Method and device for generating simulated non-standard structured scene semantic point cloud data

Abstract

The present disclosure relates to a method and device for generating simulated non-standard structured scene semantic point cloud data, the method comprising: obtaining feature attribute information of a non-standard structured scene input by a user; converting the feature attribute information into semantic terrain data; and generating non-standard structured scene point cloud data from the semantic terrain data through a simulated laser radar model, the non-standard structured scene point cloud data carrying semantic information. This can generate corresponding non-standard structured scene point cloud data from the input of the user, thereby avoiding the high acquisition cost caused by collecting point cloud data from a real scene. Furthermore, since the non-standard structured scene point cloud data carries semantic information, the obtained non-standard structured scene point cloud data can be used for model training.

Description

Technical Field

[0001] This disclosure relates to the field of point cloud technology, and in particular to a method and apparatus for generating semantic point cloud data for simulating non-standard structured scenes. Background Technology

[0002] With the continuous development of autonomous driving technology in mining scenarios, a large amount of point cloud data from mining scenarios is needed to provide training samples for model training, resulting in a large demand for point cloud samples based on mining scenarios.

[0003] In order to obtain a sufficient number of point cloud samples, related technologies require point cloud data collection from different mining scenarios with varying operational types. However, the complexity and variability of mining scenarios result in significant costs and low efficiency when obtaining a large number of point cloud samples. Summary of the Invention

[0004] This disclosure provides a method and apparatus for generating semantic point cloud data for simulating non-standard structured scenes.

[0005] According to a first aspect of this disclosure, a method for generating semantic point cloud data of simulated non-standard structured scenes is provided, the method comprising:

[0006] Obtain feature attribute information of non-standard structured scenarios input by the user;

[0007] The feature attribute information is converted into semantic terrain data;

[0008] The semantic terrain data is used to generate non-standard structured scene point cloud data by simulating a lidar model. The non-standard structured scene point cloud data carries semantic information.

[0009] Optionally, converting the feature attribute information into semantic terrain data includes:

[0010] The feature attribute information is converted into geometric feature data, which includes scene elements of different categories;

[0011] A semantic index is constructed, and the scene elements in the geometric feature data are semantically annotated using the semantic index to obtain semantic terrain data.

[0012] Optionally, constructing the semantic index includes:

[0013] Establish cached data for semantic indexes, wherein the cached data for semantic indexes includes the mapping relationship between semantic information and scene elements.

[0014] Optionally, the step of semantically annotating the scene elements in the geometric feature data using the semantic index includes:

[0015] The geometric feature data is divided into terrain blocks by reverse calculation of geometric features to obtain multiple terrain block data;

[0016] The semantic index is used to determine whether cross-domain data exists in the multiple data blocks.

[0017] If there is no cross-domain data in the multiple data blocks, semantic annotation is performed on the multiple data blocks.

[0018] Optionally, the method further includes:

[0019] In the case where cross-domain data exists among the multiple data blocks, the cross-domain data is semantically partitioned to obtain semantically partitioned data blocks.

[0020] Optionally, the scene elements include at least one of roads, intersections, retaining walls, loading areas, waiting areas, queuing areas, and spoil heaps.

[0021] Optionally, the simulated lidar model is used to simulate the performance parameters of the target type radar.

[0022] Optionally, the non-standard structured scenario includes a mining scenario.

[0023] According to a second aspect of this disclosure, an apparatus for generating semantic point cloud data of simulated non-standard structured scenes is provided, the apparatus comprising:

[0024] The information acquisition module is used to acquire feature attribute information of non-standard structured scenarios input by the user;

[0025] The conversion module is used to convert the feature attribute information into semantic terrain data;

[0026] The point cloud data generation module is used to generate non-standard structured scene point cloud data from the semantic terrain data using a simulated LiDAR model. The non-standard structured scene point cloud data carries semantic information.

[0027] Optionally, the conversion module is specifically used for:

[0028] The feature attribute information is converted into geometric feature data, which includes scene elements of different categories;

[0029] A semantic index is constructed, and the scene elements in the geometric feature data are semantically annotated using the semantic index to obtain semantic terrain data.

[0030] Optionally, the conversion module is further configured to:

[0031] Establish cached data for semantic indexes, wherein the cached data for semantic indexes includes the mapping relationship between semantic information and scene elements.

[0032] Optionally, the conversion module is further configured to:

[0033] The geometric feature data is divided into terrain blocks by reverse calculation of geometric features to obtain multiple terrain block data;

[0034] The semantic index is used to determine whether cross-domain data exists in the multiple data blocks.

[0035] If there is no cross-domain data in the multiple data blocks, semantic annotation is performed on the multiple data blocks.

[0036] Optionally, the device further includes:

[0037] The semantic segmentation module is used to perform semantic segmentation on the cross-domain data when there is cross-domain data in the multiple data blocks, so as to obtain semantically segmented data blocks.

[0038] Optionally, the scene elements include at least one of roads, intersections, retaining walls, loading areas, waiting areas, queuing areas, and spoil heaps.

[0039] Optionally, the simulated lidar model is used to simulate the performance parameters of the target type radar.

[0040] Optionally, the non-standard structured scenario includes a mining scenario.

[0041] According to a third aspect of this disclosure, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described above.

[0042] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the methods described above.

[0043] The method and apparatus for generating semantic point cloud data of simulated non-standard structured scenes provided in this disclosure acquire feature attribute information of a non-standard structured scene input by a user, and convert the feature attribute information into semantic terrain data. Then, using a simulated LiDAR model, the semantic terrain data is used to generate non-standard structured scene point cloud data, which carries semantic information. This allows for the generation of corresponding non-standard structured scene point cloud data from user input, thereby avoiding the high acquisition costs associated with collecting point cloud data from real scenes. Furthermore, because the non-standard structured scene point cloud data carries semantic information, it is convenient to use the obtained non-standard structured scene point cloud data for model training. Attached Figure Description

[0044] Further details, features, and advantages of this disclosure are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0045] Figure 1 A flowchart of a method for generating semantic point cloud data of a simulated non-standard structured scene provided as an exemplary embodiment of this disclosure;

[0046] Figure 2 A schematic block diagram of the functional modules of a device for generating semantic point cloud data of simulated non-standard structured scenes provided for an exemplary embodiment of this disclosure;

[0047] Figure 3 A structural block diagram of an electronic device provided as an exemplary embodiment of this disclosure;

[0048] Figure 4 A block diagram of a computer system provided for an exemplary embodiment of this disclosure. Detailed Implementation

[0049] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0050] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0051] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. It should be noted that the concepts of "first", "second", etc., used in this disclosure are only used to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0052] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0053] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0054] With the continuous development of autonomous driving technology in mining scenarios, semantic point cloud samples are a common data dependency for deep learning algorithm training during the development of unmanned mining algorithms, and the demand is substantial. Currently, the general method for acquiring semantic point cloud samples involves first collecting raw point cloud data of the actual scene using sensors such as LiDAR, and then manually selecting and labeling semantically informative portions from the massive point cloud data using standard software tools. This method can be considered a forward acquisition approach, but it has significant limitations in both the data acquisition and point cloud labeling stages, as detailed below.

[0055] First, the test scenarios need to be diverse, encompassing both categorical diversity and diversity of random variations. For non-standard structured scenarios like mines, the categories include different operational scenarios derived from mining, transportation, and spoil disposal processes (including multiple areas, intersections, lanes, and scenario progression changes). In terms of randomness, the scenarios also need to reflect the diversity of combinations based on different road conditions such as curves, slopes, and bumpiness. According to these requirements, there could be tens of thousands or even more such scenarios, and not all of them can necessarily be matched to the actual needs in reality. This greatly increases the difficulty of acquiring raw point cloud data. Acquiring all of them using a forward acquisition method would incur enormous costs, and the output efficiency would not meet the overall R&D needs of unmanned mining operations.

[0056] Secondly, considering the diverse scenario requirements mentioned above, semantic annotation is needed for point cloud data collected in different scenarios, and the algorithm model iteration process requires massive amounts of sample data for training. This situation also presents two problems. On the one hand, based on the amount of raw data and annotation requirements, manual annotation is extremely inefficient and has high labor costs. On the other hand, it is difficult to achieve good consistency in production standards for manually annotated sample data, which may lead to unreliable sample data quality and directly negatively impact the training of the algorithm model.

[0057] To address the aforementioned technical issues, this embodiment of the present disclosure generates terrain data with open-pit mine terrain features through reverse calculation based on the input of geometric and attribute information, and forms a set of matching semantic information index data, which is then integrated and output as semantic terrain data. Based on the semantic terrain data, point cloud data with semantic information can be quickly generated by simulating LiDAR in real time, thereby efficiently obtaining sample data for deep learning training.

[0058] In the embodiments provided in this disclosure, in order to obtain cloud samples corresponding to non-standard structured scenes, it is not necessary to collect point cloud data of real non-standard structured scenes. Instead, users can construct simulated non-standard structured scenes using scene elements, thereby generating corresponding semantic terrain data. Simulated LiDAR can then be used to generate non-standard structured scene point cloud data with semantic information from the semantic terrain data.

[0059] In the embodiments provided in this disclosure, a method for generating semantic point cloud data for simulating non-standard structured scenes is provided, such as... Figure 1 As shown, the method may include the following steps:

[0060] In step S110, the feature attribute information of the non-standard structured scene input by the user is obtained.

[0061] In the embodiments provided in this disclosure, an internal data structure can be pre-constructed, parameterized by the feature attribute information of various scene elements that may appear in non-standard structured scenarios, and each scene element in the non-standard structured scenario can be visualized as corresponding geometric feature data, converted into a data set with type identifiers, and data classification and feature transformation can be completed.

[0062] For example, scene elements appearing in non-standard structured scenes can include roads, ramps, various open areas, and retaining wall boundaries. For instance, a "road" can be visualized as its geometric shape or form, and its color, size, or type can be adjusted by modifying its attribute parameters. This allows commonly used or all possible scene elements in non-standard structured scenes to be visualized as specific combined shapes or geometric forms, making it more intuitive for users to combine these scene elements into the desired non-standard structured scene.

[0063] In this embodiment, a non-standard structured scenario, specifically a mining scenario, is used as an example for illustration.

[0064] In step S120, the feature attribute information is converted into semantic terrain data.

[0065] In this embodiment, upon receiving feature attribute information of a non-standard structured scene input by a user, the system can obtain the corresponding geometric shape or graphic based on the data structure constructed above, and visually present the scene elements corresponding to these feature attribute information in a geometric shape or graphic manner. Furthermore, the system can receive user adjustments to the size, color, or type of these geometric shapes or graphics, ultimately obtaining semantic terrain data.

[0066] It should be noted that because these scene elements carry semantic information, the resulting semantic terrain data will also carry semantic information. For example, the geometric shape or form of a "road" will carry road semantic information. Specifically, semantic annotation can be used to semantically annotate the semantic terrain data.

[0067] In step S130, non-standard structured scene point cloud data is generated from semantic terrain data using a simulated LiDAR model. This non-standard structured scene point cloud data carries semantic information.

[0068] In the embodiments provided in this disclosure, based on semantic terrain data, the semantic point cloud production stage can be entered. This stage mainly relies on a lidar mathematical model and a parallel computing module to simulate the radar acquisition process. The lidar mathematical model is used to simulate the performance parameters of a real radar to ensure that the point cloud data generated under this model is consistent with the real radar in terms of data volume and characteristics. The parallel computing module ensures computational efficiency, improving the timeliness of point cloud data generation. With the support of these two modules, combined with the semantic information in the terrain data, point cloud data that conforms to the characteristics of the lidar model and includes semantic information can ultimately be output.

[0069] The method for generating semantic point cloud data of simulated non-standard structured scenes provided in this disclosure acquires the feature attribute information of a non-standard structured scene input by the user and converts the feature attribute information into semantic terrain data. Then, it generates non-standard structured scene point cloud data using a simulated LiDAR model. This non-standard structured scene point cloud data carries semantic information. This allows for the generation of corresponding non-standard structured scene point cloud data based on user input, thereby avoiding the high acquisition costs associated with collecting point cloud data from real scenes. Furthermore, because the non-standard structured scene point cloud data carries semantic information, it is convenient to use the obtained non-standard structured scene point cloud data for model training.

[0070] Based on the above embodiments, in another embodiment provided in this disclosure, step S120 may further include the following steps:

[0071] S121, convert the feature attribute information into geometric feature data. This geometric feature data includes scene elements of different categories.

[0072] In the embodiments, during the process of converting feature attribute information into geometric feature data, the geometric feature data can specifically be the geometric shape or geometric figure in the above embodiments.

[0073] S122, Construct a semantic index, and use the semantic index to semantically annotate the scene elements in the geometric feature data to obtain semantic terrain data.

[0074] In this embodiment, a semantic index database can be constructed to establish a mapping relationship between semantic information and scene elements. Specifically, this semantic index database may include the data structure described in the above embodiment. During the process of semantically annotating scene elements in geometric feature data using the constructed semantic index, frequently used or periodic semantic indexes can be stored in the semantic index cache data.

[0075] In addition, the embodiment can also establish cached data of semantic index, which includes the mapping relationship between semantic information and scene elements.

[0076] Based on the above embodiments, in another embodiment provided in this disclosure, the scene elements in the geometric feature data can be semantically annotated using semantic indexing through the following steps, that is, step S122 above may further include the following steps:

[0077] S21, the geometric feature data is divided into terrain blocks by reverse calculation of geometric features to obtain multiple terrain block data.

[0078] Taking a non-standard structured scenario, specifically a mining scenario, as an example, in the embodiments provided in this disclosure, after the user inputs the feature attribute information of the above-mentioned mining scenario, by converting these feature attribute information into corresponding scene elements with geometric features, a visualized mining scenario, also known as a simulated mining scenario, can be obtained. The mining scenario may specifically include roads, loading areas, waiting areas, excavation areas, etc., and the number of roads and each loading area can be multiple.

[0079] Therefore, when semantically annotating scene elements in geometric feature data, the feature data can be divided into terrain blocks, and terrain with the same attributes can be divided into the same block. This can result in multiple terrain block data, and the terrain data in the same block has the same speech information, which makes it easier to semantically annotate scene elements in geometric feature data.

[0080] In the embodiments, scene elements may include at least one of roads, intersections, retaining walls, loading areas, waiting areas, queuing areas, and spoil heaps.

[0081] S22, determine whether cross-domain data exists in multiple data blocks through semantic indexing.

[0082] S23, when there is no cross-domain data in multiple data blocks, perform semantic annotation on multiple data blocks.

[0083] Taking a non-standard structured scenario, specifically a mining scenario, as an example, in the embodiment, semantic indexing can be used to determine whether there is cross-domain data in the block data, that is, whether there is different types of data in a certain block of data, such as dividing the loading area and the waiting area in the mining scenario together.

[0084] When cross-domain data exists in multiple data blocks, semantic partitioning is performed on the cross-domain data to obtain semantically partitioned data blocks.

[0085] In this way, when there is no cross-domain data in multiple data blocks, semantic annotation is performed on multiple data blocks, and each data block corresponds to various semantic information.

[0086] The method provided in this disclosure can quickly obtain terrain data with semantic information by receiving vector data input from a user. Then, based on the semantic terrain data, combined with a lidar mathematical model and a parallel computing module, point cloud data with semantic information can be quickly and directly obtained. This not only greatly improves the efficiency of point cloud data generation, but also solves the problems of inefficiency and lack of standardization in manually labeling large amounts of point clouds.

[0087] By dividing each function into corresponding functional modules, this disclosure provides a device for generating simulated non-standard structured scene semantic point cloud data. This device can be a server or a chip applied to a server. Figure 2 A schematic block diagram of the functional modules of a device for generating semantic point cloud data of simulated non-standard structured scenes, provided as an exemplary embodiment of this disclosure. Figure 2 As shown, the device for generating semantic point cloud data for simulated non-standard structured scenes includes:

[0088] Information acquisition module 10 is used to acquire feature attribute information of non-standard structured scenes input by the user;

[0089] The conversion module 20 is used to convert the feature attribute information into semantic terrain data;

[0090] The point cloud data generation module 30 is used to generate non-standard structured scene point cloud data from the semantic terrain data using a simulated lidar model. The non-standard structured scene point cloud data carries semantic information.

[0091] In another embodiment provided in this disclosure, the conversion module is specifically used for:

[0092] The feature attribute information is converted into geometric feature data, which includes scene elements of different categories;

[0093] A semantic index is constructed, and the scene elements in the geometric feature data are semantically annotated using the semantic index to obtain semantic terrain data.

[0094] In another embodiment provided in this disclosure, the conversion module is further configured to:

[0095] Establish cached data for semantic indexes, wherein the cached data for semantic indexes includes the mapping relationship between semantic information and scene elements.

[0096] In another embodiment provided in this disclosure, the conversion module is further configured to:

[0097] The geometric feature data is divided into terrain blocks by reverse calculation of geometric features to obtain multiple terrain block data;

[0098] The semantic index is used to determine whether cross-domain data exists in the multiple data blocks.

[0099] If there is no cross-domain data in the multiple data blocks, semantic annotation is performed on the multiple data blocks.

[0100] In yet another embodiment provided in this disclosure, the apparatus further includes:

[0101] The semantic segmentation module is used to perform semantic segmentation on the cross-domain data when there is cross-domain data in the multiple data blocks, so as to obtain semantically segmented data blocks.

[0102] In another embodiment provided in this disclosure, the scene elements include at least one of roads, intersections, retaining walls, loading areas, waiting areas, queuing areas, and spoil heaps.

[0103] In another embodiment provided in this disclosure, the simulated lidar model is used to simulate the performance parameters of the target type radar.

[0104] In yet another embodiment provided in this disclosure, the non-standard structured scenario includes a mining scenario.

[0105] For details regarding the apparatus, please refer to the descriptions corresponding to the above method embodiments; they will not be repeated here.

[0106] The apparatus for generating semantic point cloud data of simulated non-standard structured scenes provided in this disclosure acquires feature attribute information of a non-standard structured scene input by a user and converts the feature attribute information into semantic terrain data. Then, it generates non-standard structured scene point cloud data using a simulated LiDAR model. This non-standard structured scene point cloud data carries semantic information. This allows for the generation of corresponding non-standard structured scene point cloud data based on user input, thereby avoiding the high acquisition costs associated with collecting point cloud data from real-world scenes. Furthermore, because the non-standard structured scene point cloud data carries semantic information, it is convenient to use the obtained non-standard structured scene point cloud data for model training.

[0107] In another embodiment provided in this disclosure, the simulated lidar model is used to simulate the performance parameters of a target type radar. This disclosure also provides an electronic device, including: at least one processor; a memory for storing executable instructions of the at least one processor; wherein the at least one processor is configured to execute the instructions to implement the methods disclosed in this disclosure.

[0108] Figure 3 This is a schematic diagram of the structure of an electronic device provided as an exemplary embodiment of this disclosure. For example... Figure 3 As shown, the electronic device 1800 includes at least one processor 1801 and a memory 1802 coupled to the processor 1801. The processor 1801 can perform the corresponding steps in the methods disclosed in the embodiments of this disclosure.

[0109] The processor 1801 described above can also be called a central processing unit (CPU), which can be an integrated circuit chip with signal processing capabilities. Each step in the method disclosed in this embodiment can be implemented by the integrated logic circuitry in the processor 1801 or by software instructions. The processor 1801 can be a general-purpose processor, a digital signal processor (DSP), an ASIC, a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this embodiment can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can be located in the memory 1802, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor 1801 reads information from the memory 1802 and, in conjunction with its hardware, completes the steps of the method described above.

[0110] Furthermore, various operations / processes according to this disclosure, implemented via software and / or firmware, can be transmitted from a storage medium or network to a computer system with a dedicated hardware architecture, such as... Figure 4 The computer system 1900 shown is equipped with the programs that constitute the software. When various programs are installed, the computer system is able to perform various functions, including those described above. Figure 4 A block diagram of a computer system provided for an exemplary embodiment of this disclosure.

[0111] Computer System 1900 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0112] like Figure 4As shown, the computer system 1900 includes a computing unit 1901, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 1902 or a computer program loaded from a storage unit 1908 into a random access memory (RAM) 1903. The RAM 1903 may also store various programs and data required for the operation of the computer system 1900. The computing unit 1901, ROM 1902, and RAM 1903 are interconnected via a bus 1904. An input / output (I / O) interface 1905 is also connected to the bus 1904.

[0113] Multiple components in computer system 1900 are connected to I / O interface 1905, including: input unit 1906, output unit 1907, storage unit 1908, and communication unit 1909. Input unit 1906 can be any type of device capable of inputting information into computer system 1900. Input unit 1906 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of the electronic device. Output unit 1907 can be any type of device capable of presenting information and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 1908 may include, but is not limited to, hard disks and optical disks. Communication unit 1909 allows computer system 1900 to exchange information / data with other devices via a network such as the Internet, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0114] The computing unit 1901 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose 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 1901 performs the various methods and processes described above. For example, in some embodiments, the methods disclosed in this disclosure can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1908. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 1900 via ROM 1902 and / or communication unit 1909. In some embodiments, the computing unit 1901 can be configured to perform the methods disclosed in this disclosure by any other suitable means (e.g., by means of firmware).

[0115] This disclosure also provides a computer-readable storage medium, wherein when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is able to perform the methods disclosed in this disclosure.

[0116] The computer-readable storage medium in this disclosure can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. The aforementioned computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specifically, the aforementioned computer-readable storage medium may include electrical connections based on one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0117] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0118] This disclosure also provides a computer program product, including a computer program, wherein the computer program, when executed by a processor, implements the methods disclosed in the embodiments of this disclosure.

[0119] In embodiments of this disclosure, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network (including a local area network (LAN) or a wide area network (WAN)), or it can be connected to an external computer.

[0120] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0121] The modules, components, or units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules, components, or units do not necessarily constitute a limitation on the module, component, or unit itself.

[0122] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary hardware logic components that can be used include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0123] The above description is merely an embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0124] While specific embodiments of this disclosure have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A method for generating simulated non-standard structured scene semantic point cloud data, characterized in that, The method comprises: obtaining feature attribute information of a non-standard structured scene input by a user; converting the feature attribute information into geometric feature data, the geometric feature data comprising scene elements of different categories; establishing semantic index cache data, the semantic index cache data comprising a mapping relationship between semantic information and scene elements, and performing terrain blocking on the geometric feature data through geometric feature reverse calculation to obtain a plurality of block data; determining whether there is cross-domain data in the plurality of block data through the semantic index; in the case where there is no cross-domain data in the plurality of block data, performing semantic labeling on the plurality of block data to obtain semantic terrain data; generating non-standard structured scene point cloud data carrying semantic information through a simulated laser radar model.

2. The method of claim 1, wherein, The method further comprises: in the case where there is cross-domain data in the plurality of block data, performing semantic division on the cross-domain data to obtain block data after semantic division.

3. The method of claim 1, wherein, The scene elements comprise at least one of a road, an intersection, a retaining wall, a loading area, a waiting area, a queuing area, and a dump.

4. The method according to any one of claims 1 to 3, characterized in that, The simulated laser radar model is used to simulate performance parameters of a target model radar.

5. The method according to any one of claims 1 to 3, characterized in that, The non-standard structured scene comprises a mine scene.

6. An apparatus for generating simulated non-standard structured scene semantic point cloud data, the apparatus comprising: The device comprises: an information acquisition module configured to obtain feature attribute information of a non-standard structured scene input by a user; a conversion module configured to convert the feature attribute information into geometric feature data, the geometric feature data comprising scene elements of different categories; establish semantic index cache data, the semantic index cache data comprising a mapping relationship between semantic information and scene elements, and perform terrain blocking on the geometric feature data through geometric feature reverse calculation to obtain a plurality of block data; determine whether there is cross-domain data in the plurality of block data through the semantic index; in the case where there is no cross-domain data in the plurality of block data, perform semantic labeling on the plurality of block data to obtain semantic terrain data; a point cloud data generation module configured to generate non-standard structured scene point cloud data carrying semantic information through a simulated laser radar model.

7. An electronic device, comprising: comprises: at least one processor; a memory for storing instructions executable by the at least one processor; wherein the at least one processor is configured to execute the instructions to implement the method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, When the instructions in the computer readable storage medium are executed by the processor of the electronic device, the electronic device can perform the method of any one of claims 1-5.

Images