the object can be further adjusted according to type information of the object, so that requirements of different types of object identification are met, and identification speed is effectively improved.
In some examples, determining the third downsampling factor of the corresponding region of the target object based on the object type information includes:
setting a third downsampling coefficient corresponding to the target object as the first downsampling coefficient when the height of the target object is smaller than or equal to the first preset height threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as the perceived internal downsampling coefficient when the height of the target object is greater than the first preset height threshold and the length is greater than the length threshold or the height of the target object is greater than the first preset height threshold or the width is greater than the width threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in a downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the length is smaller than or equal to the length threshold;
and/or the number of the groups of groups,
and setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in a downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the width is smaller than the width threshold.
Exemplary, as shown in FIG. 3, at P 2 Tracking and predicting the motion trail of a perceived target object through a history perceived result in the sub-point cloud space, carrying out space division on the possible position of the target object in the current frame, and carrying out sub-space P of the range of the target according to the size of the target 5 、P 6 、P 7 The sub-point cloud space of the area where the rest non-target is located is marked as P 8 The corresponding downsampling grid parameter is a 5 、a 6 、a 7 、a 8 。a 5 、a 6 、a 7 The calculation mode of (2) is related to the size and/or the height of the target object, and the following formula is adopted:
where l is the length of the target object, w is the width of the target object, and h is the height of the target objectHeight, h 0 、l 0 、w 0 The reference value of the height threshold value, the length threshold value and the width threshold value can be h 0 ＝0.5m,l 0 ＝3m,w 0 =2m. When h is less than or equal to h 0 When the target object is smaller than or equal to the first preset height threshold, the target object is a short target, and the third downsampling coefficient is set as the first downsampling coefficient; the target object is larger than the second preset height threshold value and l>l 0 Or w>w 0 If the target object is a medium-large target of the vehicle type, the default parameters of the source can be used; otherwise, the target object may be a pedestrian or a rider, and downsampling coefficients of the downsampled boundary non-low area point cloud may be used. P (P) 8 Instead of the key sense region, the downsampling may be performed using default downsampling parameters.
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, the third downsampling coefficient of the corresponding area is readjusted according to the size of the target object, the point cloud precision of different types of objects is subjected to targeted adjustment, and the recognition speed of the target object is improved while the recognition precision is ensured.
In some examples, the above method further comprises:
when there is an overlapping region in the corresponding regions of the plurality of target objects, a downsampling process is performed on the overlapping region by selecting a smaller downsampling coefficient in the corresponding region of the overlapping target object.
In an exemplary embodiment, when there is an overlap between the areas corresponding to the plurality of target objects, the overlap area is downsampled by using a smaller downsampling coefficient, so as to ensure that the point cloud density of the overlap area is sufficient to identify any one of the overlapping target objects, and avoid degradation of identification accuracy caused by overlapping of the areas.
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, under the condition that a plurality of target objects have overlapping areas, the overlapping areas are downsampled by adopting the smaller downsampling coefficient, so that the identification accuracy of the target objects can be ensured.
In some examples, the above method further comprises:
under the condition that a smaller target object is overlapped behind a larger target object, acquiring the movement rule of the smaller target object;
and carrying out downsampling processing in the overlapped area based on the movement rule and the smaller downsampling coefficient.
For example, when a smaller target object is overlapped behind a larger target object, a larger object may shade the smaller target object in a future period of time, and after a further lapse of a period of time, the smaller object suddenly appears in a region corresponding to the non-target object, at this time, the downsampling coefficient of the region corresponding to the non-target object is larger, the point cloud data is truly larger, the recognition of the smaller object is affected, and the speed and the precision are reduced.
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, when a smaller target object is overlapped behind a larger target object, a movement rule of the smaller target object is acquired, and processing can ensure that the downsampling coefficient of a region corresponding to the smaller object is enough to meet the requirement of identifying the object when the smaller object jumps out of shielding of the larger object, so that risk caused by missing identification is avoided.
In some examples, the above method further comprises:
acquiring the moving speed of the moving object when the target object is the moving object;
and determining a fourth downsampling coefficient of the area corresponding to the moving object based on the moving speed.
In an exemplary case that the target object is a moving object, the moving speed of the moving object is obtained, the fourth downsampling coefficient is determined according to the moving speed of the object, the fourth downsampling coefficient is smaller when the moving speed of the object is higher, the potential risk of the vehicle is greater when the moving speed of the target object is higher, the fourth downsampling coefficient is smaller when the vehicle speed is higher, namely, more point cloud information is reserved, so that the motion rule and the characteristics of the object are better and more accurately identified, and the safety of automatic driving of the vehicle is improved.
In summary, according to the radar point cloud data downsampling processing method, the downsampling coefficient is dynamically adjusted according to the moving speed of the target object, and more accurate identification can be effectively made for moving objects with different speeds.
In some examples, the above method further comprises:
acquiring the moving direction of the moving object when the target object is the moving object;
the perceived internal downsampling coefficient is set to a downsampling coefficient of the moving object corresponding region when the moving direction of the object does not overlap with the moving direction of the target vehicle.
For example, when the target object is identified as a moving object, the moving direction of the moving object may be obtained according to the radar point cloud data of different time periods identified by the target vehicle, and if the moving direction of the moving object does not overlap with the moving direction of the target vehicle, that is, the moving object does not collide with the target vehicle, a larger grid parameter is not required to identify the moving object, and at this time, the downsampling coefficient of the corresponding area of the moving object is set to a perceived internal downsampling coefficient (that is, a default downsampling coefficient).
In summary, according to the radar point cloud data downsampling processing method provided by the embodiment of the application, through judging the moving direction of a moving object, when the moving direction of the object cannot interfere with the running of a vehicle, grid parameters of the object are reduced, and the speed of identifying the obstacle is improved.
In some examples, as shown in fig. 4, the point cloud space may be divided into a plurality of subspaces p= { P1, P2, P3.}, the downsampling method f= { F1, F2, F3,.} and the parameter a= { a1, a2, a3,.} of each subspace are set according to the requirement, and finally, the downsampled point clouds are combined to obtain the final point cloud downsampling result. The space division and parameter formulation are adjusted according to different specific factors such as vehicle types, operation environments, recognition targets and the like, and the invention provides a plurality of specific implementation schemes. The general flow chart is shown in fig. 4: dividing an original point cloud into different sub point clouds by three methods of boundary division, height division and perception tracking division successively; setting respective downsampling method parameters for each sub-point cloud to downsample; and finally, merging the sub-point cloud downsampling results to obtain a final point cloud downsampling result.
Referring to fig. 5, an embodiment of a radar point cloud data downsampling device in an embodiment of the present application may include:
an obtaining unit 21, configured to obtain down-sampling boundary area point cloud data of a target vehicle, where the down-sampling boundary area point cloud data is obtained by performing boundary segmentation processing on original point cloud data obtained by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient, and a perceived internal down-sampling coefficient;
the segmentation unit 22 is configured to highly segment the down-sampling boundary region point cloud data according to a first preset height threshold value according to a first down-sampling coefficient and a second down-sampling coefficient to obtain first region point cloud data and second region point cloud data, where the first region point cloud data is point cloud data with a height lower than or equal to a region corresponding to the first preset height threshold value, the second region point cloud data is point cloud data with a height higher than the region corresponding to the first preset height threshold value, the perceived internal down-sampling coefficient is greater than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is greater than the second down-sampling coefficient, and the second down-sampling coefficient is greater than the first down-sampling coefficient;
the identifying unit 23 is configured to perform target identification on the first area point cloud data and the second area point cloud data to obtain target object information.
As shown in fig. 6, the embodiment of the present application further provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and capable of running on the processor, where the processor 320 implements any one of the steps of the foregoing Lei Dadian cloud data downsampling process when executing the computer program 311.
Since the electronic device described in this embodiment is a device for implementing the radar point cloud data downsampling device in this embodiment, based on the method described in this embodiment, those skilled in the art can understand the specific implementation manner of the electronic device and various modifications thereof, so how to implement the method in this embodiment for the electronic device will not be described in detail herein, and only those devices for implementing the method in this embodiment for those skilled in the art are within the scope of protection intended by this application.
In a specific implementation, the computer program 311 may implement any of the embodiments corresponding to fig. 1 when executed by a processor.
In the foregoing embodiments, the descriptions of the embodiments are focused on, and for those portions of one embodiment that are not described in detail, reference may be made to the related descriptions of other embodiments.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Embodiments of the present application also provide a computer program product comprising computer software instructions that, when run on a processing device, cause the processing device to perform a flow of radar point cloud data downsampling process as in the corresponding embodiment of fig. 1.
The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). Computer readable storage media can be any available media that can be stored by a computer or data storage devices such as servers, data centers, etc. that contain an integration of one or more available media. Usable media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid State Disks (SSDs)), among others.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods may be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of elements is merely a logical functional division, and there may be additional divisions of actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (10)
1. The radar point cloud data downsampling processing method is characterized by comprising the following steps of:
acquiring down-sampling boundary area point cloud data of a target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient;
the down-sampling boundary area point cloud data are subjected to height segmentation according to a first preset height threshold value to obtain first area point cloud data and second area point cloud data, wherein the first area point cloud data are point cloud data of an area with the height lower than or equal to the first preset height threshold value, a first down-sampling coefficient is correspondingly arranged in the first area, the second area point cloud data are point cloud data with the height higher than the first preset height threshold value, a second down-sampling coefficient is correspondingly arranged in the second area point cloud data, the perceived internal down-sampling coefficient is larger than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient;
and carrying out target identification on the first regional point cloud data and the second regional point cloud data to acquire target object information.
2. The method of claim 1, wherein the target object information comprises object type information of a target object;
the method further comprises the steps of:
and determining a third downsampling coefficient of a corresponding region of the target object based on the object type information.
3. The method of claim 2, wherein the determining a third downsampling factor for a corresponding region of a target object based on the object type information comprises:
setting a third downsampling coefficient corresponding to the target object as the first downsampling coefficient when the height of the target object is smaller than or equal to the first preset height threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as the perceived internal downsampling coefficient when the height of the target object is greater than the first preset height threshold and the length is greater than the length threshold or the height of the target object is greater than the first preset height threshold or the height is greater than the first preset height threshold and the width is greater than the width threshold;
and/or the number of the groups of groups,
setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in a downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the length is smaller than or equal to the length threshold;
and/or the number of the groups of groups,
and setting a third downsampling coefficient corresponding to the target object as a non-low area point cloud downsampling coefficient in the downsampling boundary under the condition that the height of the target object is larger than the first preset height threshold and the width is smaller than the width threshold.
4. The method as recited in claim 2, further comprising:
and selecting a smaller downsampling coefficient in the corresponding region of the overlapped target object to downsample the overlapped region under the condition that the corresponding regions of the plurality of target objects have the overlapped region.
5. The method as recited in claim 4, further comprising:
acquiring a movement rule of a smaller target object under the condition that the smaller target object is overlapped behind a larger target object;
and carrying out downsampling processing in the overlapped area based on the movement rule and the smaller downsampling coefficient.
6. The method as recited in claim 2, further comprising:
acquiring the moving speed of the moving object under the condition that the target object is the moving object;
and determining a fourth downsampling coefficient of the area corresponding to the moving object based on the moving speed.
7. The method as recited in claim 2, further comprising:
acquiring the moving direction of the moving object under the condition that the target object is the moving object;
the perceived internal downsampling coefficient is set to a downsampling coefficient of the moving object corresponding region in a case where the moving direction of the object does not overlap with the moving direction of the target vehicle.
8. A radar point cloud data downsampling processing device, comprising:
the acquisition unit is used for acquiring the down-sampling boundary area point cloud data of the target vehicle, wherein the down-sampling boundary area point cloud data is acquired by performing boundary segmentation processing on the original point cloud data acquired by the target vehicle based on a preset down-sampling boundary width, a down-sampling boundary down-sampling coefficient and a perceived internal down-sampling coefficient;
the segmentation unit is used for carrying out height segmentation on the point cloud data of the down-sampling boundary region according to a first preset height threshold value according to a first down-sampling coefficient and a second down-sampling coefficient to obtain first region point cloud data and second region point cloud data, wherein the first region point cloud data is point cloud data with the height lower than or equal to a region corresponding to the first preset height threshold value, the second region point cloud data is point cloud data with the height higher than the corresponding first preset height threshold value, the perceived internal down-sampling coefficient is larger than the down-sampling boundary down-sampling coefficient, the down-sampling boundary down-sampling coefficient is larger than the second down-sampling coefficient, and the second down-sampling coefficient is larger than the first down-sampling coefficient;
and the identification unit is used for carrying out target identification on the first regional point cloud data and the second regional point cloud data so as to acquire target object information.
9. An electronic device, comprising: memory and processor, characterized in that the processor is adapted to carry out the steps of the radar point cloud data downsampling method according to any of the claims 1-7 when executing a computer program stored in the memory.
10. A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program, when executed by a processor, implements the radar point cloud data downsampling method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211456375.XA CN116087909A (en) | 2022-11-21 | 2022-11-21 | Radar point cloud data downsampling processing method and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211456375.XA CN116087909A (en) | 2022-11-21 | 2022-11-21 | Radar point cloud data downsampling processing method and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116087909A true CN116087909A (en) | 2023-05-09 |
Family
ID=86185722
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211456375.XA Pending CN116087909A (en) | 2022-11-21 | 2022-11-21 | Radar point cloud data downsampling processing method and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116087909A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456131A (en) * | 2023-12-26 | 2024-01-26 | 深圳市信润富联数字科技有限公司 | Down-sampling method and device for point cloud in defect scene |
-
2022
- 2022-11-21 CN CN202211456375.XA patent/CN116087909A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117456131A (en) * | 2023-12-26 | 2024-01-26 | 深圳市信润富联数字科技有限公司 | Down-sampling method and device for point cloud in defect scene |
| CN117456131B (en) * | 2023-12-26 | 2024-05-24 | 深圳市信润富联数字科技有限公司 | Down-sampling method and device for point cloud in defect scene |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7430277B2 (en) | Obstacle detection method and apparatus, computer device, and computer program | |
| CN109521757B (en) | Static obstacle identification method and device | |
| EP3767332B1 (en) | Methods and systems for radar object detection | |
| WO2011102871A1 (en) | Video surveillance system | |
| US12112554B2 (en) | Method for determining a drivable area | |
| CN114137509A (en) | Point cloud clustering method and device based on millimeter wave radar | |
| CN116087909A (en) | Radar point cloud data downsampling processing method and related equipment | |
| CN112734810B (en) | Obstacle tracking method and device | |
| CN111932644A (en) | Method and system for fusing occupied maps | |
| CN116520353B (en) | Ground detection method, device, storage medium and equipment based on laser point cloud | |
| CN112508803A (en) | Denoising method and device for three-dimensional point cloud data and storage medium | |
| CN114118276B (en) | Network training method, control method and device | |
| CN114241448A (en) | Method and device for obtaining heading angle of obstacle, electronic equipment and vehicle | |
| CN113744518B (en) | Method and device for detecting vehicle travelable area | |
| CN115546437A (en) | Slope detection method, processor and slope detection device for mechanical equipment | |
| CN115453563A (en) | Three-dimensional space dynamic object identification method, system and storage medium | |
| CN114879220A (en) | Method and device for determining ground obstacle, electronic equipment and storage medium | |
| JP6358051B2 (en) | Transition prediction data generation device and transition prediction device | |
| CN114359346A (en) | Point cloud data processing method and device, nonvolatile storage medium and processor | |
| WO2023073987A1 (en) | Electronic control device and object identification method | |
| Jadaa et al. | Object detection and localization using wireless sensor networks based on probabilistic model | |
| KR102683721B1 (en) | Apparatus and method for removing outlier of point cloud data | |
| CN113723318B (en) | Method, device and monitoring equipment for determining shielding state of target object | |
| US20240200970A1 (en) | Environmental Reference Map Building Based on Occupancy Grid | |
| CN116386001A (en) | Method for removing suspended ghost point cloud, automatic driving method and related equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |