Point cloud data processing method and device, equipment, storage medium and vehicle

Abstract

The present disclosure relates to a point cloud data processing method, device, equipment, storage medium and vehicle. The present disclosure identifies a moving foreground obstacle in front of the vehicle and determines the area passed by the foreground obstacle. Further, it can be determined that there is no target obstacle in the area through which the foreground obstacle can pass, which affects the passing of the ego vehicle. Therefore, the candidate obstacle information in the area passed by the foreground obstacle is removed, the influence of the candidate obstacle in the surrounding environment of the vehicle which does not affect the passing is excluded, thereby improving the accuracy of detecting the target obstacle in the surrounding environment of the vehicle which affects the passing, and effectively reducing the influence of obstacle false detection on the planning of the driving route of the vehicle or the control of the vehicle.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of computer, and particularly relates to a point cloud data processing method and device, equipment, storage medium and vehicle. BACKGROUND

[0002] With the development of point cloud processing technology, the existing scheme can perceive and detect the surrounding environment based on point cloud data.

[0003] In the automatic driving technology, the obstacles around the vehicle can be perceived based on the point cloud data. Generally, the green plants invading the road, the dust in the lane, the exhaust of the front vehicle, the willow catkins, etc. will not affect the normal driving of the vehicle, but since these objects will generate noise point data, the perception device will mistakenly detect these objects as obstacles affecting the vehicle passing, which will affect the planning of the vehicle driving route or the control of the vehicle. SUMMARY

[0004] To solve the above technical problems, the present disclosure provides a point cloud data processing method, device, equipment, storage medium and vehicle to improve the accuracy of detecting the obstacles around the vehicle affecting the passing, and effectively reduces the impact of obstacle mis-detection on the planning of the vehicle driving route or the control of the vehicle.

[0005] In a first aspect, an embodiment of the present disclosure provides a point cloud data processing method, comprising:

[0006] According to the point cloud data of the environment around the vehicle, data of a candidate obstacle is obtained, the candidate obstacle including a foreground obstacle moving in front of the vehicle;

[0007] According to the data of the foreground obstacle, a region passed by the foreground obstacle is determined;

[0008] The point cloud data of the candidate obstacle in the region passed by the foreground obstacle is removed to obtain point cloud data corresponding to a target obstacle.

[0009] In some embodiments, the data of the candidate obstacle includes:

[0010] The position of the candidate obstacle, the size of the candidate obstacle, and the orientation of the candidate obstacle.

[0011] In some embodiments, the region where the foreground obstacle is located includes:

[0012] According to the data of the foreground obstacle, the region passed by the foreground obstacle is determined, comprising:

[0013] According to the data of the continuous multiple frames of the foreground obstacle, a motion trajectory of the foreground obstacle is determined;

[0014] According to the data of the foreground obstacle and the motion trajectory of the foreground obstacle, a region passed by the foreground obstacle is determined.

[0015] In some embodiments, the method further comprises:

[0016] According to the point cloud data corresponding to the target obstacle, a height of the target obstacle relative to the ground is determined.

[0017] If the height of the target obstacle relative to the ground is greater than a height threshold, the point cloud data corresponding to the target obstacle is removed.

[0018] In some embodiments, the method further comprises:

[0019] According to the data of each of the candidate obstacles, an obstacle image is generated.

[0020] In some embodiments, generating the obstacle image according to the data of each of the candidate obstacles comprises:

[0021] If the candidate obstacle is located in the region passed by the foreground obstacle, a confidence of the candidate obstacle is determined as a first confidence;

[0022] If the candidate obstacle is located outside the region passed by the foreground obstacle, and the height of the candidate obstacle relative to the ground is less than or equal to a height threshold, a confidence of the candidate obstacle is determined as a second confidence, the second confidence being lower than the first confidence;

[0023] According to the confidence of the candidate obstacle, the obstacle image is generated.

[0024] In some embodiments, generating the obstacle image according to the confidence of the candidate obstacle comprises:

[0025] If the confidence of the candidate obstacle is the first confidence, the obstacle is not displayed;

[0026] If the confidence of the candidate obstacle is the second confidence, the candidate obstacle is displayed on a corresponding position of the obstacle image.

[0027] In a second aspect, the embodiments of the present disclosure provide a point cloud data processing apparatus, comprising:

[0028] An acquisition module is configured to acquire data of candidate obstacles according to point cloud data of a surrounding environment of a vehicle, the candidate obstacles including a foreground obstacle moving in front of the vehicle;

[0029] A determination module is configured to determine a region passed by the foreground obstacle according to the data of the foreground obstacle.

[0030] The screening module is configured to remove the point cloud data of the candidate obstacles in the area passed by the foreground obstacle, to obtain the point cloud data corresponding to the target obstacle.

[0031] a memory;

[0032] a processor; and

[0033] a computer program;

[0034] The computer program is stored in the memory and configured to be executed by the processor to implement the method of the first aspect.

[0035] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method of the first aspect.

[0036] In a fifth aspect, an embodiment of the present disclosure further provides a computer program product, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement the point cloud data processing method as described above.

[0037] In a sixth aspect, an embodiment of the present disclosure further provides a vehicle, which includes the point cloud data processing apparatus as described above.

[0038] The point cloud data processing method, apparatus, device, storage medium and vehicle provided by the embodiments of the present disclosure can identify the foreground obstacle moving in front of the vehicle and determine the area passed by the foreground obstacle, and further determine that there is no target obstacle affecting the passage of the ego vehicle in the area through which the foreground obstacle can pass, so that the candidate obstacle information in the area passed by the foreground obstacle is removed, the influence of the candidate obstacle in the surrounding environment of the vehicle which does not affect the passage is excluded, thereby improving the accuracy of detecting the target obstacle in the surrounding environment of the vehicle which affects the passage, and effectively reducing the influence of obstacle false detection on the planning of the driving route of the vehicle or the control of the vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0039] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and serve to explain the principles of the present disclosure together with the specification.

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, those skilled in the art can obtain other drawings from these drawings without any creative effort.

[0041] Figure 1 A point cloud data processing method flowchart provided by an embodiment of the present disclosure;

[0042] Figure 2 A point cloud data schematic diagram of a vehicle surrounding environment provided by an embodiment of the present disclosure;

[0043] Figure 3 A point cloud data processing method flowchart provided by another embodiment of the present disclosure;

[0044] Figure 4 A point cloud data schematic diagram provided by another embodiment of the present disclosure;

[0045] Figure 5 A foreground obstacle motion trajectory schematic diagram provided by an embodiment of the present disclosure;

[0046] Figure 6 An obstacle image generation schematic diagram provided by an embodiment of the present disclosure;

[0047] Figure 7 A structure schematic diagram of a point cloud data processing apparatus provided by an embodiment of the present disclosure;

[0048] Figure 8 A structure schematic diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0049] In order to enable a person skilled in the art to more clearly understand the above-mentioned purposes, features and advantages of the present disclosure, the schemes of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0050] In the following description, many specific details are set forth in order to provide a thorough understanding of the present disclosure, but the present disclosure can also be implemented in other ways different from those described herein; obviously, the embodiments in the description are only some of the embodiments of the present disclosure, not all the embodiments.

[0051] An embodiment of the present disclosure provides a point cloud data processing method, which will be introduced below in combination with specific embodiments.

[0052] Figure 1 A point cloud data processing method flowchart provided by an embodiment of the present disclosure. The method can be executed by a point cloud data processing apparatus, which can be realized in software and / or hardware, and the obstacle point cloud data processing apparatus can be configured in an electronic device, such as a server or a terminal, wherein the terminal specifically includes a car machine, a smart phone, a palm computer, a tablet computer, a notebook computer, an all-in-one machine, a smart driving device, etc. It can be understood that the point cloud data processing method provided by the embodiments of the present disclosure can also be applied in other scenarios.

[0053] The following is about Figure 1 The point cloud data processing method shown is described below, and the specific steps of this method are as follows:

[0054] S101. Based on the point cloud data of the vehicle's surrounding environment, obtain data on candidate obstacles, including foreground obstacles moving in front of the vehicle.

[0055] Point cloud data can be collected by a point cloud data processing device using a lidar device installed on the vehicle. When a laser beam shines on the surface of an object, the reflected laser light carries information such as the object's location and distance. If the laser beam is scanned along a certain trajectory, a large number of laser points can be obtained, forming laser point cloud data. Point cloud data of the vehicle's surrounding environment can be as follows: Figure 2 The image shown illustrates how a point cloud data processing device analyzes point cloud data of the vehicle's surrounding environment to obtain data on candidate obstacles. Candidate obstacles include non-collision obstacles that impede vehicle passage, such as traffic cones, roadside barriers, signs, roadside buildings, trees, etc. These are also referred to as background obstacles. Candidate obstacles also include objects that do not impede vehicle passage; however, due to limitations in obstacle recognition capabilities based on point cloud data, encroaching vegetation, dust in the lane, vehicle exhaust fumes, and willow catkins may also be identified as background obstacles. Candidate obstacles also include foreground obstacles moving in front of the vehicle, such as vehicles traveling in the same direction ahead, pedestrians, and bicycles. These foreground obstacles are not stable and differ from stable background obstacles like roadside buildings and trees.

[0056] S102. Based on the data of the foreground obstacle, determine the area traversed by the foreground obstacle.

[0057] Point cloud data contains information such as the orientation and distance of objects. Based on the corresponding point cloud data of foreground obstacles, the area traversed by the foreground obstacles can be determined.

[0058] S103. Remove the point cloud data of the candidate obstacles in the area traversed by the foreground obstacle to obtain the point cloud data corresponding to the target obstacle.

[0059] Since the foreground obstacle is not stable, and the area where the foreground obstacle is located at the current time cannot exist the background obstacle. Specifically, with the movement of the foreground obstacle, the point cloud data processing apparatus detects that the foreground obstacle is located at the first position at the first time, and detects that the foreground obstacle is located at the second position at the second time. At the first time, since the foreground obstacle occupies the first position, it proves that there is no other non-collidable obstacle in the first position; and when the second time, the foreground obstacle moves from the first position to the second position, at this time, there is neither the foreground obstacle nor other non-collidable obstacles in the first position, then the position belongs to the area where the vehicle can pass, and the point cloud data of the candidate obstacle in the first position is removed. For example, when it is detected that the foreground obstacle is a vehicle driving in the same direction in front of the vehicle, for any area around the vehicle, if the front vehicle can pass, even if the candidate obstacle in the area is identified according to the point cloud data, it can be proved that there is no non-collidable obstacle in the area that affects the passing of the vehicle, and the point cloud data of the candidate obstacle in the front vehicle driving track area is removed. The point cloud data of the environment around the vehicle is processed as described above, and the point cloud data of the candidate obstacle in the area where the foreground obstacle passes is removed, and finally the point cloud data corresponding to the target obstacle is obtained.

[0060] The embodiment of the present disclosure obtains the data of the candidate obstacle including the foreground obstacle moving in front of the vehicle according to the point cloud data of the environment around the vehicle, determines the area passed by the foreground obstacle according to the data of the foreground obstacle, removes the point cloud data of the candidate obstacle in the area passed by the foreground obstacle, and obtains the point cloud data corresponding to the target obstacle. By identifying the foreground obstacle moving in front of the vehicle and determining the area passed by the foreground obstacle, it can be further determined that there is no target obstacle affecting the passing of the vehicle in the area where the foreground obstacle can pass, so the candidate obstacle information in the area passed by the foreground obstacle is removed, the influence of the candidate obstacle not affecting the passing in the environment around the vehicle is excluded, thereby improving the accuracy of detecting the target obstacle affecting the passing around the vehicle, and effectively reducing the influence of obstacle false detection on the planning of the driving route of the vehicle or the control of the vehicle.

[0061] Figure 3 The point cloud data processing method flowchart provided for another embodiment of the present disclosure is shown in FIG. 8. As shown in the figure, the method includes the following steps: Figure 3

[0062] S301, training the obstacle detection model according to the labeled point cloud data to obtain the trained obstacle detection model.

[0063] ​The sample point cloud data is labeled in advance to obtain a large amount of labeled point cloud data. The labeled point cloud data is used to train the obstacle detection model to obtain the trained obstacle detection model. The obstacle detection model is used to extract data of a candidate obstacle from input point cloud data of a vehicle surrounding environment.

[0064] In S302, the point cloud data of the vehicle surrounding environment is input into the obstacle detection model to obtain data of the candidate obstacle.

[0065] Specifically, the data of the candidate obstacle includes a position of the candidate obstacle, a size of the candidate obstacle, and an orientation of the candidate obstacle.

[0066] Figure 4 A point cloud data schematic diagram provided by the embodiment of the present disclosure is shown. Figure 4 The various types of data of the overhead view of the region of interest with a real range of 120m*120m and a resolution of 480*480 are shown, wherein the image in window 41 represents a minimum height value of an object at each position in the region of interest, the image in window 42 represents an average height value of the object at each position in the region of interest, the image in window 43 represents a maximum height value of the object at each position in the region of interest, and the image in window 44 represents a number of laser points obtained at each position in the region of interest. It can be understood that the height value of the object can be a height value of the object relative to a plane where the point cloud data acquisition device is located, or a height value of the object relative to the ground plane, and the division range of the region of interest is only an example and does not constitute a limitation on the embodiment. A person skilled in the art can determine the best height reference plane and the best division range of the region of interest through a limited number of experiments.

[0067] The above data is input into the trained obstacle detection model to obtain data of the candidate obstacle. The data of the candidate obstacle can include a position of the candidate obstacle, such as coordinates of the candidate obstacle in the real world, and can also include a size of the candidate obstacle and an orientation of the candidate obstacle. Meanwhile, as shown in window 45 and window 46, the obstacle detection model can also obtain a confidence at each position in the region of interest and an ID of a point at each position. The confidence represents a probability of passing through the position, and the higher the confidence, the higher the probability of passing through the position. Figure 4

[0068] In S303, an area passed through by the foreground obstacle is determined according to the data of the foreground obstacle.

[0069] ​Specifically, the motion trajectory of the foreground obstacle is determined based on continuous multi-frame data of foreground obstacles; the area traversed by the foreground obstacle is determined based on the foreground obstacle data and the motion trajectory of the foreground obstacle. When the obstacle detection model confirms that the foreground obstacle is a pedestrian, bicycle, or other foreground obstacle with unstable movement speed, the location of the foreground obstacle is determined using the data of a single frame of foreground obstacle, and its current location is marked as the area where the foreground obstacle is located. When the obstacle detection model confirms that the foreground obstacle is a vehicle traveling in the same direction in front of the vehicle with relatively stable speed, the data of the foreground obstacle in the current frame of foreground obstacle is determined based on the data of the foreground obstacle in the current frame of foreground obstacle. The motion trajectory of the foreground obstacle over a period of time is obtained by superimposing the data of multiple frames of foreground obstacle. Combined with data such as the size of the foreground obstacle, the area covered by its motion trajectory, i.e., the area traversed by the foreground obstacle, can be obtained. Figure 5 This is a schematic diagram of the motion trajectory of a foreground obstacle provided in an embodiment of this disclosure. Based on the superposition of multiple frames of foreground obstacle data, it is possible to determine, as follows: Figure 5 The motion trajectory of the foreground obstacle shown is the area covered by the motion trajectory, which is the area traversed by the foreground obstacle. It is understood that the data selection method used above to determine the area traversed by the foreground obstacle is only an example. For example, in other embodiments, the motion trajectory of foreground obstacles such as pedestrians and bicycles on the roadside can also be determined by using data from multiple consecutive frames of foreground obstacles, and the motion trajectory can be further determined as the area traversed by the foreground obstacle. This disclosure does not limit this aspect.

[0070] S304. Remove the point cloud data of the candidate obstacles in the area traversed by the foreground obstacle to obtain the point cloud data corresponding to the target obstacle.

[0071] S305. Determine the height of the target obstacle relative to the ground based on the point cloud data corresponding to the target obstacle.

[0072] S306. If the height of the target obstacle relative to the ground is greater than a height threshold, then the point cloud data corresponding to the target obstacle is removed.

[0073] In order to further improve the accuracy of the detection of the obstacle affecting the passing, the target obstacle is screened according to the height data of the target obstacle from the ground outside the area passed by the foreground obstacle. According to the data of the target obstacle obtained in the above steps, the minimum value of the height of the target obstacle from the ground at each position outside the area passed by the foreground obstacle is determined as the height of the target obstacle relative to the ground at the position. If the height of the target obstacle relative to the ground is greater than the height threshold, it is considered that the obstacle does not affect the passing of the vehicle, and the point cloud data corresponding to the target obstacle at the position is removed, or the point cloud data of the target obstacle is removed. Finally, the remaining target obstacle is the obstacle affecting the passing.

[0074] S307, generating an obstacle image according to the data of each candidate obstacle in the candidate obstacles.

[0075] Optionally, if the candidate obstacle is located in the area passed by the foreground obstacle, the confidence of the candidate obstacle is determined as a first confidence; if the candidate obstacle is located outside the area passed by the foreground obstacle and the height of the candidate obstacle relative to the ground is less than or equal to the height threshold, the confidence of the candidate obstacle is determined as a second confidence; if the confidence of the candidate obstacle is the first confidence, the obstacle is not displayed; if the confidence of the candidate obstacle is the second confidence, the candidate obstacle is displayed at the corresponding position of the obstacle image, wherein the second confidence is lower than the first confidence.

[0076] After the point cloud data processing flow of S301-S306, the point cloud data of the target obstacle affecting the passing in the environment around the vehicle can be determined. On this basis, an image of the target obstacle affecting the passing in the environment around the vehicle can be directly generated according to the point cloud data of the target obstacle affecting the passing in the environment around the vehicle, or an image of the target obstacle affecting the passing in the environment around the vehicle can be generated according to the data of each candidate obstacle in the candidate obstacles.

[0077] Specifically, according to the data of the candidate obstacle, the confidence of the candidate obstacle or the confidence of each position in the area covered by the candidate obstacle can be determined. If the candidate obstacle is located in the area passed by the foreground obstacle, it is considered that the candidate obstacle does not affect the normal traffic of the vehicle, and the confidence of the candidate obstacle is determined as a first confidence. If the candidate obstacle is located outside the area passed by the foreground obstacle, and the height of the obstacle relative to the ground is less than or equal to a preset threshold, it is considered that the candidate obstacle belongs to the target obstacle that may affect the traffic, and the confidence of the candidate obstacle is determined as a second confidence, wherein the second confidence is lower than the first confidence. According to the confidence information of each position in the region of interest, an obstacle image in the region of interest affecting the traffic is generated. Figure 6 The obstacle image generation schematic diagram provided by the embodiment of the present disclosure is shown in the following figure. The left image represents the confidence information of each position in the region of interest, and the right image is the finally generated obstacle image in the region of interest affecting the traffic. As shown in the figure, Figure 6 If the confidence of the candidate obstacle is the first confidence, the obstacle is not displayed in the finally generated obstacle image in the region of interest affecting the traffic. If the confidence of the candidate obstacle is the second confidence, the candidate obstacle is displayed on the corresponding position in the finally generated obstacle image in the region of interest affecting the traffic.

[0078] Optionally, if the motion trajectory of the foreground obstacle is determined according to the data of the continuous multiple frames of the foreground obstacle, it can be determined that the candidate obstacle in the area covered by the motion trajectory of the foreground obstacle has a higher confidence, for example, the first confidence. If the position of the foreground obstacle is determined by the data of a single frame of the foreground obstacle, due to the limited information represented by the data of the single frame of the foreground obstacle, in order to avoid missing the detection of the obstacle affecting the traffic, the confidence of the position of the foreground obstacle can be obtained by the obstacle detection model. A suitable confidence threshold is set in advance, and the confidence threshold is lower than the first confidence. When generating the obstacle image in the region of interest affecting the traffic, if the confidence of the candidate obstacle is less than or equal to the confidence threshold, the candidate obstacle is displayed on the corresponding position in the finally generated obstacle image in the region of interest affecting the traffic. If the confidence of the candidate obstacle is greater than the confidence threshold, the generation of the obstacle image on the corresponding position is inhibited in the process of generating the obstacle image, and the candidate obstacle on the corresponding position is not finally displayed.

[0079] This embodiment trains an obstacle detection model using labeled point cloud data to obtain a trained obstacle detection model. Point cloud data of the vehicle's surrounding environment is input into the obstacle detection model to obtain candidate obstacle data. Based on the foreground obstacle data, the area traversed by the foreground obstacle is determined. The point cloud data of candidate obstacles within the area traversed by the foreground obstacle is removed to obtain the point cloud data corresponding to the target obstacle. The height of the target obstacle relative to the ground is determined based on the point cloud data corresponding to the target obstacle. If the height of the target obstacle relative to the ground is greater than a height threshold, the point cloud data corresponding to the target obstacle is removed. In areas with foreground obstacle data, candidate obstacles are filtered based on the point cloud data of the foreground obstacles; in areas without foreground obstacle data, candidate obstacles are filtered based on the height of the candidate obstacles above the ground. This filters out noise data in the vehicle's surrounding environment that affects obstacle recognition, such as roadside vegetation, dust in the lane, exhaust fumes from vehicles ahead, and willow catkins, effectively improving the accuracy of obstacle detection and thus enhancing the accuracy of vehicle route planning or vehicle control.

[0080] Figure 7 This is a schematic diagram of the structure of a point cloud data processing apparatus provided in an embodiment of this disclosure. The point cloud data processing apparatus may be an electronic device as described in the above embodiments, or it may be a component or assembly within that electronic device. The point cloud data processing apparatus provided in this disclosure can execute the processing flow provided in the point cloud data processing method embodiments, such as... Figure 7 As shown, the point cloud data processing device 70 includes: an acquisition module 71, a determination module 72, and a filtering module 73; wherein, the acquisition module 71 is used to acquire candidate obstacle data based on the point cloud data of the vehicle's surrounding environment, the candidate obstacles including foreground obstacles moving in front of the vehicle; the determination module 72 is used to determine the area traversed by the foreground obstacle based on the foreground obstacle data; the filtering module 73 is used to remove the point cloud data of candidate obstacles within the area traversed by the foreground obstacle to obtain the point cloud data corresponding to the target obstacle.

[0081] Optionally, the acquisition module 71 includes a training unit 711 and a detection unit 712; the training unit 711 is used to train the obstacle detection model based on the labeled point cloud data to obtain the trained obstacle detection model; the detection unit 712 is used to input the point cloud data of the vehicle's surrounding environment into the obstacle detection model to obtain the data of candidate obstacles.

[0082] Optionally, the determining module 72 is further configured to determine a motion trajectory of the foreground obstacle according to the data of the continuous multiple frames of the foreground obstacle; and determine a region passed by the foreground obstacle according to the data of the foreground obstacle and the motion trajectory of the foreground obstacle.

[0083] Optionally, the screening module 73 is further configured to determine a height of the target obstacle relative to the ground according to the point cloud data corresponding to the target obstacle; and remove the point cloud data corresponding to the target obstacle if the height of the target obstacle relative to the ground is greater than a height threshold.

[0084] Optionally, the point cloud data processing apparatus 70 further comprises a generating module 74 configured to generate an obstacle image according to the data of each of the candidate obstacles.

[0085] Optionally, the generating module 74 comprises a determining unit 741 and a generating unit 742. The determining unit 741 is configured to determine a confidence of the candidate obstacle as a first confidence if the candidate obstacle is located in the region passed by the foreground obstacle; and determine the confidence of the candidate obstacle as a second confidence if the candidate obstacle is located outside the region passed by the foreground obstacle and the height of the candidate obstacle relative to the ground is less than or equal to a height threshold, the second confidence being lower than the first confidence. The generating unit 742 is configured to generate the obstacle image according to the confidence of the candidate obstacle.

[0086] Optionally, the generating unit 742 is configured to not display the obstacle if the confidence of the candidate obstacle is the first confidence; and display the candidate obstacle at a corresponding position of the obstacle image if the confidence of the candidate obstacle is the second confidence.

[0087] Figure 7 The point cloud data processing apparatus of the illustrated embodiment can be used to execute the technical solutions of the above-mentioned method embodiments, and has similar principles and technical effects, which will not be described here again.

[0088] In addition, the present disclosure also provides a vehicle, which comprises the point cloud data processing apparatus as described in the above embodiments.

[0089] Figure 8 A structural schematic diagram of an electronic device provided by the present disclosure is shown. The electronic device can be a server or a terminal as described in the above embodiments. The electronic device provided by the present disclosure can execute the processing flow provided by the point cloud data processing method embodiments, such as Figure 8As shown, the electronic device 80 includes a memory 81, a processor 82, a computer program and a communication interface 83; wherein the computer program is stored in the memory 81 and is configured to be executed by the processor 82 to perform the point cloud data processing method as described above.

[0090] The memory 81 is a non-transitory computer readable storage medium, which can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the point cloud data processing method in the embodiments of the present disclosure. The processor 82 performs various functional applications and data processing of the server by running the software programs, instructions and modules stored in the memory 81, that is, implements the point cloud data processing method of the above method embodiments.

[0091] The memory 81 can include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required by a function; the data storage area can store data created according to the use of the vehicle, etc. In addition, the memory 81 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory 81 can optionally include a memory disposed remotely with respect to the processor 82, and these remote memories can be connected to the terminal device through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0092] In addition, the embodiments of the present disclosure also provide a computer readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the point cloud data processing method described in the above embodiments.

[0093] In addition, the embodiments of the present disclosure also provide a computer program product, which includes a computer program or instructions, and the computer program or instructions are executed by a processor to implement the point cloud data processing method as described above.

[0094] Computer program code for carrying out operations of the present disclosure can be written in any one or more programming languages or combinations of languages including object or visual programming languages such as Java, Smalltalk, C++, or conventional procedural programming languages such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0095] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams or block diagrams can represent a module, a procedure, or a portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or in the reverse order, depending on the functionality involved. Also, each block in the block diagrams and / or flow diagrams, and combinations of blocks in the block diagrams and / or flow diagrams, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and

[0096] It should be noted that, in the present document, the terms such as "first" and "second" and the like are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between such entities or operations. Also, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without more limitations, an element preceded by "comprises... a" does not, without more limitations, foreclose the existence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0097] The foregoing is merely illustrative of the various implementations of the present disclosure and the general principles thereof. Numerous modifications can be made to these illustrations, and equivalents can be substituted therefor, without departing from the scope of the present disclosure. The specific embodiments commensurate with the specific application are intended to be illustrative only and not limiting of the scope of the application as set forth in the following claims.

Claims

1. A method for processing point cloud data, characterized in that, The method comprises: acquiring data of candidate obstacles according to point cloud data of a surrounding environment of a vehicle, the candidate obstacles including a foreground obstacle moving in front of the vehicle; determining an area passed by the foreground obstacle according to the data of the foreground obstacle; removing point cloud data of candidate obstacles in the area passed by the foreground obstacle to obtain point cloud data corresponding to a target obstacle.

2. The method of claim 1, wherein, The data of the candidate obstacles comprises: a position of the candidate obstacle, a size of the candidate obstacle, and an orientation of the candidate obstacle.

3. The method of claim 1, wherein, The determining of the area passed by the foreground obstacle according to the data of the foreground obstacle comprises: determining a motion trajectory of the foreground obstacle according to data of the foreground obstacle in continuous multiple frames; determining the area passed by the foreground obstacle according to the data of the foreground obstacle and the motion trajectory of the foreground obstacle.

4. The method of claim 1, wherein, The method further comprises: determining a height of the target obstacle relative to the ground according to the point cloud data corresponding to the target obstacle; if the height of the target obstacle relative to the ground is greater than a height threshold, removing the point cloud data corresponding to the target obstacle.

5. The method of claim 1, wherein, The method further comprises: generating an obstacle image according to the data of each of the candidate obstacles.

6. The method of claim 5, wherein, The generating of the obstacle image according to the data of each of the candidate obstacles comprises: if the candidate obstacle is located in the area passed by the foreground obstacle, determining a confidence of the candidate obstacle as a first confidence; if the candidate obstacle is located outside the area passed by the foreground obstacle and a height of the candidate obstacle relative to the ground is less than or equal to a height threshold, determining a confidence of the candidate obstacle as a second confidence, the second confidence being lower than the first confidence; generating the obstacle image according to the confidence of the candidate obstacle.

7. The method of claim 6, wherein, The generating of the obstacle image according to the confidence of the candidate obstacle comprises: if the confidence of the candidate obstacle is the first confidence, not displaying the obstacle; if the confidence of the candidate obstacle is the second confidence, displaying the candidate obstacle at a corresponding position of the obstacle image.

8. A point cloud data processing apparatus, characterized by comprising: The device comprises: an acquiring module configured to acquire data of candidate obstacles according to point cloud data of a surrounding environment of a vehicle, the candidate obstacles including a foreground obstacle moving in front of the vehicle; a determining module configured to determine an area passed by the foreground obstacle according to the data of the foreground obstacle; a screening module configured to remove point cloud data of candidate obstacles in the area passed by the foreground obstacle to obtain point cloud data corresponding to a target obstacle.

9. An electronic device, comprising: comprise: a memory; a processor; and a computer program; wherein the computer program is stored in the memory and configured to be executed by the processor to implement the method of any one of claims 1-7.

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

11. A vehicle comprising: The point cloud data processing device of claim 8.

Images