CN117292354A

CN117292354A - Processing method, device, equipment and storage medium of point cloud data

Info

Publication number: CN117292354A
Application number: CN202311219887.9A
Authority: CN
Inventors: 司马兵; 刘京凯
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2023-09-20
Filing date: 2023-09-20
Publication date: 2023-12-26

Abstract

The disclosure provides a processing method, a processing device, processing equipment and processing media for point cloud data, relates to the technical field of artificial intelligence, and particularly relates to the technical field of automatic driving. The specific implementation scheme is as follows: extracting data to be processed according to class information of the point cloud data; constructing a fitting reference line according to first coordinate information of data to be processed; determining a second coordinate system according to the fitting reference line to obtain second coordinate information of the data to be processed; according to the second coordinate information, determining the coordinate distance between two adjacent point cloud data in the data to be processed; and determining the obstacle identifier corresponding to the point cloud data according to the coordinate interval between the two adjacent point cloud data. The processing method can accurately distinguish the obstacle identifier of each point cloud data, improves the identification precision of the point cloud data, and improves the perception precision of the autonomous vehicle to the obstacle, thereby improving the passing efficiency.

Description

Processing method, device, equipment and storage medium of point cloud data

Technical Field

The disclosure relates to the technical field of data processing, in particular to the technical field of automatic driving, and particularly relates to a method, a device, equipment and a storage medium for processing point cloud data.

Background

In the running process of an open road, fences and the like often appear at two sides of the road, and the hard characteristic of the automatic driving vehicle determines that the automatic driving vehicle cannot collide with the fences or even scratch, so that in an automatic driving system, besides normal perception of social vehicles, non-motor vehicles, pedestrians and other traffic participants, stable perception and stable profile description are also required for the hard and low fences, fences and the like.

Disclosure of Invention

The disclosure provides a processing method, a device, equipment and a storage medium for point cloud data, which effectively improve the perceived stability of a fence.

According to a first aspect of the present disclosure, there is provided a method for processing point cloud data, including:

extracting data to be processed according to class information of the point cloud data;

constructing a fitting reference line according to first coordinate information of data to be processed;

determining a second coordinate system according to the fitting reference line to obtain second coordinate information of the data to be processed;

according to the second coordinate information, determining the coordinate distance between two adjacent point cloud data in the data to be processed;

and determining the obstacle identifier corresponding to the point cloud data according to the coordinate interval between the two adjacent point cloud data.

According to a second aspect of the present disclosure, there is provided a processing apparatus for point cloud data, including:

the extraction module is configured to extract data to be processed according to category information of the point cloud data;

the construction module is configured to construct a fitting reference line according to first coordinate information of data to be processed;

the first coordinate conversion module is configured to determine a second coordinate system according to the fitting reference line to obtain second coordinate information of data to be processed;

the first determining module is configured to determine the coordinate distance between two adjacent point cloud data in the data to be processed according to the second coordinate information;

and the second determining module is configured to determine the obstacle identifier corresponding to the point cloud data according to the coordinate distance between the two adjacent point cloud data.

According to a third aspect of the present disclosure, there is provided an electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method provided in the first aspect.

According to a fourth aspect of the present disclosure there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method as provided in the first aspect.

According to a fifth aspect of the present disclosure, there is provided a computer program product comprising a computer program which, when executed by a processor, implements the method provided according to the first aspect.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is an exemplary system architecture diagram to which the processing method of point cloud data of the present disclosure may be applied;

FIG. 2 is a schematic diagram of one embodiment of a method of processing point cloud data according to the present disclosure;

FIG. 3 is a schematic diagram of a second embodiment of a method of processing point cloud data according to the present disclosure;

FIG. 4 is an implementation flow of a method of processing point cloud data according to the present disclosure;

fig. 5 is a schematic diagram of one embodiment of a processing apparatus for point cloud data according to the present disclosure.

Fig. 6 is a block diagram of an electronic device for implementing a method of processing point cloud data according to an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The scenes of real fences, simple fences and the like are all the more wonder, and the encountered problems are also endless. Taking the fence as an example, because the fence has special road attributes of short and small, easy moving and the like, the real fence position of the road can deviate relative to the position of a pre-acquired high-precision map, so that the influence of the fence on the vehicle cannot be avoided by simply relying on map labeling. In some scenes, the real road has some radians, and accordingly, the fence beside the road is curved, but the convex hull nature of the obstacle profile determines that the whole profile generated by the curved fence can certainly invade a vehicle travelling lane, so that the vehicle is stopped before, and the vehicle travelling is seriously influenced. In addition, in some scenes with narrower lanes and barriers on both sides, the general perception algorithm easily clusters the barriers on both sides into a large barrier, and if the adjustment parameters easily generate large back reflection on other general scenes, the scenes need to be subjected to special treatment.

The disclosure provides a processing method of point cloud data, which aims to solve the problems of traffic and perception of automatic driving vehicles caused by some fences or fence scenes, and reasonably optimizes the point cloud data of fence type barriers besides detection and identification of common barriers.

According to the method, the device and the system, the data to be processed are extracted according to the category information of the point cloud data, a fitting reference line is constructed according to the first coordinate information of the point cloud data, a second coordinate system is determined, then the obstacle identifications corresponding to the point cloud data are determined through the coordinate distance between two adjacent point cloud data in the second coordinate system, and therefore the obstacle identifications of each point cloud data are accurately distinguished, the perception accuracy of an automatic driving vehicle on the obstacle is improved, and the passing efficiency is further improved.

Fig. 1 illustrates an exemplary system architecture 100 to which embodiments of a processing method of point cloud data or a processing apparatus of point cloud data of the present disclosure may be applied.

As shown in fig. 1, system architecture 100 may include a terminal device 101, a network 102, and a server 103. The network 102 is used to provide a communication link between the terminal device 101 and the server 103, and may include various connection types, for example, a wired communication link, a wireless communication link, or an optical fiber cable, etc.

A user can interact with the server 103 through the network 102 using the terminal device 101 to receive or transmit information or the like. Various client applications may be installed on the terminal device 101.

The terminal device 101 may be hardware or software. When the terminal device 101 is hardware, it may be various electronic devices including, but not limited to, a smart phone, a tablet computer, a laptop portable computer, a desktop computer, and the like, and may also include additional electronic devices on a vehicle such as an in-vehicle terminal. When the terminal apparatus 101 is software, it may be installed in the above-described electronic apparatus. Which may be implemented as a plurality of software or software modules, or as a single software or software module. The present invention is not particularly limited herein.

The server 103 may be hardware or software. When the server 103 is hardware, it may be implemented as a distributed server cluster composed of a plurality of servers, or may be implemented as a single server. When the server 103 is software, it may be implemented as a plurality of software or software modules (for example, to provide distributed services), or may be implemented as a single software or software module. The present invention is not particularly limited herein.

The processing method of the point cloud data provided in the embodiments of the present disclosure is generally executed by the server 103, and accordingly, the processing device of the point cloud data is generally disposed in the server 103.

It should be noted that the numbers of the terminal device 101, the network 102, and the server 103 in fig. 1 are merely illustrative. There may be any number of terminal devices 101, networks 102, and servers 103, as desired for implementation.

In the embodiment of the present disclosure, the processing method of the point cloud data is executed by the server 103, and the processing result is sent to the terminal device 101 installed with the client after being saved, for example, the processing result is sent to the terminal device of the autopilot vehicle, and the processing result may also be sent to the mobile terminal device of the user end, and so on.

Fig. 2 shows a flow 200 of one embodiment of a method of processing point cloud data according to the present disclosure, and referring to fig. 2, the method of processing point cloud data includes the steps of:

step S201, extracting data to be processed according to category information of point cloud data.

In the embodiment of the present disclosure, an execution body of a processing method of point cloud data, for example, a server 103 shown in fig. 1, extracts data to be processed from the point cloud data according to category information of the point cloud data.

Illustratively, the executing body extracts the point cloud data of the target class as the data to be processed according to the class information of the point cloud data.

In this scheme, the category information of the point cloud data may be category information marked for the point cloud data in the high-definition map data. For example, category information of each object in the image marked in each frame of image data acquired for constructing a high definition map.

Illustratively, the category information includes, but is not limited to, text, voice, and the like.

In some optional implementations of embodiments of the present disclosure, the category information of the point cloud data includes at least one of: obstacle category, obstacle identification, and obstacle specification.

Wherein the obstacle category is an obstacle category in which point cloud data is marked in high definition map data, for example, the obstacle represented by the obstacle category in which the point cloud data is marked may include, but is not limited to, pedestrians, vehicles, non-vehicles, buildings, plants, construction facilities, roadblock facilities, and the like.

In the present disclosure, the obstacle category to which the extracted data to be processed belongs includes one or more of construction facilities, roadblock facilities, and the like, for example, fences, low movable fences, and the like.

In some optional implementations of embodiments of the present disclosure, the obstacle category includes at least one of: fence and enclosure.

Wherein, the fence can include a short and easily movable spliced fence, and also can include an easily movable and adjustable telescopic fence, which is not limited herein.

Illustratively, the enclosure type may include a short, easily movable, simple enclosure, and may also include a continuous spliced enclosure in a temporary construction facility, such as a color-steel-plate spliced enclosure, without limitation.

In the present disclosure, the obstacle identification of the point cloud data characterizes which obstacle individual the point cloud data belongs to is, i.e., the obstacles represented by the same point cloud data are the same. For example, if the obstacle identifiers corresponding to the point cloud data with the same obstacle type are different, it indicates that the obstacles represented by the point cloud data are not the same.

Illustratively, the obstacle identifier may be any one or a combination of a plurality of forms of a text label, a numeric serial number, a letter sequence, and the like.

The obstacle specifications include the size of the space occupied by the obstacle, the external dimensions of the obstacle, etc., for example, the length, width, height, aspect ratio, etc., of the obstacle.

In some optional implementations, the executing entity screens data to be processed from each frame of image data of the high-definition map data.

Firstly, the execution subject screens point cloud data of the same target class according to the obstacle class to serve as candidate data to be processed. For example, the target class is a fence class.

Illustratively, the executing subject preferentially determines that the obstacle class of the point cloud data is not a foreground obstacle of a pedestrian, a motor vehicle, a non-motor vehicle, or the like.

And then, the execution subject selects data to be processed according to the proportion of the point cloud data with the barrier type as a fence in the image frame.

For example, if the proportion of the point cloud data with the barrier type being the barrier in the current frame image exceeds the preset proportion threshold, the execution subject may confirm that the point cloud data belonging to the barrier in the image frame is candidate to be processed data.

For example, if the proportion of the point cloud data with the obstacle category of the fence in the current frame image of a certain position is lower and is lower than or far lower than the preset proportion threshold, but the proportion of the point cloud data with the obstacle category of the fence in the historical frame image of the position is higher, for example, higher than the preset proportion threshold, at this time, the executing body may update the obstacle category of the point cloud data of the position by combining the current frame image of the position and the point cloud data category information in the historical frame image, redetermine the point cloud data proportion with the obstacle category of the fence, and if the updated point cloud data proportion with the obstacle category of the fence exceeds the preset proportion threshold or the sum of the point cloud data proportions with the obstacle category of the fence and the static obstacle (for example, plants, fences, etc.) exceeds the preset proportion threshold, the executing body may still determine the corresponding point cloud data as candidate to be processed data.

Then, the executing body may further perform grouping processing on the candidate to-be-processed data based on the obstacle identifier, for example, confirm the point cloud data with the same obstacle identifier in the candidate to-be-processed data as the point cloud data of the same obstacle, so as to serve as a group of candidate to-be-processed data.

And the execution main body determines the specification of the obstacle corresponding to each obstacle identifier according to the grouping result, extracts the obstacle identifier of which the obstacle length is larger than a preset length threshold value and the aspect ratio of the obstacle is larger than a preset aspect ratio threshold value, and determines the point cloud data corresponding to the obstacle identifier as a group of data to be processed.

For each group of data to be processed extracted through the screening process, the processing method of the point cloud data provided by the disclosure can be adopted for processing.

According to the class information of the point cloud data, the data to be processed meeting the requirements is extracted and processed through the screening process, so that unnecessary consumption of resources and calculation power can be effectively reduced, and energy is saved.

It should be noted that, in the technical solution of the present disclosure, the acquisition, storage, application, etc. of the related personal information of the user, for example, the personal information of the user related to the high-definition map data, all conform to the rules of the related laws and regulations, and do not violate the popular regulations.

Step S202, a fitting reference line is constructed according to first coordinate information of data to be processed.

In the embodiment of the present disclosure, an execution body of a processing method of point cloud data, for example, a server 103 shown in fig. 1, constructs a fitting reference line according to first coordinate information of data to be processed in a first coordinate system.

After extracting the data to be processed of the target class, the execution body may convert the coordinates of the data to be processed into the first coordinate system, so as to obtain first coordinate information of the data to be processed in the first coordinate system. The first coordinate system may be a world coordinate system or an auxiliary coordinate system constructed by using data in the current scene.

The first coordinate system may be, for example, a coordinate system constructed by fitting according to actual position coordinates of the data to be processed in the world coordinate system. According to the first coordinate system, the execution body may acquire first coordinate information corresponding to each point cloud data in the data to be processed.

In the embodiment of the disclosure, the executing body determines, according to the first coordinate information of the data to be processed, the reference extending direction of the obstacle represented by the executing body, so that the extending direction of the fitting reference line constructed according to the reference extending direction can be initially considered as the reference extending direction of the obstacle.

In some optional implementations, after the executing body obtains the first coordinate information of the data to be processed, a fitting reference line is constructed according to the first coordinate information of part or all of the point cloud data in the data to be processed.

The fit reference line is illustratively a fit straight line, and the data to be processed is distributed along the extension direction of the fit straight line, on both sides of the fit straight line or on the fit straight line.

In some optional implementations of embodiments of the present disclosure, constructing a fitting reference line according to first coordinate information of data to be processed includes: determining target data according to first coordinate information of data to be processed; and constructing a fitting reference line according to the first coordinate information of the target data.

In this implementation manner, the execution body selects a plurality of point cloud data meeting the requirements as target data according to first coordinate information of each point cloud data in the data to be processed, and then constructs a fitting reference line according to the first coordinate information of each target data.

According to the implementation mode, the data to be processed are further screened according to the first coordinate information, the target data are selected to construct the fitting reference line, the construction precision of the fitting reference line can be effectively improved, and the reliability of the constructed fitting reference line is improved.

In some alternative implementations, the manner of constructing the fitted reference line according to the first coordinate information of each target data may be as follows: according to the first coordinate information of each target data, determining the average value of the x direction and the y direction of the target data in the packet data in the first coordinate system to obtain a covariance matrix of 2 x 2(wherein dx represents the difference between the x coordinate and the x average value of each target data, dy represents the difference between the y coordinate and the y average value of each target data, pc represents the number of target data); and then calculating the characteristic value and the characteristic vector by utilizing an SVD (singular value decomposition) method to obtain the characteristic vector corresponding to the minimum characteristic value, namely, the parameter of the linear equation of the fitting reference line, so that the linear equation of the fitting reference line can be obtained according to the parameter, and the construction of the fitting reference line corresponding to the grouping data is completed.

In some optional implementations of embodiments of the present disclosure, determining target data according to first coordinate information of data to be processed includes: determining a coordinate difference between a maximum coordinate value and a minimum coordinate value of the data to be processed in the first target direction according to the first coordinate information; dividing the data to be processed into at least one group of group data according to the coordinate difference and a preset length threshold; and respectively determining target data according to the proportion of the candidate point cloud data in each group data, wherein the absolute value of the coordinate value of the candidate point cloud data in the second target direction is smaller than or equal to a preset coordinate threshold value, and the second target direction is perpendicular to the first target direction.

In this implementation manner, the executing body determines a maximum value and a minimum value of coordinates in a first target direction according to first coordinate information of each point cloud data in the data to be processed; determining the appearance length of the obstacle represented by the data to be processed according to the coordinate difference between the maximum value and the minimum value of the coordinates; comparing the data with a preset length threshold value, and dividing the data to be processed into at least one group of data according to a comparison result; and then, respectively determining target data in each group of data according to the proportion of the candidate point cloud data in each group of data.

The first target direction is illustratively the length direction of the obstacle represented by the data to be processed. The second target direction is perpendicular to the first target direction, that is, the second target direction may be the width direction of the obstacle represented by the data to be processed.

Illustratively, the first target direction may be one axis direction, such as an x-axis direction, of the first coordinate system in which the first coordinate information is located; accordingly, the second target direction may be another axis direction of the first coordinate system, for example, a y-axis direction. Therefore, the coordinate value of the candidate point cloud data in the second target direction is the distance between the candidate point cloud data and the first target direction (x axis), and the discrete condition of the candidate point cloud data in the second target direction can be represented.

In the scheme, the smaller the absolute value of the coordinate value of the point cloud data in the y-axis direction is, the closer the point cloud data is to the x-axis, namely the higher the referent value of the point cloud data in the construction of the fitting reference line is. Therefore, a preset coordinate threshold is set as a threshold for the discrete condition of the point cloud data in the second target direction, that is, as a boundary value for whether the point cloud data can function in constructing the fitting reference line.

And if the absolute value of the coordinate value of the point cloud data in the second target direction is smaller than or equal to a preset coordinate threshold, that is, the distance from the point cloud data to the x axis is smaller than or equal to the preset coordinate threshold, the executing main body confirms that the fitting reference line of the point cloud data is more reliable, and the point cloud data is determined as candidate point cloud data. And then, respectively determining target data corresponding to each group of data according to the proportion of candidate point cloud data in each group of data so as to respectively construct more reliable fitting reference lines for each group of data and improve the referential of each fitting reference line to the group of data.

According to the method, the data to be processed are grouped according to the length of the obstacle represented by the data to be processed, and corresponding fitting reference lines are respectively constructed, so that the identification precision of the point cloud data in the obstacle can be improved, and the determination precision of the obstacle identification can be improved; and then determining target data for constructing a fitting reference line in each piece of grouping data according to the proportion of the candidate point cloud data in each piece of grouping data, and improving the construction precision of the fitting reference line, thereby further improving the recognition precision of the obstacle mark corresponding to each point cloud data.

In some optional implementations of embodiments of the present disclosure, dividing the data to be processed into at least one set of packet data according to the coordinate difference and the preset length threshold includes: responding to the coordinate difference being greater than a preset length threshold, dividing the data to be processed according to the preset length threshold and first coordinate information of the data to be processed, and obtaining at least two groups of different grouping data; and dividing the data to be processed into the same group of group data in response to the coordinate difference being smaller than or equal to a preset length threshold.

In this implementation manner, the executing body groups the data to be processed along the length direction of the obstacle according to a preset length threshold, and if the length of the obstacle is greater than the preset length threshold, the executing body divides the point cloud data in the data to be processed into at least two groups of different group data according to the preset length threshold; and if the length of the obstacle is smaller than the preset length threshold value, the data to be processed is processed as a group of group data.

In a real scene, the real fence type barriers may have problems of bending, irregularity and the like, so that whether to segment and optimize the data to be processed is needed to be considered when the fitting reference line is constructed, and the accuracy and reliability of the constructed fitting reference line are improved.

Taking the barrier class of the point cloud data as an example, the corresponding preset length threshold value can be 5m, and if the coordinate difference between the maximum coordinate value and the minimum coordinate value of the data to be processed in the second coordinate system is 18m, dividing the data to be processed into 4 groups of different grouping data according to the length of 5 m.

In some optional implementations, the executing body may further incorporate a preset width threshold when dividing the packet data.

For example, the execution subject may determine the first target direction of the obstacle represented by the data to be processed as the length direction thereof in the second coordinate system, and the width direction of the obstacle in the direction perpendicular to the first target direction; in this case, the execution body may determine that the coordinate difference between the maximum coordinate value and the minimum coordinate value in the longitudinal direction is the length value of the obstacle, and the coordinate difference between the maximum coordinate value and the minimum coordinate value in the width direction is the width value of the obstacle.

And when the execution body determines that the obstacle length value corresponding to the data to be processed exceeds the preset length threshold value and the obstacle width value exceeds the preset width threshold value, determining that the data to be processed needs to be grouped. At this time, the execution body may segment the data to be processed according to a preset length threshold. For example, if the preset length threshold is 5m and the preset width threshold is 0.3m, and if the length of the obstacle represented by the data to be processed is greater than 5m and the width is greater than 0.3m, the data to be processed is grouped according to 5m in the length direction of the obstacle.

In the implementation mode, whether the data to be processed are grouped or not can be judged by combining the widths of the corresponding barriers, but in the grouping process, only the barrier length direction is needed to be segmented, and the width direction is not needed to be segmented.

In this implementation manner, the execution body groups the data to be processed according to the first coordinate information of the data to be processed and the preset length threshold value, so as to reduce the processing constituent unit of the data to be processed, reduce the influence of bending and other conditions on data processing, and improve the processing precision, thereby improving the selection accuracy of the target data and the construction precision of the fitting reference line, and further improving the identification accuracy of the cloud data of each point in the data to be processed.

In some optional implementations of the embodiments of the present disclosure, determining the target data according to the proportion of the candidate point cloud data in each packet data includes: determining candidate point cloud data as target data of the group data in response to the proportion of the candidate point cloud data in the group data being greater than or equal to a preset proportion threshold; and determining all the point cloud data in the group data as target data in response to the proportion of the candidate point cloud data in the group data being smaller than a preset proportion threshold.

In this implementation manner, the execution body determines a proportion of candidate point cloud data in each group of the group data. For example, the proportion of the candidate point cloud data in a certain group of group data is greater than or equal to a preset proportion threshold (for example, 80%), at this time, the execution subject considers that the fitting reference line constructed only according to the candidate point cloud data is relatively true, and the referenceability is sufficiently high, so that the candidate point cloud data in the group data can be determined as the target data of the group data, and noise point cloud data other than the candidate point cloud data in the group data can be ignored or even removed.

If the proportion of the candidate point cloud data in a certain group of data is smaller than a preset proportion threshold value, the execution body considers that the candidate point cloud data is insufficient to construct a fitting reference line with enough referential property, and at the moment, the execution body determines all the point cloud data in the group of data as target data of the group of data. That is, the execution body constructs the fitting reference line according to all the point cloud data in the grouping data together, so as to avoid low referential of the fitting reference line caused by insufficient quantity of the point cloud data for constructing the fitting reference line.

In the implementation manner, the target data of the fitting reference line is constructed according to the proportion determination of the candidate point cloud data in the grouping data, so that the reliability of the target data can be effectively ensured, and the reliability of the constructed fitting reference line is improved.

Step S203, determining a second coordinate system according to the fitting reference line to obtain second coordinate information of the data to be processed.

In the embodiment of the present disclosure, the execution body of the processing method of the point cloud data, for example, the server 103 shown in fig. 1, determines the second coordinate system according to the fitting reference line constructed in step S202, so as to obtain second coordinate information of the point cloud data in the to-be-processed data in the second coordinate system.

As an example, the manner in which the execution body determines the second coordinate system according to the fitting reference line may be to use any one of the extending direction of the fitting reference line, the extending direction parallel to the fitting reference line, and the direction forming a preset included angle with the extending direction of the fitting reference line as the extending direction of one coordinate axis of the second coordinate system.

In some optional implementations of the embodiments of the present disclosure, determining a second coordinate system according to the fitted reference line, to obtain second coordinate information of the data to be processed includes: determining a second coordinate system by taking the extending direction and the normal vector direction of the fitting reference line as coordinate axes; and determining second coordinate information of the data to be processed in a second coordinate system.

In this implementation manner, the execution body directly uses the extending direction and the normal vector direction of the fitting reference line as two coordinate axis directions of the second coordinate system, so as to construct the second coordinate system; and then, projecting the position of each point cloud data in the data to be processed on the coordinate axis of the second coordinate system to obtain a corresponding coordinate value, thereby determining the second coordinate information of each point cloud data in the data to be processed in the second coordinate system.

According to the method, the second coordinate system is built according to the extending direction and the normal vector direction of the fitting reference line, so that coordinate system conversion of the data to be processed is achieved, second coordinate information of the data to be processed is obtained, coordinate conversion accuracy of the data to be processed is effectively guaranteed, and accuracy of the second coordinate information of the data to be processed is guaranteed.

Step S204, according to the second coordinate information, the coordinate distance between two adjacent point cloud data in the data to be processed is determined.

In the embodiment of the present disclosure, an execution body of the processing method of point cloud data, for example, the server 103 shown in fig. 1, determines, in a second coordinate system, a coordinate distance between every two adjacent point cloud data in the data to be processed according to second coordinate information of the data to be processed.

According to the scheme, whether a larger gap exists in the barrier represented by the data of the to-be-processed data or not can be determined by determining the coordinate distance between the two adjacent point cloud data, for example, whether the middle of the barrier is broken or not is determined, so that whether the barrier represented by the data to-be-processed data is a plurality of barriers arranged at intervals is determined.

Step S205, determining obstacle identifications corresponding to the point cloud data according to the coordinate spacing between two adjacent point cloud data.

In the embodiment of the present disclosure, an execution body of a processing method of point cloud data, for example, the server 103 shown in fig. 1, determines whether two adjacent point cloud data are point cloud data of the same obstacle according to a coordinate distance between the two adjacent point cloud data in a second coordinate system, so as to determine an obstacle identifier corresponding to each point cloud data.

In some optional implementations of the embodiments of the present disclosure, determining, according to a coordinate interval between two adjacent point cloud data, an obstacle identifier corresponding to the point cloud data includes: determining that the two adjacent point cloud data respectively have different obstacle identifications in response to the coordinate spacing between the two adjacent point cloud data being greater than a preset spacing threshold; and determining that the two adjacent point cloud data have the same obstacle identifier in response to the coordinate spacing between the two adjacent point cloud data being less than or equal to a preset spacing threshold.

In this implementation manner, the executing body compares the coordinate distance between every two adjacent point cloud data with a preset distance threshold, and determines whether the two adjacent point cloud data belong to the same obstacle according to the comparison result, so as to determine whether the two adjacent point cloud data have the same obstacle identifier, and further determine the obstacle identifier corresponding to each point cloud data.

According to the method, the coordinate distance between two adjacent point cloud data is compared with the preset distance threshold value, whether the two adjacent point cloud data belong to the same obstacle is determined, whether the data to be processed are the same obstacle is determined, the accuracy of determining the obstacle identification of the data to be processed is improved, and the accuracy of identifying the obstacle is improved.

In some optional implementation manners, when determining that the coordinate distance between two adjacent point cloud data is greater than the preset distance threshold, the executing body confirms that the two adjacent point cloud data respectively belong to two obstacles, where the executing body may further segment the data to be processed according to the space between the two adjacent point cloud data, then re-construct a reference line according to second coordinate information corresponding to the segmented data set to be processed, then re-determine a third coordinate system according to the newly constructed reference line, determine third coordinate information of each point cloud data in the data set to be processed in the third coordinate system, and further determine the distance between each adjacent point cloud data, thereby determining the obstacle to which each point cloud data belongs, and further determining the obstacle identifier corresponding to each point cloud data.

In the implementation manner, the identification accuracy of the cloud data of each point can be further improved by cutting the data to be processed again, constructing the reference line and transforming the coordinate system, the determination accuracy of the obstacle identification is improved, and the identification accuracy of the obstacle is further improved.

According to the processing method of the point cloud data, the to-be-processed data of the target class is extracted according to the class information of the point cloud data, then a fitting reference line which accords with the actual extension rule of the obstacle is constructed according to the first coordinate information of the to-be-processed data, a second coordinate system is determined according to the fitting reference line, and coordinate conversion of the to-be-processed data is achieved; and then determining whether the obstacles to which the point cloud data belong are the same according to the coordinate distance between two adjacent point cloud data in the second coordinate system environment, thereby determining the obstacle identifications of the point cloud data, realizing further accurate division of the data to be processed according to the obstacle identifications, improving the recognition accuracy of the obstacles, avoiding the vehicle congestion caused by inaccurate obstacle recognition, and improving the vehicle passing efficiency.

Fig. 3 illustrates a flow 300 of one embodiment of a method of processing point cloud data according to the present disclosure, and referring to fig. 3, the method of processing point cloud data includes the steps of:

step S301, extracting data to be processed according to category information of the point cloud data.

In the embodiment of the present disclosure, an execution body of a processing method of point cloud data, for example, a server 103 shown in fig. 1, extracts data to be processed of a target class according to class information of the point cloud data.

Step S301 is substantially identical to step S201 in the embodiment shown in fig. 2, and the detailed implementation may refer to the foregoing description of step S201, which is not repeated herein.

Step S302, an initial reference line is constructed according to initial coordinate information of the data to be processed.

In the embodiment of the present disclosure, the execution body of the processing method of the point cloud data, for example, the server 103 shown in fig. 1, acquires initial coordinate information of the data to be processed according to the data to be processed extracted in step S301, and constructs an initial reference line accordingly.

For example, the initial coordinate information of the data to be processed may be initial coordinate information of each point cloud data in the data to be processed in the world coordinate system.

In some optional implementations, when the executing body constructs an initial reference line according to initial coordinate information of the data to be processed, according to a distribution structure feature of the data to be processed in the world coordinate system, an average value of each point cloud data in an x direction and a y direction of the world coordinate system is calculated, so as to obtain a covariance matrix of 2×2 (wherein dx ' represents the difference between the x coordinate and the x average value of each point cloud data in the world coordinate system, dy ' represents the difference between the y coordinate and the y average value of each point cloud data in the world coordinate system, and pc ' represents the number of point cloud data in the data to be processed); and then calculating the characteristic value and the characteristic vector by utilizing an SVD (singular value decomposition) method to obtain the characteristic vector corresponding to the minimum characteristic value, namely, the parameter of the linear equation of the initial reference line, so that the linear equation of the initial reference line can be obtained according to the parameter, and the construction of the initial reference line corresponding to the data to be processed is completed.

It should be noted that, in a real scene, the real barrier may have problems such as bending, irregularity, etc., so the initial reference line is only a rough result, and still needs to be further segmented and optimized.

Step S303, determining a first coordinate system according to the initial reference line to obtain first coordinate information of the data to be processed.

In the embodiment of the present disclosure, the execution body of the processing method of the point cloud data, for example, the server 103 shown in fig. 1, determines a first coordinate system according to the initial reference line constructed in step S302, so as to obtain first coordinate information of the data to be processed in the first coordinate system.

And the execution main body determines a first coordinate system according to the constructed initial reference line, then converts coordinate information of the data to be processed, and projects the position of each point cloud data in the data to be processed into a corresponding coordinate axis in the first coordinate system to obtain the corresponding first coordinate information.

In some optional implementations of the embodiments of the present disclosure, determining a first coordinate system according to an initial reference line, to obtain first coordinate information of data to be processed includes: determining a first coordinate system by taking the extending direction and the normal vector direction of an initial reference line as coordinate axes; first coordinate information of data to be processed in a first coordinate system is determined.

In this implementation manner, since the extending direction of the initial reference line is roughly consistent with the extending direction of the obstacle represented by the data to be processed, the executing body uses the extending direction of the initial reference line and the normal vector direction as coordinate axes, and constructs the first coordinate system, so that the obstacle represented by the data to be processed is gathered near one coordinate axis of the first coordinate system as much as possible, so that the data to be processed is further processed, and the discrete condition of the point cloud data in the data to be processed is accurately distinguished, so that the obstacle to which each point cloud data belongs is determined, and the accuracy of identifying the obstacle is improved.

Step S304, a fitting reference line is constructed according to first coordinate information of the data to be processed.

In the embodiment of the present disclosure, the execution subject of the processing method of the point cloud data, for example, the server 103 shown in fig. 1, constructs the fitting reference line according to the first coordinate information obtained in step S303.

Step S304 is substantially identical to step S202 in the embodiment shown in fig. 2, and the detailed implementation may refer to the foregoing description of step S202, which is not repeated herein.

Step S305, determining a second coordinate system according to the fitting reference line to obtain second coordinate information of the data to be processed.

In the embodiment of the present disclosure, the execution body of the processing method of the point cloud data, for example, the server 103 shown in fig. 1, determines the second coordinate system according to the fitting reference line constructed in step S304, so as to obtain second coordinate information of the point cloud data in the to-be-processed data in the second coordinate system.

Step S305 is substantially identical to step S203 of the embodiment shown in fig. 2, and the detailed implementation may refer to the foregoing description of step S203, which is not repeated herein.

Step S306, according to the second coordinate information, the coordinate distance between two adjacent point cloud data in the data to be processed is determined.

In the embodiment of the present disclosure, an execution body of the processing method of point cloud data, for example, the server 103 shown in fig. 1, determines a coordinate distance between two adjacent point cloud data according to second coordinate information of each point cloud data in the data to be processed.

Step S306 is substantially identical to step S204 of the embodiment shown in fig. 2, and the detailed implementation may refer to the foregoing description of step S204, which is not repeated herein.

Step S307, determining the obstacle identification corresponding to the point cloud data according to the coordinate distance between the two adjacent point cloud data.

In the embodiment of the present disclosure, an execution body of a processing method of point cloud data, for example, the server 103 shown in fig. 1, determines whether two adjacent point cloud data belong to the same obstacle according to a coordinate interval between the two adjacent point cloud data, so as to determine an obstacle identifier corresponding to each point cloud data.

Step S307 is substantially identical to step S205 of the embodiment shown in fig. 2, and the detailed implementation may refer to the foregoing description of step S205, which is not repeated herein.

In the embodiment of the disclosure, after the execution main body extracts the data to be processed, an initial reference line is roughly fitted and constructed according to initial coordinate information of the data to be processed in a world coordinate system, and the extending direction of the initial reference line is close to the distribution direction of the data to be processed, so that a first coordinate system is constructed according to the initial reference line, and first coordinate information of the data to be processed is obtained; then, constructing a fitting reference line according to the first coordinate information, and further improving the position distribution precision of the data to be processed; and then constructing a second coordinate system according to the coordinate space between adjacent point cloud data in the second coordinate system, determining the obstacle to which the point cloud data belong, and effectively improving the accuracy of the obstacle identification corresponding to the point cloud data, thereby improving the accuracy of identifying the obstacle, avoiding traffic jam caused by incorrect obstacle identification and improving traffic efficiency.

FIG. 4 is a flowchart 400 of one implementation of a point cloud data processing method according to the present disclosure, and referring to FIG. 4, an executing body determines whether the point cloud data is a fence class according to class information of the point cloud data, and if not, directly marks an obstacle type and a corresponding obstacle identifier of the point cloud data according to the class information thereof; extracting point cloud data belonging to fence types as data to be processed, and forming a covariance matrix according to initial coordinate information of the point cloud data in the data to be processed, so as to obtain a first linear equation by fitting, and obtain an initial reference line; then, according to an initial reference line, projecting the point cloud data to the extending direction and the normal vector direction of the initial reference line to obtain coordinate difference values of the point cloud data in two directions respectively; dividing according to the coordinate difference value and the preset length of 5m to obtain a plurality of groups of grouping data; respectively and independently performing straight line fitting on each group of data, determining target data according to the proportion of candidate point cloud data, and then constructing a fitting reference line according to the target data; determining a second coordinate system according to the fitting reference line, and judging the coordinate distance between two adjacent point cloud data according to the second coordinate system, so as to judge whether the fence corresponding to each group of group data is an integer; if yes, directly marking the obstacle type and the corresponding obstacle identifier of the point cloud data in the grouping data; if not, namely the fence corresponding to the grouping data is not a whole, segmenting the group of data at the maximum value of the coordinate interval according to the maximum value of the coordinate interval between two adjacent point cloud data, and performing straight line fitting again according to the segmented data to determine an obstacle identifier corresponding to each point cloud data; and then marking the obstacle type and the corresponding obstacle identification of each point cloud data according to the obstacle type and the corresponding obstacle identification.

The processing method of the point cloud data, provided by the disclosure, is used for solving the problems of perception and passing of the automatic driving vehicle caused by the fence scene. Taking an actual road test scene of an automatic driving BUS project as an example, the problems of abrupt change in size, unstable shape and the like of fences at two sides of a road are solved, so that the problem of sudden braking and sudden braking of an automatic driving vehicle is endless; in addition, the normal operation of the automatic driving vehicle is seriously affected by the situations that the curved fence and the fence are interrupted due to the fence or the fence temporarily laid in municipal construction frequently.

According to the method, through a series of real vehicle tests and scene reproduction, the large fence is segmented into the small fences by utilizing strategies such as large and small, centralized interior point reconstruction, interrupt segmentation fitting and the like, the shape and the identification of the fences are corrected and optimized, and the traffic efficiency problem caused by the fence scene is effectively solved.

In the scheme of the disclosure, the setting of relevant parameters follows the principle of full verification and absolute safety. The scheme disclosed by the invention has a general meaning on the fence scene, and on the premise of safety and reliability, the fence identification and the fence identification are distinguished by adopting the scheme disclosed by the invention, so that the passing efficiency of vehicles in the fence scene is greatly improved, and the closed-loop capacity of the automatic driving vehicles is effectively improved.

As an implementation of the method illustrated in the above figures, fig. 5 illustrates an embodiment of a processing apparatus for point cloud data according to the present disclosure. The processing device 500 of the point cloud data corresponds to the method embodiment shown in fig. 2, and the device can be applied to various electronic devices.

Referring to fig. 5, a processing apparatus 500 for point cloud data according to an embodiment of the present disclosure includes: the system comprises an extraction module 501, a first construction module 502, a first coordinate conversion module 503, a first determination module 504 and a second determination module 505. Wherein, the extracting module 501 is configured to extract data to be processed according to class information of the point cloud data; the first construction module 502 is configured to construct a fitting reference line according to first coordinate information of data to be processed; the first coordinate conversion module 503 is configured to determine a second coordinate system according to the fitting reference line, so as to obtain second coordinate information of the data to be processed; the first determining module 504 is configured to determine, according to the second coordinate information, a coordinate distance between two adjacent point cloud data in the data to be processed; the second determining module 505 is configured to determine the obstacle identifier corresponding to the point cloud data according to the coordinate space between two adjacent point cloud data.

In the processing device 500 for point cloud data according to the embodiment of the present disclosure, the specific processes and the technical effects of the extracting module 501, the first constructing module 502, the first coordinate converting module 503, the first determining module 504 and the second determining module 505 may refer to the relevant descriptions of steps S201 to S205 in the corresponding embodiment of fig. 2, and are not repeated herein.

In some alternative implementations of embodiments of the present disclosure, the first building block 501 includes: a first determination sub-module and a first construction sub-module. The first determining submodule is configured to determine target data according to first coordinate information of data to be processed; the first construction sub-module is configured to construct a fitting reference line from first coordinate information of the target data.

In some optional implementations of embodiments of the present disclosure, the first determining submodule includes: a first determination unit, a dividing unit, and a second determination unit. The first determining unit is configured to determine a coordinate difference between a maximum coordinate value and a minimum coordinate value of the data to be processed in the first target direction according to the first coordinate information; the dividing unit is configured to divide the data to be processed into at least one group of group data according to the coordinate difference and a preset length threshold; the second determining unit is configured to determine target data according to proportions of candidate point cloud data in the respective group data, wherein an absolute value of coordinate values of the candidate point cloud data in a second target direction is smaller than or equal to a preset coordinate threshold, and the second target direction is perpendicular to the first target direction.

In some optional implementations of embodiments of the present disclosure, the partitioning unit is configured to: responding to the coordinate difference being greater than a preset length threshold, dividing the data to be processed according to the preset length threshold and first coordinate information of the data to be processed, and obtaining at least two groups of different grouping data; and dividing the data to be processed into the same group of group data in response to the coordinate difference being smaller than or equal to a preset length threshold.

In some optional implementations of embodiments of the present disclosure, the second determining unit is configured to: determining candidate point cloud data as target data of the group data in response to the proportion of the candidate point cloud data in the group data being greater than or equal to a preset proportion threshold; and determining all the point cloud data in the group data as target data in response to the proportion of the candidate point cloud data in the group data being smaller than a preset proportion threshold.

In some optional implementations of embodiments of the present disclosure, the first coordinate conversion module is configured to: determining a second coordinate system by taking the extending direction and the normal vector direction of the fitting reference line as coordinate axes; and determining second coordinate information of the data to be processed in the second coordinate system.

In some optional implementations of embodiments of the present disclosure, the second determining module is configured to: determining that the two adjacent point cloud data respectively have different obstacle identifications in response to the coordinate spacing between the two adjacent point cloud data being greater than a preset spacing threshold; and determining that the two adjacent point cloud data have corresponding obstacle identifications in response to the coordinate spacing between the two adjacent point cloud data being smaller than or equal to a preset spacing threshold.

In some optional implementations of embodiments of the present disclosure, the category information includes at least one of: obstacle category, obstacle identification, and obstacle specification.

In some optional implementations of the embodiments of the present disclosure, the processing apparatus of point cloud data further includes: the system comprises a second construction module and a second coordinate conversion module. The second construction module is configured to construct an initial reference line according to initial coordinate information of the data to be processed; the second coordinate conversion module is configured to determine a first coordinate system according to the initial reference line to obtain first coordinate information of data to be processed.

In the processing device for point cloud data in the embodiment of the present disclosure, the specific processing of the second construction module and the second coordinate conversion module and the technical effects thereof may refer to the description related to steps S302 to S303 in the corresponding embodiment of fig. 3, and are not described herein again.

In some optional implementations of embodiments of the present disclosure, the second coordinate conversion module is configured to: determining a first coordinate system by taking the extending direction and the normal vector direction of an initial reference line as coordinate axes; first coordinate information of data to be processed in a first coordinate system is determined.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

Fig. 6 illustrates a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 6, the apparatus 600 includes a computing unit 601 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 may also be stored. The computing unit 601, ROM 602, and RAM 603 are connected to each other by a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

Various components in the device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, mouse, etc.; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the respective methods and processes described above, for example, a processing method of point cloud data. For example, in some embodiments, the method of processing point cloud data may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the above-described processing method of point cloud data may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the processing method of the point cloud data in any other suitable way (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel or sequentially or in a different order, provided that the desired results of the technical solutions of the present disclosure are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A method of processing point cloud data, comprising:

constructing a fitting reference line according to the first coordinate information of the data to be processed;

determining coordinate intervals between two adjacent point cloud data in the data to be processed according to the second coordinate information;

And determining an obstacle identifier corresponding to the point cloud data according to the coordinate distance between the two adjacent point cloud data.

2. The method of claim 1, wherein the constructing a fitting reference line from the first coordinate information of the data to be processed comprises:

determining target data according to the first coordinate information of the data to be processed;

and constructing a fitting reference line according to the first coordinate information of the target data.

3. The method of claim 2, wherein the determining target data according to the first coordinate information of the data to be processed comprises:

determining a coordinate difference between a maximum coordinate value and a minimum coordinate value of the data to be processed in a first target direction according to the first coordinate information;

dividing the data to be processed into at least one group of group data according to the coordinate difference and a preset length threshold;

and respectively determining target data according to the proportion of the candidate point cloud data in each group data, wherein the absolute value of the coordinate value of the candidate point cloud data in a second target direction is smaller than or equal to a preset coordinate threshold value, and the second target direction is perpendicular to the first target direction.

4. A method according to claim 3, wherein the dividing the data to be processed into at least one set of packet data according to the coordinate difference and a preset length threshold comprises:

dividing the data to be processed according to the preset length threshold and the first coordinate information of the data to be processed to obtain at least two groups of different grouping data in response to the coordinate difference being greater than the preset length threshold;

and dividing the data to be processed into the same group of data in response to the coordinate difference being smaller than or equal to a preset length threshold.

5. The method of claim 3, wherein the determining the target data according to the proportions of the candidate point cloud data in each of the group data includes:

determining the candidate point cloud data as target data of the group data in response to the proportion of the candidate point cloud data in the group data being greater than or equal to a preset proportion threshold;

and determining all the point cloud data in the group data as target data in response to the proportion of the candidate point cloud data in the group data being smaller than the preset proportion threshold.

6. The method according to claim 1, wherein the determining a second coordinate system according to the fitting reference line, to obtain second coordinate information of the data to be processed, includes:

Determining a second coordinate system by taking the extending direction and the normal vector direction of the fitting reference line as coordinate axes;

and determining second coordinate information of the data to be processed in the second coordinate system.

7. The method of claim 1, wherein the determining, according to the coordinate space between the two adjacent point cloud data, the obstacle identifier corresponding to the point cloud data includes:

determining that the two adjacent point cloud data respectively have different obstacle identifications in response to the coordinate spacing between the two adjacent point cloud data being greater than a preset spacing threshold;

and determining that the two adjacent point cloud data have the same obstacle identifier in response to the coordinate spacing between the two adjacent point cloud data being less than or equal to the preset spacing threshold.

8. The method of any of claims 1-7, the category information comprising at least one of: obstacle category, obstacle identification, and obstacle specification.

9. The method of claim 8, wherein the obstacle category comprises at least one of: fence and enclosure.

10. The method of any of claims 1-9, further comprising:

Constructing an initial reference line according to the initial coordinate information of the data to be processed;

and determining a first coordinate system according to the initial reference line to obtain first coordinate information of the data to be processed.

11. The method of claim 10, wherein determining a first coordinate system according to the initial reference line, to obtain first coordinate information of the data to be processed, includes:

determining a first coordinate system by taking the extending direction and the normal vector direction of the initial reference line as coordinate axes;

and determining first coordinate information of the data to be processed in the first coordinate system.

12. A processing apparatus for point cloud data, comprising:

the first construction module is configured to construct a fitting reference line according to first coordinate information of the data to be processed;

the first coordinate conversion module is configured to determine a second coordinate system according to the fitting reference line to obtain second coordinate information of the data to be processed;

13. The apparatus of claim 12, wherein the first build module comprises:

the first determining submodule is configured to determine target data according to first coordinate information of the data to be processed;

and the first construction submodule is configured to construct a fitting reference line according to the first coordinate information of the target data.

14. The apparatus of claim 13, wherein the first determination submodule comprises:

a first determining unit configured to determine a coordinate difference between a maximum coordinate value and a minimum coordinate value of the data to be processed in a first target direction, based on the first coordinate information;

the dividing unit is configured to divide the data to be processed into at least one group of group data according to the coordinate difference and a preset length threshold;

and a second determining unit configured to determine target data according to a proportion of candidate point cloud data in each group data, wherein an absolute value of a coordinate value of the candidate point cloud data in a second target direction is smaller than or equal to a preset coordinate threshold, and the second target direction is perpendicular to the first target direction.

15. The apparatus of claim 14, wherein the partitioning unit is configured to:

16. The apparatus of claim 14, wherein the second determination unit is configured to:

17. The apparatus of claim 12, wherein the first coordinate conversion module is configured to:

18. The apparatus of claim 12, wherein the second determination module is configured to:

19. The apparatus of any of claims 12-18, wherein the category information comprises at least one of: obstacle category, obstacle identification, and obstacle specification.

20. The apparatus of claim 19, wherein the obstacle category comprises at least one of: fence and enclosure.

21. The apparatus of any of claims 12-20, further comprising:

the second construction module is configured to construct an initial reference line according to the initial coordinate information of the data to be processed;

and the second coordinate conversion module is configured to determine a first coordinate system according to the initial reference line to obtain first coordinate information of the data to be processed.

22. The apparatus of claim 21, wherein the second coordinate conversion module is configured to:

23. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-11.

24. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-11.

25. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-11.