Multi-task scene dynamic generation method, server, system and medium

Abstract

The invention provides a multi-task scene dynamic generation method, a server, a system and a medium. The method comprises the steps of obtaining a to-be-trained medical task of nuclear emergency medical rescue training; each medical task corresponds to one partition mode, various terrain categories and various ground feature categories; dividing the original scene into a plurality of regions according to a partitioning mode and a plurality of terrain categories corresponding to the to-be-trained medical task, and generating landforms of the plurality of regions; selecting a plurality of ground feature models of each region from a preset model library according to a plurality of ground feature categories corresponding to the to-be-trained medical task and the landform of each region, and generating a basic scene; simulating a radiation environment in each area of the basic scene; based on the basic scene of the radiation environment and the medical task to be trained, multiple wounded persons are generated in the multiple areas, and the injury condition of each wounded person is generated. According to the invention, complex and diversified rescue scenes can be dynamically generated, and the complex requirements of multiple scenes and multiple tasks in nuclear emergency medical rescue are met.

Description

Technical Field

[0001] The present application relates to the field of medical training technology, and in particular to a method, server, system, and medium for dynamically generating multi-task scenarios. Background Art

[0002] Nuclear and radiation emergencies are characterized by suddenness, wide-ranging impacts, and significant difficulty in treatment. The core of nuclear emergency medical rescue is to rapidly remove radioactive contamination from the casualty's body, proactively treat radiation damage, and save lives. Rescuers must not only master the essential knowledge of nuclear medicine diagnosis and treatment but also possess practical skills in detection, protection, decontamination, and treatment within the context of nuclear emergency medical rescue. Therefore, nuclear emergency medical rescue training is essential for relevant personnel to enhance their capabilities.

[0003] With the continuous advancement of technology, virtual reality technology has been adopted in simulation and training. It can simulate real-world scenarios and facilitate nuclear emergency medical rescue training. However, nuclear emergency medical rescue is highly dangerous and the scenarios encountered are complex and varied. Traditional training scenario simulations are primarily achieved through 3D modeling of real-world scenarios or by pre-designing virtual scenarios of real-world scenarios. These training scenarios are relatively focused and single, primarily focusing on practicing similar scenarios. They lack complexity, diversity, and immersion, making them unable to meet the complex multi-scenario and multi-task requirements of nuclear emergency medical rescue training and hindering the development of the capabilities of nuclear emergency medical rescue personnel. Summary of the Invention

[0004] The embodiments of the present application provide a method, server, system, and medium for dynamically generating multi-task scenarios to address the problem that existing nuclear and radiation emergency training scenarios are relatively simple and lack complexity, diversity, and immersion, meet the complex requirements of nuclear emergency medical rescue training for multiple scenarios and multiple tasks, and improve the capabilities of relevant personnel.

[0005] In a first aspect, an embodiment of the present application provides a method for dynamically generating a multi-task scenario, comprising:

[0006] Obtain medical tasks to be trained for nuclear emergency medical rescue training; each medical task corresponds to a zoning method, multiple terrain types, and multiple ground feature types;

[0007] Divide the original scene into multiple areas according to the partitioning method corresponding to the medical task to be trained and multiple terrain categories, and generate the terrain features of the multiple areas;

[0008] According to the multiple ground object categories corresponding to the medical task to be trained and the topography of each area, multiple ground object models of each area are selected from a preset model library to generate a basic scene;

[0009] simulating a radiation environment in each area of ​​the base scenario;

[0010] Based on the basic scenario of setting the radiation environment and the medical task to be trained, a plurality of wounded persons are generated in the plurality of areas, and the injury condition of each wounded person is generated.

[0011] In a possible implementation, the partitioning mode includes continuous partitioning and spliced ​​partitioning;

[0012] The method of dividing the original scene into multiple regions according to the partitioning method and multiple terrain categories corresponding to the medical task to be trained, and generating the terrain features of the multiple regions, includes:

[0013] If the partitioning method corresponding to the medical task to be trained is continuous partitioning, randomly selecting at least one from the multiple terrain categories corresponding to the medical task to be trained to generate the topography of the original scene; partitioning is performed based on the characteristics of the topography in the original scene to obtain multiple regions and the topography of each region;

[0014] If the partitioning method corresponding to the medical task to be trained is splicing partitioning, multiple task requirements are determined according to the medical task to be trained; according to the task requirements, the original scene is partitioned to obtain at least one area corresponding to each task requirement; according to the task requirements of each area and the multiple terrain categories corresponding to the medical task to be trained, the terrain and landforms of each area are generated respectively.

[0015] In a possible implementation, the step of generating multiple injured persons in the multiple areas based on the basic scenario of the radiation environment and the medical task to be trained, and generating the injury condition of each injured person includes:

[0016] Determining the initial number and initial positions of the wounded based on the population density of the terrain model in the basic scenario, the radiation environment, and the medical task to be trained;

[0017] Generating an initial injury condition of each injured person based on a terrain model and radiation environment at the initial position of each injured person; the initial injury condition includes an injury level;

[0018] The preset injury statistical model corresponding to the medical task to be trained is used to adjust the initial number and injury levels of all the injured to obtain the adjusted number of injured and the injury condition of each injured.

[0019] In a possible implementation, after generating multiple injured persons in the multiple areas and generating the injury condition of each injured person based on the basic scenario of the radiation environment and the medical task to be trained, the method further includes:

[0020] Obtain the current training time and rescue operations performed by the trainer;

[0021] The injury condition of each injured person is dynamically evolved according to the current training duration and the rescue operation.

[0022] In a possible implementation, each random number corresponds to a meteorological category, and each meteorological category corresponds to a parameter selection range of a meteorological element;

[0023] Before simulating the radiation environment in each area of ​​the basic scenario, the following steps are also included:

[0024] Generate a random number within the preset data range;

[0025] Determining the meteorological category corresponding to the random number as the meteorological category of the basic scene;

[0026] Determining the parameter values ​​of the meteorological elements in each area according to the partitioning method corresponding to the medical task to be trained and the parameter selection range corresponding to the meteorological category;

[0027] Adjust the parameter values ​​of meteorological elements in each area according to the topography of each area;

[0028] Based on the adjusted parameter values ​​of meteorological elements in each region, meteorological simulation is performed in each region to generate the meteorological environment.

[0029] In a possible implementation, the object model in the basic scene includes a building model;

[0030] The radiation environment is simulated in each area of ​​the base scenario, including:

[0031] Determining radiation information of a radioactive source in each partition of the basic scene according to a partitioning method corresponding to the medical task to be trained;

[0032] Determining, based on the radiation information corresponding to each partition, the radiation diffusion range outside the building model in each partition and the first radiation intensity at each position outside the building model;

[0033] Determining the second radiation intensity at each location within each building model based on the radiation information corresponding to each partition and the building model within each partition;

[0034] Based on the radiation diffusion range, the first radiation intensity, and the second radiation intensity, a radiation environment in each area of ​​the basic scene is simulated.

[0035] In a possible implementation, each terrain category corresponds to a distribution probability of a ground feature category;

[0036] The method comprises selecting multiple ground object models for each area from a preset model library according to the multiple ground object categories corresponding to the medical task to be trained and the topography of each area to generate a basic scene, including:

[0037] Determine the ground feature category of each area according to the multiple ground feature categories corresponding to the medical task to be trained and the distribution probability of the ground feature categories corresponding to the terrain category of the topography of each area;

[0038] According to the topography and corresponding land feature categories of each area, multiple land feature models of each area are selected from the preset model library to generate a basic scene.

[0039] In a second aspect, an embodiment of the present application provides a server comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method in the first aspect or any possible implementation of the first aspect is implemented.

[0040] In a third aspect, an embodiment of the present application provides a system comprising a virtual reality (VR) device and the server as described in the second aspect above.

[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method in the first aspect or any possible implementation of the first aspect.

[0042] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method in the first aspect or any possible implementation of the first aspect.

[0043] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0044] The embodiment of the present application divides the original scene in the simulation training into multiple areas by obtaining the medical tasks to be trained for nuclear emergency medical rescue training, and generates the topography of the multiple areas, so that the partitions of the original scene can be adapted to the medical tasks to be trained so as to train for each medical task; through the multiple types of land objects corresponding to the medical tasks to be trained and the topography of each area, multiple land object models of each area are selected from the preset model library to generate a basic scene, and land objects on the terrain, such as buildings, roads and trees, can be generated according to the adaptability of the topography of each area to transform the virtual scene and increase the scene diversity; by simulating the radiation environment in each area of ​​the basic scenario, the radiation situation at the nuclear emergency medical rescue site can be simulated in order to carry out the medical tasks to be trained and improve the authenticity of the scenario; by generating multiple wounded in multiple areas based on the basic scenario and the medical tasks to be trained of the radiation environment, and generating the injuries of each wounded, the radiation environment and the basic scenario can be considered when generating the wounded and their injuries, so that the wounded and their injuries are adapted to the scenario, and the diversity of the wounded in the scenario is improved, thereby enriching the formed virtual scene and improving the immersion of the trainees, so as to better conduct nuclear emergency medical rescue training. This application generates corresponding virtual scenes by the medical tasks to be trained, and can carry out diversified training for the medical tasks to be trained; at the same time, different medical tasks to be trained can generate virtual scenes with different task requirements, which can realize the training of various types of medical tasks, meet the complex requirements of multiple scenes and multiple tasks in nuclear emergency medical rescue, and thus improve the actual combat capability of relevant personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1 This is an application scenario diagram of the multi-task scenario dynamic generation method provided by an embodiment of the present application;

[0047] Figure 2 This is a flowchart of the implementation of the multi-task scenario dynamic generation method provided in an embodiment of the present application;

[0048] Figure 3 is a schematic diagram of a spliced ​​partition provided in an embodiment of the present application;

[0049] Figure 4 It is a schematic diagram of the server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0051] The inventors of this application have discovered that in the training of nuclear emergency medical rescue, traditional scenario simulation is mainly obtained by modeling by collecting data from real scenes or pre-designed event scenes. This scenario is relatively simple and is generally only suitable for achieving a specific task. It lacks complexity and diversity and is difficult to meet the complex requirements of multiple scenarios and multiple tasks in nuclear emergency medical rescue.

[0052] In order to increase the diversity of virtual training scenarios, in the implementation mode of the present application, corresponding virtual training scenarios are generated by obtaining the medical tasks to be trained for nuclear emergency medical rescue training, including the terrain, objects, radiation environment and wounded in the scenario. It is possible to randomly generate appropriate virtual scenarios for the medical tasks to be trained, thereby increasing the diversity and immersion of the scenarios. At the same time, corresponding virtual scenarios can be generated for a variety of medical tasks to meet the training and training needs of different medical tasks.

[0053] In order to make the purpose, technical solutions and advantages of this application clearer, specific embodiments will be described below with reference to the accompanying drawings.

[0054] Figure 1 This is an application scenario diagram of the multi-task scenario dynamic generation method provided in an embodiment of the present application. In the virtual training of nuclear emergency medical rescue, the architecture of the virtual reality-based training system mainly includes virtual reality (VR) equipment and servers.

[0055] VR equipment includes a data acquisition unit (such as an optical data acquisition unit and an inertial data acquisition unit), a VR wearable device (such as a head-mounted display and motion capture equipment), a backpack host, and a motion capture computing device. The head-mounted display provides an immersive virtual reality experience for users, while the backpack host provides the core computing power for virtual scene rendering, data processing, and interactive logic calculations. The backpack host can also be a wireless host, a cloud host, or a wearable host, and can be selected and configured based on the actual scenario.

[0056] The motion capture device is a wearable device that captures the trainee's body movements and transmits them to a data acquisition unit. The data acquisition unit can be set up in the trainee's area to collect signals from the motion tracking device, capture the trainee's movements, and transmit them in real time to the motion capture computing device for processing. The data acquisition unit also provides network clock synchronization to ensure consistent time.

[0057] Motion capture computing equipment uses the data it receives to capture the trainee's current position and real-time movements, creating a virtual avatar of the user within the virtual scene. Using the limb movement data, it reverse-calculates the entire body's skeletal posture, and through motion estimation and inference compensation for missing information, generates accurate motion capture data.

[0058] The server is mainly used to implement key tasks in VR technology, such as data processing, information storage, and system monitoring, including the implementation of the multi-task scene dynamic generation method provided by this application. The server can include computing servers, storage servers, and GPU computing servers.

[0059] Data can be transmitted between VR devices and servers through switches. The server can transmit the rendered virtual scenes and interactive data to each user's head-mounted display device in real time through the switch and backpack host. The motion capture computing device can transmit the trainee's current position and real-time movement posture to the server in real time through the switch, generate a virtual avatar of the user in the virtual scene, and update the entire virtual scene in real time.

[0060] Figure 2 The following is a flowchart of the implementation of the method for dynamically generating a multi-task scenario provided in the embodiment of the present application:

[0061] Step 201: Obtain medical tasks to be trained for nuclear emergency medical rescue training; each medical task corresponds to a zoning method, multiple terrain types, and multiple ground feature types.

[0062] In this embodiment, the training tasks of nuclear emergency medical rescue training include environmental assessment zoning tasks, casualty triage tasks, forward emergency support tasks, radioactive contamination decontamination tasks, on-site emergency treatment tasks, sample collection and processing tasks, and contaminated casualty evacuation tasks.

[0063] Since the task scenarios and task focuses of each task are different, the corresponding virtual scenes are different. In order to achieve targeted task training, the partitioning of the virtual scenes will also be different.

[0064] For example, an environmental assessment zoning task requires assessing and zoning the environment at a nuclear or radiation emergency site. A relatively complete on-site scenario can be constructed to allow trainees to assess and zon the environment within. In a forward emergency support mission, the specific location of the injured has been determined. Focusing on the individual injured, assessing and treating their condition, the scenario needs to be divided into zones. Within each zone, injured individuals can be located at different locations within the incident scene. This allows multiple trainees to quickly and accurately assess the conditions of multiple injured individuals in multiple locations and provide emergency assistance within a virtual, confined space and tense atmosphere.

[0065] Here, each medical mission requires different scenes, and the corresponding terrain and objects will also be different. Among them, objects can include buildings, roads, rivers, and trees.

[0066] For example, in a forward emergency support mission, casualties may be located inside buildings, on roads, or in fields. Therefore, appropriate modeling of these possible locations is necessary. Contaminated casualty evacuation missions require rapid transport of casualties from the incident scene to various levels of medical facilities in the rear. This requires training personnel to ensure the safety and stability of the casualties during transport. The scenarios primarily involve the routes or roads involved in the transfer process, which can involve a variety of terrains and obstacles such as rivers, trees, and hills.

[0067] Step 202 : Divide the original scene into multiple regions according to the partitioning method corresponding to the medical task to be trained and the multiple terrain categories, and generate the terrain features of the multiple regions.

[0068] In this embodiment, different medical tasks require different scenarios, and the corresponding partitioning methods are different. Using the partitioning method corresponding to the medical task to be trained to divide the regions, the resulting multiple regions can be suitable for the medical task to be trained, so that a scenario that can execute the medical task can be generated in each region.

[0069] Partitioning methods include continuous partitioning and spliced ​​partitioning. Continuous partitioning means that the boundaries between partitions are connected to each other, forming a complete and continuous scene. Spliced ​​partitioning means that the boundaries between some partitions are independent and are directly spliced ​​together to form a complete scene.

[0070] The corresponding terrain and landforms are generated according to the terrain category corresponding to the medical task to be trained and the characteristics or requirements of the medical task to be trained itself.

[0071] For example, an environmental assessment zoning mission requires a relatively complete nuclear and radiation emergency scene. The terrain of each area can be varied, with smooth, continuous transitions. For a forward emergency support mission, individual casualties can be targeted, with each area containing casualties in different locations. Therefore, the terrain of each area can be relatively independent and clearly separated.

[0072] In step 203 , a plurality of ground object models for each area are selected from a preset model library according to the various ground object categories corresponding to the medical task to be trained and the topography of each area, to generate a basic scene.

[0073] In this embodiment, terrain features can include various buildings, roads, vehicles, streetlights, trees, rocks, and rivers. The model library is a pre-established database containing various terrain feature models. The terrain feature models in the library have different types, sizes, shapes, and appearances to meet the needs of different tasks. For example, the library may include building models of various styles and heights to accommodate simulations of different urban or rural environments; it may also include various types of vehicle models, such as ambulances, fire trucks, and private cars, to simulate different traffic scenarios.

[0074] Topography is the basic geographical feature of each region. Since the topography of each region is different, the distribution and placement of the feature models thereon will also be different. For example, when placing a building model in a mountainous area, it is necessary to take into account the slope and undulations of the topography to avoid the building model appearing unreasonably suspended in the air or embedded in the ground.

[0075] By placing multiple terrain models of each selected area into the corresponding area according to certain layouts and rules, and combining them with the topography of the area, a complete basic scene is formed, building a simulation environment for the medical tasks to be trained, and restoring the scenarios that medical tasks may face in the real world as much as possible, allowing trainees to train in an environment close to reality and improve their ability to deal with various actual situations.

[0076] Step 204: simulate the radiation environment in each area of ​​the basic scene.

[0077] In this embodiment, radiation-enhancing environmental effects are created within each area of ​​the basic scenario. Parameters such as the central location of the radiation source, radiation type, and concentration can be set. Simulations are then performed using a radiation diffusion model, allowing trainees to experience and learn how to conduct nuclear emergency medical rescue operations in such hazardous environments, such as how to properly don protective gear and how to safely approach and evacuate casualties. By simulating radiation environments within each area of ​​the basic scenario, the realism and complexity of the training scenarios can be further enhanced, improving medical personnel's ability to handle radiation-related medical tasks.

[0078] Optionally, before simulating the radiation environment, the meteorological environment can be simulated within each area of ​​the base scenario, including temperature, humidity, wind direction, wind speed, sunlight, rainfall, and cloud cover. By simulating different weather conditions, the complex environments that medical missions may encounter can be more comprehensively and realistically reproduced, allowing trainees to train under realistic conditions and improve their ability to handle nuclear emergency medical rescue missions in various meteorological conditions. Furthermore, radiation diffusion varies under different weather conditions; for example, radiation diffusion may be smaller in the direction of headwind.

[0079] Here, each random number corresponds to a meteorological category, and each meteorological category corresponds to a parameter selection range of a meteorological element. Before simulating the radiation environment in each area of ​​the basic scene, the meteorological environment simulation is performed in each area of ​​the basic scene, specifically:

[0080] Generate a random number within a preset data range; determine the meteorological category corresponding to the random number as the meteorological category of the basic scenario; determine the parameter values ​​of the meteorological elements in each area based on the partitioning method corresponding to the medical task to be trained and the parameter selection range corresponding to the meteorological category; adjust the parameter values ​​of the meteorological elements in each area based on the topography of each area; and perform meteorological simulation in each area based on the adjusted parameter values ​​of the meteorological elements in each area to generate a meteorological environment.

[0081] In this embodiment, the preset data range is the range for selecting random numbers. Weather categories can include sunny, rainy, windy, and foggy. Weather categories can be further divided, such as snow, hail, dust, and haze. Weather categories can also be associated with seasons, such as spring, winter, and spring and autumn.

[0082] For example, the correspondence between random numbers and weather categories can be, for example, a preset data range of natural numbers from 1 to 10, where each number corresponds to a weather category, 1, 2, and 3 correspond to sunny days, 4, 5, and 6 correspond to rainy days, 5 and 6 correspond to strong wind days, 7 and 8 correspond to foggy days, etc. The preset data range can also be 0-1, where random numbers in [0, 0.4] correspond to sunny days, random numbers in (0.4, 0.6] correspond to rainy days, random numbers in (0.6, 0.8] correspond to strong wind days, and random numbers in (0.8, 1] correspond to foggy days, etc.

[0083] If the zoning method is continuous, the meteorological environment is also continuous in each area. You can select the parameter values ​​of meteorological elements for all areas to ensure the continuity of the meteorological environment. If the zoning method is spliced, you can select the parameter values ​​of meteorological elements for each area separately.

[0084] Temperatures vary with altitude, with the temperature dropping by approximately 0.6 degrees Celsius for every 100-meter increase in altitude. Therefore, the parameters of meteorological elements can be adjusted based on the topography of each region to better match the corresponding topography, thereby generating the overall meteorological environment.

[0085] Step 205 : Based on the basic scenario of the radiation environment and the medical task to be trained, multiple wounded persons are generated in multiple areas, and the injury condition of each wounded person is generated.

[0086] In this embodiment, different medical training missions require different numbers of casualties and the severity of their injuries. For example, in a forward emergency support mission, casualties may be located in a variety of locations, including various locations in buildings, on roads, and in fields, and their distribution is relatively dispersed. In a contaminated casualty evacuation mission, the casualties may be located in a relatively safe area near the incident site before the mission begins, and their distribution is relatively concentrated.

[0087] The distribution and severity of injuries will vary depending on factors such as topography and radiation intensity. For example, areas close to radiation sources may have a higher number of casualties with more severe injuries, while relatively safer areas may have fewer casualties with less severe injuries. When generating casualties and their injuries, it is also necessary to comprehensively consider the radiation environment and other possible factors (such as building collapse and fire in the scenario) to ensure that the casualties and their injuries are consistent with the environment and make the training more realistic for nuclear emergency medical rescue scenarios. For example, some casualties may suffer radiation injuries such as skin burns and internal organ damage due to high radiation doses; some casualties may also suffer from fractures, traumatic bleeding, and other injuries caused by building collapse.

[0088] The embodiment of the present application divides the original scene in the simulation training into multiple areas by obtaining the medical tasks to be trained for nuclear emergency medical rescue training, and generates the topography of the multiple areas, so that the partitions of the original scene can be adapted to the medical tasks to be trained so as to train for each medical task; through the multiple types of land objects corresponding to the medical tasks to be trained and the topography of each area, multiple land object models of each area are selected from the preset model library to generate a basic scene, and land objects on the terrain, such as buildings, roads and trees, can be generated according to the adaptability of the topography of each area to transform the virtual scene and increase the scene diversity; by simulating the radiation environment in each area of ​​the basic scenario, the radiation situation at the nuclear emergency medical rescue site can be simulated in order to carry out the medical tasks to be trained and improve the authenticity of the scenario; by generating multiple wounded in multiple areas based on the basic scenario and the medical tasks to be trained of the radiation environment, and generating the injuries of each wounded, the radiation environment and the basic scenario can be considered when generating the wounded and their injuries, so that the wounded and their injuries are adapted to the scenario, and the diversity of the wounded in the scenario is improved, thereby enriching the formed virtual scene and improving the immersion of the trainees, so as to better conduct nuclear emergency medical rescue training. This application generates corresponding virtual scenes by the medical tasks to be trained, and can carry out diversified training for the medical tasks to be trained; at the same time, different medical tasks to be trained can generate virtual scenes with different task requirements, which can realize the training of various types of medical tasks, meet the complex requirements of multiple scenes and multiple tasks in nuclear emergency medical rescue, and thus improve the actual combat capability of relevant personnel.

[0089] In some embodiments, the partitioning methods include continuous partitioning and spliced ​​partitioning.

[0090] Here, continuous partitioning means that the boundaries between partitions are connected to each other, forming a complete and continuous scene. Partitioning is for the purpose of setting up feature models for each area later.

[0091] Spliced ​​zoning means that some zones have independent boundaries (some can be connected, such as in a scenario where the transport route for a contaminated casualty evacuation mission must be continuous), and the entire scene is formed by directly splicing them together. Zoning allows for targeted settings based on the terrain and features of each area, eliminating some scenes that are not currently needed for the medical mission being trained.

[0092] This embodiment divides the original scene into multiple regions according to the partitioning method corresponding to the medical task to be trained and multiple terrain categories, and generates the terrain features of the multiple regions, which can be:

[0093] If the partitioning method corresponding to the medical task to be trained is continuous partitioning, at least one is randomly selected from the multiple terrain categories corresponding to the medical task to be trained to generate the topography of the original scene; partitioning is performed based on the characteristics of the topography in the original scene to obtain multiple areas and the topography of each area.

[0094] If the partitioning method corresponding to the medical task to be trained is splicing partitioning, multiple task requirements are determined based on the medical task to be trained; according to the task requirements, the original scene is partitioned to obtain at least one area corresponding to each task requirement; based on the task requirements of each area and the multiple terrain categories corresponding to the medical task to be trained, the terrain of each area is generated separately.

[0095] In this embodiment, when the medical task to be trained corresponds to a continuous partition, one of multiple terrain types can be randomly selected to generate the terrain of the entire scene. Partitions can then be created based on the characteristics of the terrain. For example, partitions can be created based on the direction of mountains and rivers, while plains can be divided into regular grids.

[0096] When the medical task to be trained corresponds to a spliced ​​partition, due to the different task requirements of the medical task to be trained, each area in the spliced ​​partition may be independent, or some areas in the spliced ​​partition may be independent and some areas may be continuous. The task requirements of the medical task to be trained can be clarified first.

[0097] For example, in a forward emergency support mission, it is necessary to assess and treat the injuries of the wounded at a certain location. The wounded at each location are independent of each other, that is, each partition in the virtual scene is independent. The original scene can be directly partitioned, for example, according to a regular grid, and each grid can be used as an independent area, such as Figure 3 As shown, A1, A2, A3...A16 represent an area respectively.

[0098] Since some partitions in the spliced ​​partition are independent and discontinuous, the terrain can be set separately for each discontinuous partition, and the terrain can be kept continuous or smoothly transitioned for the continuous partitions.

[0099] In some embodiments, based on a basic scenario of setting a radiation environment and a medical task to be trained, multiple wounded are generated in multiple areas, and the injury condition of each wounded is generated. The initial number and initial position of the wounded are determined based on the population density, radiation environment and medical task to be trained of the terrain model in the basic scenario; the initial injury condition of each wounded is generated based on the terrain model and radiation environment at the initial position of each wounded; the initial injury condition includes the injury level; finally, the preset injury statistical model corresponding to the medical task to be trained is used to adjust the initial number and injury level of all wounded to obtain the adjusted number of wounded and the injury condition of each wounded.

[0100] In this embodiment, the terrain models in the basic scene (such as buildings and vehicles) affect the distribution of people, resulting in different population density areas. For example, in densely populated areas in the city center, the population density is relatively high; while in remote mountainous areas or open areas, the population density is lower. The difference in population density affects the number and distribution of casualties, and areas with higher population density may have more casualties.

[0101] The radiation environment has a significant impact on casualty generation. People near radiation sources or in areas with high radiation doses are more vulnerable, so more casualties are likely to spawn in these areas. For example, near nuclear facilities, due to the high radiation levels, people in these areas are more likely to become casualties.

[0102] Different medical missions will result in different numbers and distributions of casualties. For example, in a large-scale nuclear and radiation emergency rescue mission, a larger number of casualties may need to be simulated to test the rescue team's response capabilities; in smaller incidents, the number of casualties may be relatively small. In a forward emergency support mission, all casualties located in the environment must be treated, and the number of casualties may be higher. In a casualty transport mission, the critically injured and those exposed to high doses of radiation must be prioritized for transport. Therefore, for a nuclear and radiation emergency of the same severity, the number of casualties required may be lower than in a forward emergency support mission.

[0103] When determining the initial location of the injured, a random selection can be made within the basic scenario. In densely populated areas, where the number of injured is likely to be greater, a larger number of locations can be selected as the initial locations of the injured. Alternatively, a probability-based spatial random allocation algorithm can be used to generate the initial locations of the injured.

[0104] Among them, the basic scene can be divided into multiple small spatial units, and the probability of casualty generation can be set for each spatial unit. The setting of the probability can be based on factors such as radiation intensity, spatial function and personnel density. For example, in a spatial unit with dense personnel (such as a conference room, office) and high radiation intensity, the probability of casualty generation is higher; while in an area with sparse personnel (such as a warehouse, equipment room), the probability of casualty generation is lower. According to the set probability, casualties are randomly generated in each spatial unit. For each spatial unit, a random judgment is made based on its set probability whether to generate a casualty. If a casualty is generated, the number of casualties and the specific location of their distribution are further determined.

[0105] The terrain model (e.g., collapsed buildings, explosions of flammable materials) and the radiation environment (radiation dose, radiation type, etc.) at the casualty's initial location will affect the casualty's injury. For example, a casualty in an area with collapsed buildings may suffer fractures, traumatic bleeding, or other injuries. A casualty in an area with high radiation levels may suffer radiation burns or acute radiation sickness. Different factors, such as radiation intensity and type, can cause varying degrees of radiation damage. Based on these conditions, each casualty's initial injury profile is generated.

[0106] The initial injury condition includes the type of injury and the level of injury. The level of injury can be minor injury (such as slight abrasions, initial reaction to low radiation doses, etc.), serious injury (such as severe fractures, organ damage caused by high-dose radiation, etc.) and critical injury (such as unstable vital signs, multiple organ failure, etc.).

[0107] Since the initial number and initial positions of the wounded are randomly generated and may deviate from the actual situation, the preset injury statistical model corresponding to the medical task to be trained can be used to adjust the initial number and injury levels of all the wounded, so that the generated wounded are more consistent with the distribution and change patterns of the wounded's injuries in real situations, and more realistically simulate the actual nuclear emergency medical rescue situation.

[0108] Among them, the preset injury statistical model is pre-set according to the characteristics and requirements of the medical task to be trained.

[0109] Optionally, the medical mission includes a forward emergency support mission. When the medical mission to be trained is a forward emergency support mission, after adjusting the initial number and initial injury conditions of all injured individuals using the preset injury statistical model corresponding to the medical mission to be trained, and obtaining the adjusted number of injured individuals and the injury condition of each injured individual, each injured individual's disaster avoidance behavior pattern can be determined based on their injury condition and initial position. According to each injured individual's disaster avoidance behavior pattern, each injured individual is moved from their initial position to a direction with low radiation intensity in their radiation environment to obtain their final position. Based on each injured individual's disaster avoidance behavior pattern and final position, each injured individual's injury condition is updated.

[0110] In this embodiment, it is considered that after a nuclear or radiation emergency occurs, people will have different levels of evacuation behavior, which will affect their final location and injury severity. For example, injured people may quickly seek shelter, such as entering a building, tunnel, or lying in a low-lying area, which may reduce their injuries; while others may panic and run around, suffering secondary injuries during movement, aggravating their injuries.

[0111] Therefore, during forward emergency support missions, evacuation behavior patterns can be tailored to each casualty. Those near radiation sources or in dangerous areas may move as quickly as possible away from the source; whereas those in relatively safe areas may choose to evacuate less urgently. Furthermore, some casualties with less severe injuries, who can move freely to evacuate, may choose to run quickly away from the danger zone. Others, who may have suffered blast injuries or fractures and are unable to move quickly, may choose to move slowly and seek cover.

[0112] Disaster avoidance behavior can affect the severity of a patient's injuries. For example, a patient running quickly may experience worsening injuries due to physical exertion or encountering other hazards during movement (such as falls or collisions). Crawling on rough terrain may cause new abrasions or displace existing fractures.

[0113] Final location also affects injury severity. Patients in different radiation intensity zones experience varying degrees of radiation damage, which can lead to changes in injury severity. For example, a patient who moves to an area with lower radiation intensity may gradually improve, while a patient who remains in a high-intensity zone may continue to experience worsening injuries.

[0114] Specifically, each disaster avoidance behavior pattern can correspond to multiple injury change directions, and the injury status of the injured person can be updated by randomly selecting an injury change direction. When updating the injury status, the injury type can be added or the injury level can be changed.

[0115] In some embodiments, the feature model in the base scene includes a building model. The building model can be configured with building materials, wall thickness, and other parameters. Different building materials and wall thicknesses can affect the propagation and attenuation of radiation within the building. For example, reinforced concrete walls offer greater radiation blocking capabilities than wooden walls.

[0116] This embodiment simulates the radiation environment in each area of ​​the basic scene. First, the radiation information of the radiation source in each area of ​​the basic scene can be determined according to the partitioning method corresponding to the medical task to be trained; then, based on the radiation information corresponding to each area, the radiation diffusion range outside the building model in each area and the first radiation intensity at each position outside the building model are determined; based on the radiation information corresponding to each area and the building model in each area, the second radiation intensity at each position within each building model is determined; finally, based on the radiation diffusion range, the first radiation intensity, and the second radiation intensity, the radiation environment in each area of ​​the basic scene is simulated.

[0117] In this embodiment, radiation may diffuse outward from the radiation source. At locations outside the building model, radiation intensity may vary depending on the distance from the radiation source, environmental factors, and other factors. For example, radiation intensity may be higher closer to the radiation source; and in windy conditions, radiation intensity downwind may vary. Therefore, a radiation diffusion model can be used to predict radiation diffusion at various locations outside the building model.

[0118] Radiation intensity attenuates as it passes through a building. For example, when radiation passes through a thick concrete wall, its intensity decreases significantly. Therefore, it's possible to simulate the radiation environment within the building model, taking into account its inherent characteristics, to accurately simulate the radiation intensity at different locations within the building model.

[0119] By comprehensively determining information such as the radiation diffusion range, the first radiation intensity, and the second radiation intensity, the radiation environment in each area of ​​the basic scenario is constructed, which can more realistically simulate the radiation conditions in different locations in the actual scenario and provide an accurate radiation environment for the simulation of medical tasks.

[0120] In some embodiments, each terrain category corresponds to a distribution probability of a feature category. Different terrain categories (e.g., mountainous, plain, watery, etc.) affect the distribution of feature categories. For example, in mountainous terrain, the probability of features such as trees and rocks appearing is relatively high; whereas in plain terrain, the probability of features such as farmland and villages appearing is likely to be higher.

[0121] This embodiment selects multiple terrain models of each area from a preset model library based on the multiple terrain categories corresponding to the medical tasks to be trained and the topography of each area to generate a basic scene. The terrain category of each area can be determined separately based on the multiple terrain categories corresponding to the medical tasks to be trained and the distribution probability of the terrain category corresponding to the terrain category of the topography of each area; according to the topography of each area and the corresponding terrain category, multiple terrain models of each area are selected from the preset model library to generate a basic scene.

[0122] In this embodiment, the feature categories for each region are determined by combining the distribution probabilities of the various feature categories corresponding to the medical tasks to be trained and the feature categories corresponding to the terrain categories of the topography. This allows for the appropriate selection of feature categories within each region. Feature categories with high distribution probabilities have a higher probability of being selected, while feature categories with low distribution probabilities may have a lower probability of being selected.

[0123] There may be multiple types of land features in an area. For land feature types with high distribution probability, the number of possible appearances is greater; for land feature types with low distribution probability, the number of possible appearances is smaller.

[0124] For the same type of land feature, the terrain and landforms of each area are different, and the specific land feature models that are suitable will also be different. You can select the land feature model that matches the terrain and landforms in the land feature category to generate the basic scene to build a simulation environment that conforms to the actual situation.

[0125] In some embodiments, after generating multiple wounded persons in multiple areas based on the basic scenario of setting the radiation environment and the medical tasks to be trained and generating the injury condition of each wounded person, the current training time and rescue operations performed by the trainees can also be obtained; the injury condition of each wounded person is dynamically evolved based on the current training time and rescue operations.

[0126] In this embodiment, to more realistically simulate the medical mission scenario, the patient's injury can be dynamically evolved to improve the authenticity and effectiveness of the medical mission simulation training. Trainers are individuals participating in the medical mission simulation training. Current training duration refers to the time the trainer has spent training from the start of training to the current moment. Completed rescue operations include various actions taken by the trainer during training, such as wound care, patient transfer, and diagnosis using medical equipment.

[0127] During the current training duration, the casualty's condition can change accordingly with the trainer's rescue operations. Due to the large number of casualties in the scenario, trainers may not be able to search and rescue all of them in a timely manner. Over time, the condition of some casualties may worsen due to lack of timely treatment. In certain radiation environments, the effects of radiation on the casualties may gradually become apparent and worsen over time.

[0128] The trainees may provide treatment to some of the injured. Correct and effective rescue operations, such as timely stopping of bleeding, bandaging of wounds, and radiation protection, may improve the condition of the injured; while incorrect or untimely operations may worsen the condition of the injured.

[0129] Therefore, according to the wounded selected by the trainers and the treatment provided to the wounded, the dynamic changes of each wounded's condition can be simulated accordingly, making the training scenario more realistic and challenging.

[0130] For example, during forward emergency support missions, as the duration of first aid increases, the casualty's condition may worsen due to exposure to the environment. For example, a casualty whose bleeding is not promptly stopped may go into shock due to excessive blood loss. A wound left untreated for a long time may also become infected, worsening the condition. The casualty's environmental conditions can also alter the severity of their injury. For example, in a radioactive environment, the casualty's radiation dose increases over time, potentially causing radiation sickness and impairing organ function. Exposure to high temperatures can lead to dehydration and heatstroke, exacerbating existing injuries.

[0131] On the other hand, during forward emergency support missions, inaccurate injury assessments can lead to inappropriate emergency treatment measures, worsening the casualty's condition. Accurate and effective treatment, such as timely blood transfusions and proper fixation of fractures, can control and improve the casualty's condition. Furthermore, due to the large number of casualties on-site, misjudging the injury severity of all casualties can lead to the condition of those who aren't treated promptly worsening over time, further impacting their overall health.

[0132] During the evacuation of contaminated casualties, the jolting of the vehicle or stretcher during transport can cause secondary injuries to the casualty, worsening their condition. For example, a casualty with a head injury may suffer further brain damage from the jolting, worsening their coma. Prolonged transport can delay optimal treatment, worsening their condition. The temperature and humidity of the transport environment can also affect the patient's condition. Transporting in cold environments can lead to hypothermia, which can affect metabolism and blood circulation, worsening their condition.

[0133] Furthermore, when prioritizing patients for transfer, prioritizing critically ill patients allows them to reach medical centers for specialized treatment as quickly as possible, effectively preventing their condition from worsening. While critically injured patients are not currently in danger of death, their condition may worsen if they are not promptly transferred due to wound infection or prolonged periods of ineffective treatment. For patients with relatively stable conditions, delayed transfer generally does not significantly impact their condition. However, prolonged waits for transfer may cause these patients to experience anxiety, stress, and other emotional issues, leading to psychological problems and potentially hindering their recovery.

[0134] The embodiment of the present application divides the original scene in the simulation training into multiple areas by obtaining the medical tasks to be trained for nuclear emergency medical rescue training, and generates the topography of the multiple areas, so that the partitions of the original scene can be adapted to the medical tasks to be trained so as to train for each medical task; through the multiple types of land objects corresponding to the medical tasks to be trained and the topography of each area, multiple land object models of each area are selected from the preset model library to generate a basic scene, and land objects on the terrain, such as buildings, roads and trees, can be generated according to the adaptability of the topography of each area to transform the virtual scene and increase the scene Diversity; by simulating the radiation environment in each area of ​​the basic scenario, the radiation situation at the nuclear emergency medical rescue site can be simulated to carry out the medical tasks to be trained and improve the authenticity of the scenario; by generating multiple casualties in multiple areas based on the basic scenario and the medical tasks to be trained, and generating the injuries of each casualty, the radiation environment and the basic scenario can be taken into account when generating the casualties and their injuries, so that the casualties and their injuries are adapted to the scenario, increasing the diversity of the casualties in the scenario, thereby enriching the formed virtual scenario and improving the immersion of the trainees, so as to better carry out nuclear emergency medical rescue training. This application generates corresponding virtual scenarios by the medical tasks to be trained, which can carry out diversified training for the medical tasks to be trained; at the same time, different medical tasks to be trained can generate virtual scenarios with different task requirements, which can realize the training of various types of medical tasks and meet the complex requirements of multiple scenarios and multiple tasks in nuclear emergency medical rescue. During the training process of the trainees, the injuries of the casualties are dynamically evolved according to the trainees' operations to improve the authenticity, effectiveness and complexity of the medical task simulation training, thereby improving the practical ability of the relevant personnel.

[0135] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0136] Figure 4 Schematic diagram of the server provided in the embodiment of the present application. Figure 4 As shown, the server 40 of this embodiment includes a processor 41 and a memory 42. The memory 42 stores a computer program 43. When the processor 41 executes the computer program 43, the steps of the above-described method embodiments are implemented. Alternatively, when the processor 41 executes the computer program 43, the functions of the modules / units in the above-described device embodiments are implemented.

[0137] For example, computer program 43 may be divided into one or more modules / units, one or more of which are stored in memory 42 and executed by processor 41 to implement the present application. One or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of computer program 43 in server 40.

[0138] The server 40 may include, but is not limited to, a processor 41 and a memory 42. Those skilled in the art will appreciate that Figure 4 This is only an example of the server 40 and does not constitute a limitation on the server 40. The server 40 may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0139] The processor 41 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0140] The memory 42 can be an internal storage unit of the electronic device 40, such as a hard drive or memory of the electronic device 40. The memory 42 can also be an external storage device of the electronic device 40, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the electronic device 40. Furthermore, the memory 42 can include both an internal storage unit of the electronic device 40 and an external storage device. The memory 42 is used to store computer programs and other programs and data required by the electronic device. The memory 42 can also be used to temporarily store data that has been output or is about to be output.

[0141] For the sake of convenience and brevity, the division of the above functional modules / units is only used as an example. In actual applications, the above functions can be assigned to different functional modules / units as needed. The above modules / units can be implemented in the form of hardware, software, or a combination of hardware and software.

[0142] The embodiment of the present invention also provides a system, including a VR device and a server, which can be found in Figure 1 The description in the above embodiments will not be repeated here.

[0143] Optionally, the system can also include an electronic classroom, which is an area for theoretical teaching and can be equipped with equipment such as a demonstration system, a multimedia podium, a virtual simulation training terminal, and a dynamic platform.

[0144] The demonstration system is used for centralized teaching and for observing virtual training sessions. The multimedia podium can be used for teaching. The virtual simulation training terminal can be used by trainers and trainees for independent learning and exploration, accessing learning materials, completing individual learning tasks, and consolidating and expanding knowledge. The dynamic platform allows trainers and trainees to conduct individual specialized training.

[0145] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in the above-mentioned method embodiments.

[0146] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the methods in the above-mentioned method embodiments.

[0147] The computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. Computer-readable media may include any entity or device capable of carrying computer program code, recording media, USB flash drives, mobile hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media.

[0148] In the above embodiments, the descriptions of each embodiment have their own focus. For parts not described or recorded in detail in one embodiment, please refer to the relevant descriptions of other embodiments. Unless otherwise specified or there is a logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other. The technical features of different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0149] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for dynamically generating a multi-task scenario, characterized in that: include: Obtain medical tasks to be trained for nuclear emergency medical rescue training; Each medical mission corresponds to a zoning method, multiple terrain categories, and multiple ground feature categories; Divide the original scene into multiple areas according to the partitioning method corresponding to the medical task to be trained and multiple terrain categories, and generate the terrain features of the multiple areas; According to the multiple ground object categories corresponding to the medical task to be trained and the topography of each area, multiple ground object models of each area are selected from a preset model library to generate a basic scene; simulating a radiation environment in each area of ​​the base scenario; Based on the basic scenario of setting the radiation environment and the medical task to be trained, a plurality of wounded persons are generated in the plurality of areas, and the injury condition of each wounded person is generated.

2. The multi-task scene dynamic generation method according to claim 1, characterized in that: The partitioning methods include continuous partitioning and splicing partitioning; The method of dividing the original scene into multiple regions according to the partitioning method and multiple terrain categories corresponding to the medical task to be trained, and generating the topography of the multiple regions, includes: If the partitioning method corresponding to the medical task to be trained is continuous partitioning, randomly selecting at least one from the multiple terrain categories corresponding to the medical task to be trained to generate the topography of the original scene; partitioning is performed based on the characteristics of the topography in the original scene to obtain multiple regions and the topography of each region; If the partitioning method corresponding to the medical task to be trained is splicing partitioning, multiple task requirements are determined according to the medical task to be trained; according to the task requirements, the original scene is partitioned to obtain at least one area corresponding to each task requirement; according to the task requirements of each area and the multiple terrain categories corresponding to the medical task to be trained, the terrain and landforms of each area are generated respectively.

3. The method for dynamically generating a multi-task scenario according to claim 1, wherein: The step of generating a plurality of injured persons in the plurality of areas based on the basic scenario of the radiation environment and the medical task to be trained, and generating the injury condition of each injured person, includes: Determining the initial number and initial positions of the wounded based on the population density of the terrain model in the basic scenario, the radiation environment, and the medical task to be trained; Generating an initial injury condition of each injured person based on a terrain model and radiation environment at the initial position of each injured person; the initial injury condition includes an injury level; The preset injury statistical model corresponding to the medical task to be trained is used to adjust the initial number and injury levels of all the injured to obtain the adjusted number of injured and the injury condition of each injured.

4. The method for dynamically generating a multi-task scenario according to claim 3, wherein: After generating a plurality of wounded persons in the plurality of areas and generating the injury condition of each wounded person based on the basic scenario of the radiation environment and the medical task to be trained, the method further includes: Obtain the current training time and rescue operations performed by the trainer; The injury condition of each injured person is dynamically evolved according to the current training duration and the rescue operation.

5. The method for dynamically generating a multi-task scenario according to any one of claims 1 to 4, characterized in that: Each random number corresponds to a meteorological category, and each meteorological category corresponds to a parameter selection range of meteorological elements; Before simulating the radiation environment in each area of ​​the basic scene, the method further includes: Generate a random number within the preset data range; Determining the meteorological category corresponding to the random number as the meteorological category of the basic scene; Determining the parameter values ​​of the meteorological elements in each area according to the partitioning method corresponding to the medical task to be trained and the parameter selection range corresponding to the meteorological category; Adjust the parameter values ​​of meteorological elements in each area according to the topography of each area; Based on the adjusted parameter values ​​of meteorological elements in each region, meteorological simulation is performed in each region to generate the meteorological environment.

6. The method for dynamically generating a multi-task scenario according to any one of claims 1 to 4, characterized in that: The object model in the basic scene includes a building model; The radiation environment is simulated in each area of ​​the base scenario, including: Determining radiation information of a radioactive source in each partition of the basic scene according to a partitioning method corresponding to the medical task to be trained; Determining, based on the radiation information corresponding to each partition, the radiation diffusion range outside the building model in each partition and the first radiation intensity at each position outside the building model; Determining the second radiation intensity at each location within each building model based on the radiation information corresponding to each partition and the building model within each partition; Based on the radiation diffusion range, the first radiation intensity, and the second radiation intensity, a radiation environment in each area of ​​the basic scene is simulated.

7. The method for dynamically generating a multi-task scenario according to any one of claims 1 to 4, characterized in that: The distribution probability of each terrain category corresponding to a ground feature category; The method comprises selecting multiple ground object models for each area from a preset model library according to the multiple ground object categories corresponding to the medical task to be trained and the topography of each area to generate a basic scene, including: Determine the ground feature category of each area according to the multiple ground feature categories corresponding to the medical task to be trained and the distribution probability of the ground feature categories corresponding to the terrain category of the topography of each area; According to the topography and corresponding land feature categories of each area, multiple land feature models of each area are selected from the preset model library to generate a basic scene.

8. A server, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

9. A system, characterized in that: The device comprises a VR device and the server as claimed in claim 8.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images