Scene synthesis data generation method, computer device and readable storage medium

Abstract

The present invention discloses a scene synthesis data generation method, a computer device and a readable storage medium, and relates to the field of computer technology. The scene synthesis data generation method of the present invention comprises: obtaining preset basic parameter information, generating a basic terrain based on terrain parameters, generating a natural ecology on the basic terrain based on natural ecology information, and generating humanistic building facilities on the basic terrain based on humanistic facilities information and algorithm model information; obtaining a synthesized scene based on the basic terrain, the natural ecology and the humanistic building facilities; obtaining scene synthesis data based on the synthesized scene, so that the algorithm model is trained based on the scene synthesis data. The present invention realizes the rapid and efficient batch production of scene synthesis data, and greatly improves the richness and generalization of the scene while ensuring the rationality of the synthesis data.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a scene synthesis data generation method, a computer device and a readable storage medium. Background Art

[0002] With the advancement of technology, the application scenarios of artificial intelligence are constantly expanding, especially in the fields of smart cities, environmental protection, emergency rescue, etc. The use of algorithm models that realize artificial intelligence in combination with specific application fields will be more extensive. The improvement of the application performance of algorithm models in various scenarios is closely related to the training data set. In the prior art, the acquisition of scene data for algorithm model training usually uses a deep learning model to use raw sensor data (such as multi-channel camera images, laser point clouds) as input to identify scene structure information (including terrain, vegetation, water bodies, etc. in the natural ecological environment). It can be seen that the traditional virtual scene generation method is inefficient, with a small data set, and it is difficult to improve the generalization and robustness of algorithm models used in various scenarios. In particular, for algorithm models used in application scenarios such as automatic cruising and disaster relief, due to the complexity and variability of the natural ecological environment, the scene image data has a problem of sample scarcity. The main means currently adopted to address this problem include image affine transformation, image mirror inversion, horizontal inversion, etc. These methods can only increase the scale of training sets and test sets to a limited extent, and the complexity of training data is difficult to approach the real application scenario.

[0003] Therefore, how to obtain batch and reasonable scene data for algorithm model training has become an urgent problem that needs to be solved. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the scene synthesis data generation method, computer device and readable storage medium provided by the present invention can solve or at least partially solve the problems of scarcity of existing scene data used for algorithm model training and poor matching between scene synthesis data and actual application scenarios.

[0005] In a first aspect, the present invention provides a method for generating scene synthetic data, wherein the synthetic data is applied to the training of an algorithm model, comprising:

[0006] Obtaining preset basic parameter information; the basic parameter information includes terrain parameters and scene parameters; the scene parameters include algorithm model information, cultural facilities information and natural ecology information;

[0007] generating a basic terrain based on the terrain parameters;

[0008] generating a natural ecology on the basic terrain based on the natural ecology information;

[0009] Generate humanistic building facilities on the basic terrain based on the humanistic facility information and the algorithm model information;

[0010] Obtaining a synthetic scene based on the basic terrain, the natural ecology and the cultural architectural facilities;

[0011] Scene synthesis data is obtained based on the synthesized scene, so that the algorithm model is trained based on the scene synthesis data.

[0012] In some technical solutions, the algorithm model information includes the purpose of algorithm model training and the starting point and end point coordinates used for training;

[0013] The cultural facilities information includes cultural facilities parameters and cultural facilities construction rules; the generation of cultural building facilities on the basic terrain based on the cultural facilities information and algorithm model information includes:

[0014] Determine a target area in the base terrain based on the algorithm model information;

[0015] Based on the cultural facility parameters and the cultural facility construction rules, generating cultural building facilities in the target area;

[0016] And / or, the natural ecological information includes ecological type, natural ecological parameters and natural ecological construction rules, and the generating of natural ecology on the basic terrain based on the natural ecological information includes:

[0017] Based on the ecological type, the natural ecological parameters and the natural ecological construction rules, a natural ecology is generated on the basic terrain.

[0018] In some technical solutions, generating a humanistic building facility in the target area based on the humanistic facility parameter and the humanistic facility construction rule includes:

[0019] Based on the geographical environment parameters of the target area, the cultural facility parameters and the cultural facility construction rules, at least one first-level cultural building facility sampling point is set in the target area, and a first-level cultural building facility is placed at the first-level cultural building facility sampling point;

[0020] And / or, generating a natural ecology on the basic terrain based on the ecological type, the natural ecological parameters and the natural ecological construction rules includes:

[0021] At least one first-level natural ecological sampling point is set in the basic terrain based on the geographical environment parameters of the basic terrain, the ecological type, the natural ecological parameters and the natural ecological construction rules, and a first-level natural ecology is placed at the first-level natural ecological sampling point.

[0022] In some technical solutions, the geographical environment parameter includes a ground slope parameter; and the step of setting at least one first-level humanistic building facility sampling point in the target area based on the geographical environment parameter of the target area, the humanistic facility parameter and the humanistic facility construction rule includes:

[0023] Select the area in the target area where the ground slope parameter is less than the preset slope parameter as the sampling area for humanistic building facilities;

[0024] At least one first-level humanistic building facility sampling point is set in the humanistic building facility sampling area according to the humanistic facility parameters and the humanistic facility construction rules.

[0025] In some technical solutions, the geographical environment parameters include ground material parameters; the geographical environment parameters based on the basic terrain, the ecological type, the natural ecological parameters and the natural ecological construction rules are used to set at least one primary natural ecological sampling point in the basic terrain, including:

[0026] Based on the ground material parameters, different ground materials included in the basic terrain are determined; according to the ecological type, the natural ecological parameters and the natural ecological construction rules, the sampling intervals used for different ground materials are determined; and according to the sampling intervals, at least one first-level natural ecological sampling point is set in the basic terrain.

[0027] In some technical solutions, the method further comprises:

[0028] Based on the first-level humanistic building facilities placed at the first-level humanistic building facility sampling points, generating second-level humanistic building facility sampling points around the first-level humanistic building facility sampling points, and placing second-level humanistic building facilities at the second-level humanistic building facility sampling points;

[0029] And / or, based on the primary natural ecology placed at the primary natural ecological sampling point, a secondary natural ecological sampling point is generated around the primary natural ecological sampling point, and a secondary natural ecology is placed at the secondary natural ecological sampling point.

[0030] In some technical solutions, the method also includes a step of transforming the position of the first-level cultural building facility sampling points, and / or the second-level cultural building facility sampling points, and / or the first-level natural ecological sampling points, and / or the second-level natural ecological sampling points to change the sampling point attributes, and screening according to preset conditions after changing the sampling point attributes, wherein the position transformation includes at least one of translation, rotation and scaling.

[0031] In some technical solutions, obtaining preset basic parameter information includes:

[0032] Obtaining a data generalization range corresponding to at least one of the algorithm model information, cultural facilities information, and natural ecological information;

[0033] At least one of the algorithm model information, the cultural facilities information and the natural ecological information is determined based on the data generalization scope.

[0034] In a second aspect, the present invention provides a computer device comprising a processor and a memory, wherein the memory is suitable for storing multiple program codes, and the program codes are suitable for being loaded and run by the processor to execute the method described in any one of the technical solutions of the above-mentioned scene synthesis data generation method.

[0035] In a third aspect, the present invention provides a computer-readable storage medium storing a plurality of program codes, wherein the program codes are suitable for being loaded and run by a processor to execute the method described in any one of the technical solutions of the above-mentioned scene synthesis data generation method.

[0036] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects: on the basis of the basic terrain generated based on terrain parameters, the present invention can combine algorithm model information, cultural facilities information and natural ecological information to generate synthetic scenes, obtain scene synthesis data based on the synthetic scenes, and the data can be randomly generalized in real time during online simulation of the virtual environment and production on the cloud, realizing fast and efficient batch production of scene synthesis data, and while ensuring the rationality of the synthetic data, greatly improving the richness and generalization of the scene, with high customizability and flexibility, by adjusting the algorithm model information in the scene generation process and the parameters and rules contained in the cultural facilities information and natural ecological information, synthetic data that meets specific application requirements can be generated. At the same time, when the method of the present invention is used to produce algorithm model training data, there is no need to pre-store the generated content, and the virtual environment online simulation and the scene synthesis data generated in real time on the cloud can be directly output to the pre-trained algorithm model used for algorithm model training, thereby significantly saving storage space. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The disclosure of the present invention will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. In addition, similar numbers in the figures are used to represent similar components, among which:

[0038] Figure 1 is a schematic flow chart of main steps of a method for generating scene synthesis data according to an embodiment of the present invention;

[0039] Figure 2is a flowchart of the main steps of an algorithm model training method according to an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of an application scenario of the scene synthesis data generation method provided by the present invention;

[0041] Figure 4 It is a schematic diagram of the main structural block diagram of a scene synthesis data generating device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] Some embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0043] In the description of the present invention, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, and memories, and may also include software parts, such as program codes, or a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware, or a combination of the two. Non-temporary computer-readable storage media include any suitable medium that can store program codes, such as a magnetic disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, and the like. The term "A and / or B" means all possible combinations of A and B, such as only A, only B, or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B", and may include only A, only B, or A and B. The singular terms "one" and "the" may also include plural forms.

[0044] Although the currently used 3D software can complete the extraction of virtual scene data for algorithm model training, the actual use of such data in combination with products that use the algorithm model is limited. For example, when the product using the algorithm model is applied to aerial photography, automatic cruising, disaster relief and other scenes, the acquisition of scene data often becomes unrealistic due to the constant changes in the terrain in the environment. Combined with the complexity of the natural ecological environment, the sudden change of the terrain and other factors, the use of software for data processing will change the matching degree between the data and the actual natural ecological environment. Therefore, the existing technology of using software to produce scene data still has the problem that the data complexity is difficult to approach the real application scene.

[0045] Refer to the attached Figure 1, shows a flowchart of the steps of a method for generating scene synthesis data according to an embodiment of the present invention. The scene synthesis data generation process of this embodiment includes the following steps S11 to S15:

[0046] Step S11: obtaining preset basic parameter information; the basic parameter information includes terrain parameters and scene parameters, and the scene parameters include algorithm model information, cultural facilities information and natural ecology information;

[0047] Specifically in this embodiment, the terrain parameters may include terrain height, terrain size, landform type, etc., and the algorithm model information may include the purpose of algorithm model training and the starting and ending point coordinates used for training. For example, the purpose of algorithm model training may be aerial photography, map construction, disaster relief, detection, etc. The starting and ending point coordinates are the starting and ending point coordinates of the flight path that can cross the aerial photography area, map construction area, disaster relief or detection area. The human facilities information may include human facilities parameters and human facilities construction rules, and the natural ecology information may include ecological types, natural ecology parameters and natural ecology construction rules. It should be understood that the human facilities parameters may include parameters for characterizing the characteristics of generable human buildings, such as style characteristics; the natural ecology parameters may include parameters for characterizing the characteristics of generable natural ecology, such as size characteristics.

[0048] Step S12: generating a basic terrain based on the terrain parameters;

[0049] Specifically, in this embodiment, by setting terrain parameters, UE (Unreal Engine) can be used to generate a scene of basic terrain with rich details and realism. It should be understood that the generated basic terrain contains adaptive terrain and geographical environment, such as surface material and basic surface vegetation, rivers and stones, etc. Therefore, the generated basic terrain is equivalent to the geographical environment parameters of the basic terrain. Further, the geographical environment parameters may include ground slope parameters, ground material parameters, etc. In this embodiment, the basic terrain generated based on the terrain parameters is used to provide a terrain foundation for the construction of subsequent natural ecological scenes, as well as areas where humanistic buildings and facilities can be generated and areas where natural ecology of various vegetation can be generated.

[0050] Step S13: generating a natural ecology on the basic terrain based on the natural ecology information; generating a humanistic building facility on the basic terrain based on the humanistic facility information and the algorithm model information;

[0051] Step S14: obtaining a synthesized scene based on the basic terrain, the natural ecology and the cultural architectural facilities;

[0052] Step S15: Obtaining scene synthesis data based on the synthesized scene, so as to train the algorithm model based on the scene synthesis data.

[0053] In a preferred implementation manner based on the above step S11, obtaining preset basic parameter information may include:

[0054] Obtaining a data generalization range corresponding to at least one of the algorithm model information, cultural facilities information, and natural ecological information;

[0055] At least one of the algorithm model information, the cultural facilities information and the natural ecological information is determined based on the data generalization scope.

[0056] Specifically, the data generalization scope may include the data selection scope for the algorithm model training purpose, the data selection scope of the starting point and end point coordinates used for the training, the human facilities information, and the data selection scope of various parameters and rules in the natural ecological information.

[0057] In a preferred implementation manner based on the above step S13, generating cultural building facilities on the basic terrain based on the cultural facility information and the algorithm model information includes:

[0058] Step a, determining a target area in the basic terrain based on the algorithm model information;

[0059] Step b: generating cultural building facilities in the target area based on cultural facility parameters and cultural facility construction rules.

[0060] And / or, generating a natural ecology on the basic terrain based on the natural ecology information includes:

[0061] Step c: generating a natural ecology on the basic terrain based on the ecological type, natural ecological parameters and natural ecological construction rules.

[0062] It should be understood that the generation of natural ecology on the basic terrain based on the natural ecological information in this embodiment may also include: determining the target area in the basic terrain based on the algorithm model information; generating natural ecology in the target area based on the ecological type, the natural ecological parameters and the natural ecological construction rules.

[0063] Exemplarily, a preferred implementation method for determining a target area in the basic terrain based on the algorithm model information is: selecting the target area according to the start and end points of the algorithm model flight path set by the user and the purpose of algorithm model training.

[0064] Optionally, a specific implementation method of the above step b is: obtaining the geographical environment parameters of the target area in the basic terrain, setting at least one first-level cultural building facility sampling point in the target area based on the geographical environment parameters of the target area, the cultural facility parameters and the cultural facility construction rules, and placing a first-level cultural building facility at the first-level cultural building facility sampling point.

[0065] Specifically in this embodiment, the geographical environment parameters of the target area include a ground slope parameter; an area in the target area where the ground slope parameter is less than a preset slope parameter is selected as a humanistic building facility sampling area; and at least one first-level humanistic building facility sampling point is set in the humanistic building facility sampling area according to the humanistic facility parameters and the humanistic facility construction rules.

[0066] Optionally, a specific implementation method of the above step c is: based on the geographical environment parameters of the basic terrain, the ecological type, the natural ecological parameters and the natural ecological construction rules, at least one first-level natural ecological sampling point is set in the basic terrain, and a first-level natural ecology is placed at the first-level natural ecological sampling point.

[0067] Specifically in this embodiment, the geographical environment parameters of the basic terrain include ground material parameters; based on the ground material parameters, the different ground materials contained in the basic terrain are determined, and the sampling intervals adopted by the different ground materials are determined according to the ecological type, the natural ecological parameters and the natural ecological construction rules; and at least one first-level natural ecological sampling point is set in the basic terrain according to the sampling interval.

[0068] Optionally, another specific implementation method of the above step b is: based on the geographical environment parameters of the target area, the cultural facilities parameters and the cultural facilities construction rules, at least one first-level cultural building facility sampling point is set in the target area, and a first-level cultural building facility is placed at the first-level cultural building facility sampling point; based on the first-level cultural building facilities placed at the first-level cultural building facility sampling point, second-level cultural building facility sampling points are generated around the first-level cultural building facility sampling points, and second-level cultural building facilities are placed at the second-level cultural building facility sampling points. Exemplarily, humanistic building facilities classified as large assets are placed at the first-level humanistic building facility sampling points, such as assets such as buildings, walls and other large facilities; then, second-level humanistic building facility sampling points are derived around the first-level humanistic building facility sampling points according to the different assets placed at the first-level humanistic building facility sampling points, and humanistic building facilities classified as small assets are placed at the second-level humanistic building facility sampling points. For example, second-level humanistic building facility sampling points can be generated according to the different attributes of the assets placed at the first-level humanistic building facility sampling points (such as schools, hospitals, bridges, etc.) and second-level humanistic building facilities corresponding to the assets placed at the first-level humanistic building facility sampling points (such as flagpoles in schools, parking spaces in hospitals, etc.) can be placed.

[0069] Optionally, another specific implementation of the above step c is: at least one first-level natural ecology sampling point is set in the basic terrain based on the geographical environment parameters of the basic terrain, the ecological type, the natural ecology parameters and the natural ecology construction rules, and a first-level natural ecology is placed at the first-level natural ecology sampling point; based on the category of the first-level natural ecology placed at the first-level natural ecology sampling point, a second-level natural ecology sampling point is generated around the first-level natural ecology sampling point, and a second-level natural ecology is placed at the second-level natural ecology sampling point. Exemplarily, a natural ecology with a category of large assets, such as trees, stones and other assets, is placed at the first-level natural ecology sampling point, and then a second-level natural ecology sampling point is derived around the first-level natural ecology sampling point, and a second-level natural ecology with a category of small assets, such as shrubs, gravel and other assets, is placed at the second-level natural ecology sampling point.

[0070] Furthermore, in the specific implementation of the above step b and step c, it may also include: the step of changing the position of the first-level humanistic building facility sampling point, and / or the second-level humanistic building facility sampling point, and / or the first-level natural ecological sampling point, and / or the second-level natural ecological sampling point to change the sampling point attributes, and filtering according to preset conditions after changing the sampling point attributes, wherein the position transformation includes at least one of translation, rotation, and scaling. It is understandable that filtering the sampling points according to the preset conditions can be specifically filtering suitable first-level or second-level sampling points according to pre-set conditions such as ground height, slope, water area, etc., and then placing corresponding assets on the filtered sampling points, so that the obtained scene can be more reasonable.

[0071] In a preferred implementation of step S15 above, obtaining scene synthesis data based on the synthesized scene may include: automatically updating the scene parameters based on the data generalization range to obtain batches of synthesized scenes, and obtaining a scene synthesis data set based on the batches of synthesized scenes, and then performing algorithm model training based on the scene synthesis data set.

[0072] Specifically in this embodiment, the automatically updating the scene parameters based on the data generalization scope includes: automatically updating at least one of the algorithm model information, the cultural facilities distribution information and the ecological distribution information based on the data generalization scope.

[0073] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effects of the present invention, different steps do not have to be performed in such an order, and they can be performed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present invention.

[0074] See attached Figure 2 The present invention also provides an algorithm model training method, which mainly includes the following steps S21 and S22.

[0075] Step S21: Acquire scene synthesis data as training data;

[0076] Specifically in this embodiment, obtaining the use Figure 1 The batch of scene synthesis data generated by the scene synthesis data generation method is used as the training data.

[0077] Step S22: input the acquired training data into the pre-training model for training.

[0078] Specifically in this embodiment, the pre-trained model can be an algorithm model in various intelligent devices in the prior art. For example, the pre-trained model can be an MPC (Model Predictive Control) algorithm model or a SLAM (localization and mapping) algorithm model in an autonomous driving device.

[0079] like Figure 3 As shown, it is a schematic diagram of the application scenario of the scene synthesis data generation method provided by the present application. The scene synthesis data generation method of the present application can be combined with online simulation of virtual environment or cloud generation of synthetic data to achieve real-time random generalization in the data generation process to generate batches of scene synthesis data, and used for the training of algorithm models in smart devices, for example, it can be specifically used for AI model algorithm training, adaptive algorithm training in different environments, and online learning of simulation platforms. Exemplarily, the smart device can be a device with automatic driving function, and the algorithm model in the device can be an existing path planning algorithm model, positioning and navigation algorithm model, obstacle avoidance algorithm model, etc. In practical applications, the smart device can train the algorithm model in the smart device based on the scene image data collected by itself or the scene image data obtained by interaction with other devices or servers, or based on the scene synthesis data obtained by the method of the present application, and improve the performance of the smart device by improving the accuracy of the algorithm model. Preferably, the scene synthesis data obtained by the scene synthesis data generation method provided by the present application can greatly improve the richness and generalization of the scene while ensuring the rationality of the synthesis data, and has a high degree of customizability and flexibility. By adjusting the algorithm and rules used for scene construction, synthetic data that meets specific needs can be generated. At the same time, there is no need to store the generated content in advance, and the storage space can be significantly saved. Therefore, using the scene synthesis data obtained by the method of the present application to train the algorithm model in the smart device has a better application effect in improving the accuracy of the algorithm model and improving the performance of the smart device.

[0080] Furthermore, the present invention also provides a scene synthesis data generating device.

[0081] See attached Figure 4 , Figure 4 FIG. 1 is a main structural block diagram of a scene synthesis data generating device according to an embodiment of the present invention. Figure 4 As shown, the scene synthesis data generating device in the embodiment of the present invention mainly includes an input module 401 , a basic terrain generating module 402 , a scene generating module 403 and an output module 404 .

[0082] In some embodiments, one or more of the input module 401 , the basic terrain generation module 402 , the scene generation module 403 , and the output module 404 may be combined into one module.

[0083] In some embodiments, the input module 401 can be configured to: obtain preset basic parameter information, the basic parameter information includes terrain parameters and scene parameters; the scene parameters include algorithm model information, cultural facilities information and natural ecology information.

[0084] The basic terrain generation module 402 may be configured to generate a basic terrain according to the terrain parameters acquired by the input module 401 .

[0085] The scene generation module 403 may be configured to: generate a natural ecology on the basic terrain generated by the basic terrain generation module 402 based on the natural ecology information acquired by the input module 401; and generate humanistic building facilities on the basic terrain generated by the basic terrain generation module 402 based on the humanistic facility information and algorithm model information acquired based on the input module 401; and then obtain a synthesized scene based on the basic terrain, the natural ecology and the humanistic building facilities;

[0086] The output module 405 may be configured to obtain scene synthesis data according to the synthesized scene obtained by the scene generation module 403 .

[0087] The scene synthesis data generating device is used to perform Figure 1 The embodiment of the scene synthesis data generation method shown in the figure has similar technical principles, technical problems solved and technical effects produced. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process and related instructions of the scene synthesis data generation device can refer to the contents described in the embodiment of the scene synthesis data generation method, which will not be repeated here.

[0088] Further, it should be understood that since the setting of each module is only for illustrating the functional units of the device of the present invention, the physical devices corresponding to these modules may be the processor itself, or a part of the software in the processor, a part of the hardware, or a part of the combination of software and hardware. Therefore, the number of each module in the figure is only schematic.

[0089] Those skilled in the art will appreciate that the modules in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principle of the present invention, and therefore, the technical solutions after splitting or merging will fall within the protection scope of the present invention.

[0090] It is understood by those skilled in the art that the present invention implements all or part of the processes in the method of the above embodiment, and can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.

[0091] Furthermore, the present invention also provides a computer device. In one embodiment of a computer device according to the present invention, the computer device includes a processor and a memory, the memory can be configured to store a program for executing the scene synthesis data generation method of the above method embodiment, and the processor can be configured to execute the program in the memory, which includes but is not limited to the program for executing the scene synthesis data generation method of the above method embodiment. For ease of explanation, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The computer device can be a control device device formed by various electronic devices.

[0092] Furthermore, the present invention also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present invention, the computer-readable storage medium can be configured to store a program for executing the scene synthesis data generation method of the above method embodiment, and the program can be loaded and run by the processor to implement the above scene synthesis data generation method. For ease of explanation, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The computer-readable storage medium can be a storage device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present invention is a non-temporary computer-readable storage medium.

[0093] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A method for generating scene synthetic data, wherein the synthetic data is used for training an algorithm model, characterized in that: The method comprises: Obtaining preset basic parameter information; the basic parameter information includes terrain parameters and scene parameters; the scene parameters include algorithm model information, cultural facilities information and natural ecology information; generating a basic terrain based on the terrain parameters; generating a natural ecology on the basic terrain based on the natural ecology information; Generate humanistic building facilities on the basic terrain based on the humanistic facility information and the algorithm model information; Obtaining a synthetic scene based on the basic terrain, the natural ecology and the cultural architectural facilities; Obtaining scene synthesis data based on the synthesized scene, so that the algorithm model is trained based on the scene synthesis data; The algorithm model information includes the purpose of algorithm model training and the starting point and end point coordinates used for training; the cultural facility information includes cultural facility parameters and cultural facility construction rules; The generating of cultural building facilities on the basic terrain based on the cultural facility information and the algorithm model information comprises: Determine a target area in the base terrain based on the algorithm model information; Based on the cultural facility parameters and the cultural facility construction rules, cultural building facilities are generated in the target area.

2. The method according to claim 1, characterized in that The natural ecological information includes ecological type, natural ecological parameters and natural ecological construction rules, and the generation of natural ecology on the basic terrain based on the natural ecological information includes: Based on the ecological type, the natural ecological parameters and the natural ecological construction rules, a natural ecology is generated on the basic terrain.

3. The method according to claim 2, characterized in that The generating of the humanities building facilities in the target area based on the humanities facility parameters and the humanities facility construction rules comprises: Based on the geographical environment parameters of the target area, the cultural facility parameters and the cultural facility construction rules, at least one first-level cultural building facility sampling point is set in the target area, and a first-level cultural building facility is placed at the first-level cultural building facility sampling point; And / or, generating a natural ecology on the basic terrain based on the ecological type, the natural ecological parameters and the natural ecological construction rules includes: At least one first-level natural ecological sampling point is set in the basic terrain based on the geographical environment parameters of the basic terrain, the ecological type, the natural ecological parameters and the natural ecological construction rules, and a first-level natural ecology is placed at the first-level natural ecological sampling point.

4. The method according to claim 3, characterized in that The geographical environment parameter includes a ground slope parameter; and the setting of at least one first-level humanistic building facility sampling point in the target area based on the geographical environment parameter of the target area, the humanistic facility parameter and the humanistic facility construction rule includes: Select the area in the target area where the ground slope parameter is less than the preset slope parameter as the sampling area for humanistic building facilities; At least one first-level humanistic building facility sampling point is set in the humanistic building facility sampling area according to the humanistic facility parameters and the humanistic facility construction rules.

5. The method according to claim 3, characterized in that: The geographical environment parameters include ground material parameters; the geographical environment parameters based on the basic terrain, the ecological type, the natural ecological parameters and the natural ecological construction rules are used to set at least one first-level natural ecological sampling point in the basic terrain, including: Based on the ground material parameters, different ground materials included in the basic terrain are determined; according to the ecological type, the natural ecological parameters and the natural ecological construction rules, the sampling intervals used for different ground materials are determined; and according to the sampling intervals, at least one first-level natural ecological sampling point is set in the basic terrain.

6. The method according to claim 3, characterized in that The method further comprises: Based on the first-level humanistic building facilities placed at the first-level humanistic building facility sampling points, generating second-level humanistic building facility sampling points around the first-level humanistic building facility sampling points, and placing second-level humanistic building facilities at the second-level humanistic building facility sampling points; And / or, based on the primary natural ecology placed at the primary natural ecological sampling point, a secondary natural ecological sampling point is generated around the primary natural ecological sampling point, and a secondary natural ecology is placed at the secondary natural ecological sampling point.

7. The method according to claim 6, characterized in that The method also includes the steps of transforming the positions of the first-level cultural building facility sampling points, and / or the second-level cultural building facility sampling points, and / or the first-level natural ecological sampling points, and / or the second-level natural ecological sampling points to change the sampling point attributes, and screening according to preset conditions after changing the sampling point attributes, wherein the position transformation includes at least one of translation, rotation and scaling.

8. The method according to any one of claims 1 to 7, characterized in that: The obtaining of preset basic parameter information includes: Obtaining a data generalization range corresponding to at least one of the algorithm model information, cultural facilities information, and natural ecological information; At least one of the algorithm model information, the cultural facilities information and the natural ecological information is determined based on the data generalization scope.

9. A computer device comprising a processor and a memory, wherein the memory is suitable for storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by the processor to execute the scene synthesis data generating method according to any one of claims 1 to 8.

10. A computer-readable storage medium storing a plurality of program codes, characterized in that: The program code is suitable for being loaded and run by a processor to execute the scene synthesis data generation method according to any one of claims 1 to 8.

Images