Virtual reality-based workshop virtual scene rendering method and system

Abstract

The application provides a kind of workshop virtual scene rendering method and system based on virtual reality, it is related to industrial manufacturing and virtual reality technical field, first execute workshop multi-source data collaborative access processing and generate multi-source data collaborative access dataset, based on real-time video stream and internet of things operation data in multi-source data collaborative access dataset generate scene dynamic adaptation parameter set.Layered rendering module is called, and computer aided design layout data and parameter set are combined to carry out layered virtual-real fusion rendering, to generate initial rendering result.Real-time feedback information is collected, parameter set is iteratively adjusted, optimized parameter set is generated and is substituted into rendering to obtain final result, and based on final result and equipment state data, equipment three-dimensional model state synchronous update is driven.The application improves the precision and response speed of workshop virtual scene rendering, realizes real-time synchronization of equipment state, and helps efficient management of workshop.

Description

Technical Field

[0001] This invention relates to the fields of industrial manufacturing and virtual reality technology, and more specifically, to a method and system for rendering virtual workshop scenes based on virtual reality. Background Technology

[0002] As the core location for production activities, the efficient monitoring and precise management of the workshop's operational status are crucial for improving production efficiency and ensuring product quality. With the continuous development of virtual reality technology, constructing virtual workshop scenarios to achieve intuitive display and simulation analysis of the workshop's operational status has become an effective means.

[0003] However, existing methods for rendering virtual workshop scenes have several shortcomings. Firstly, workshop data comes from a wide range of sources, including real-time video streams, IoT operational data, computer-aided design layout data, and equipment status data. However, this multi-source data is often isolated, lacking an effective collaborative processing mechanism, making it difficult to fully leverage the comprehensive value of the data. This results in virtual scenes failing to accurately reflect the actual operation of the workshop. Secondly, during virtual scene rendering, traditional methods typically use fixed scene parameters, unable to intelligently adjust according to real-time dynamic changes in the workshop. This leads to significant deviations between the rendered results and the actual scene, low image blending accuracy, and slow scene response speed, failing to meet the needs of real-time workshop monitoring and management. Furthermore, existing methods lack a dynamic update mechanism for equipment status after rendering, failing to reflect the actual operating status of workshop equipment in a timely manner, thus reducing the practicality and reliability of the virtual scene. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for rendering a virtual workshop scene based on virtual reality, the method comprising: The workshop multi-source data collaborative access processing is performed. The dependency structure between multi-source data is established through the data middleware system, and a multi-source data collaborative access dataset is generated. The workshop multi-source data includes workshop real-time video stream, workshop IoT operation data, workshop computer-aided design layout data, and workshop equipment status data. Based on the real-time video stream characteristics of the workshop and the workshop IoT operation data in the multi-source data collaborative access dataset, intelligent adaptation processing is performed on the scene parameters of the preset workshop virtual scene to generate a dynamic scene adaptation parameter set. The layered rendering module of the virtual reality rendering system is invoked. The workshop computer-aided design layout data in the multi-source data collaborative access dataset and the scene dynamic adaptation parameter set are combined to perform a layered virtual-real fusion rendering operation on the real-time video stream of the workshop and the preset workshop virtual scene to generate the initial workshop virtual scene rendering result. Real-time feedback information of the initial workshop virtual scene rendering result is collected. The real-time feedback information includes image fusion accuracy information and scene response speed information. Based on the real-time feedback information, the scene dynamic adaptation parameter set is iteratively adjusted to generate an optimized scene dynamic adaptation parameter set. The optimized scene dynamic adaptation parameter set is substituted into the layered virtual-real fusion rendering operation to generate the final workshop virtual scene rendering result. Based on the final workshop virtual scene rendering result and workshop equipment status data, the workshop equipment 3D model is driven by a finite state machine to perform synchronous status updates.

[0005] Furthermore, embodiments of the present invention also provide a virtual workshop scene rendering system based on virtual reality, characterized in that it includes: A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described virtual reality-based workshop virtual scene rendering method by executing the machine-executable instructions.

[0006] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of a virtual reality-based workshop virtual scene rendering system reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the virtual reality-based workshop virtual scene rendering system to execute the aforementioned virtual reality-based workshop virtual scene rendering method.

[0007] Based on the above, by performing multi-source data collaborative access processing in the workshop, a dependency structure between multi-source data is established using a data platform system, generating a multi-source data collaborative access dataset. This fully integrates multi-source information such as real-time video streams from the workshop, IoT operation data, computer-aided design layout data, and equipment status data. Based on the characteristics of real-time video stream images and IoT operation data from the multi-source data, intelligent adaptation processing is performed on the scene parameters of the preset workshop virtual scene, generating a dynamic scene adaptation parameter set. This allows the virtual scene to automatically adjust parameters according to the real-time dynamic changes in the workshop, effectively improving the matching degree between the rendering results and the actual scene. The layered rendering module of the virtual reality rendering system is invoked, combined with computer-aided design layout data... By dynamically adapting the parameter set to the scene, a layered virtual-real fusion rendering operation is performed on the real-time video stream of the workshop and the preset virtual scene to generate an initial rendering result, which further improves the accuracy of image fusion and the response speed of the scene. Real-time feedback information of the initial rendering result is collected and the scene dynamically adapting parameter set is iteratively adjusted to generate an optimized parameter set. The rendering effect is continuously optimized to ensure the accuracy and stability of the rendering result. The optimized parameter set is substituted into the rendering operation to generate the final rendering result of the workshop virtual scene. Based on this result and the equipment status data, the workshop equipment 3D model is driven by a finite state machine to perform state synchronization update, realizing real-time synchronization between the virtual scene and the actual workshop equipment status, which significantly improves the efficiency of workshop operation and management. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the execution flow of the virtual workshop scene rendering method based on virtual reality provided in the embodiment of the present invention.

[0009] Figure 2 This is a schematic diagram of exemplary hardware and software components of a virtual reality-based workshop virtual scene rendering system provided in an embodiment of the present invention. Detailed Implementation

[0010] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating a virtual workshop scene rendering method based on virtual reality provided in one embodiment of the present invention. The following is a detailed description of the virtual workshop scene rendering method based on virtual reality.

[0011] Step S110: Perform workshop multi-source data collaborative access processing, establish the dependency structure between multi-source data through the data middleware system, and generate a multi-source data collaborative access dataset. The workshop multi-source data includes workshop real-time video stream, workshop IoT operation data, workshop computer-aided design layout data, and workshop equipment status data.

[0012] This embodiment uses a large-scale hydropower equipment manufacturing workshop as an application scenario. This workshop is primarily responsible for the processing and manufacturing of core hydropower equipment such as turbine runners and guide vanes. It includes a processing area (equipped with multi-axis CNC machine tools, large vertical lathes, etc.), a storage area (storing raw materials and semi-finished products), a passageway area (logistics channels connecting various functional areas), and an assembly and debugging area (for final assembly and performance testing of equipment components). When performing collaborative access processing of multi-source data within the workshop, it is necessary to integrate various scattered data resources. The data platform system, as the core hub for data interaction, needs to uniformly manage different types of data and build dependencies between data to ensure the synergy of data calls and updates. For example, changes in equipment status data in the processing area need to trigger updates to the corresponding 3D model of the equipment in the virtual scene, and video stream data in the storage area needs to be combined with 3D layout data to achieve a virtual-real fusion display.

[0013] Step S111: Call the video access interface to receive raw video streams transmitted from multiple camera devices in the workshop. The video access interface supports real-time streaming protocols and proprietary protocols of multiple mainstream manufacturers' software development kits. Perform protocol conversion processing on the raw video streams of different protocols to generate a unified protocol video stream.

[0014] In this hydropower equipment production workshop, high-definition cameras supporting real-time streaming protocols are deployed next to the CNC machine tools in the processing area to monitor the cutting process. Intelligent cameras using proprietary protocols with infrared night vision capabilities are installed in the shelving aisles of the storage area. The assembly and debugging area is equipped with panoramic cameras supporting multiple protocols. The video access interface has a built-in protocol parsing module. For real-time streaming protocol video streams transmitted from the cameras in the processing area, the format is directly verified before proceeding to the next processing step. For proprietary protocol video streams from the storage area, the corresponding manufacturer's software development kit (SDK) parsing plugin is used to extract video frame data and encoding information, converting them to a real-time streaming protocol format. The panoramic video streams from the assembly and debugging area are parsed dynamically based on their protocol type. During protocol conversion, the frame rate, resolution, and other parameters of the video stream must remain unchanged; only the data encapsulation format is converted to generate a unified protocol video stream, ensuring compatibility with subsequent processing modules.

[0015] Step S112: Perform preloading processing on the unified protocol video stream, determine the number of preloaded video frames based on historical rendering requirement data, read the corresponding number of video frames in advance and store them in the cache area to generate a preloaded video stream.

[0016] After the unified protocol video stream is generated, preloading is required to avoid stuttering or delays during rendering. Historical rendering demand data includes the video stream call frequency, average rendering time, and smoothness feedback for different time periods (such as equipment start-up and shutdown periods, peak processing periods, and lunch breaks) over the past three months. Analysis of historical data reveals that the processing area experiences higher video frame call frequency during equipment start-up and shutdown periods, necessitating the preloading of more video frames to handle sudden rendering demands. Conversely, the call frequency is lower during lunch breaks, allowing for a reduction in the number of preloaded frames. Based on this, a dynamic adjustment rule for the number of preloaded video frames is established: when the real-time video stream frame rate exceeds a preset threshold, the number of preloaded video frames increases; when the frame rate falls below the preset threshold, the number of preloaded frames decreases. Preloaded video frames are stored in a circular queue in the cache area according to their timestamp order. When a new video frame arrives, it overwrites the oldest expired frame, ensuring that the cache area always stores the latest preloaded video stream data.

[0017] Step S113: Deploy an IoT data receiving gateway to receive raw IoT data transmitted from temperature sensors, vibration sensors, and humidity sensors in the workshop. Perform data cleaning processing on the raw IoT data to remove duplicate and invalid data and generate cleaned IoT data.

[0018] In the processing area of ​​the hydropower equipment production workshop, temperature and vibration sensors are installed on key components such as the spindle box and guide rails of each CNC machine tool; temperature and humidity sensors are deployed in the storage area; and humidity sensors are also installed at the air outlets of the ventilation system in the passageway area. The IoT data receiving gateway adopts a distributed deployment approach, setting up sub-gateways in each functional area, connecting to sensors via industrial Ethernet, and supporting industrial bus protocols such as Modbus and Profinet. The received raw IoT data may contain duplicate data due to sensor malfunctions (e.g., a temperature sensor continuously sending three identical temperature values ​​within one second) and invalid data due to communication interference (e.g., vibration sensors returning values ​​exceeding the normal measurement range). During data cleaning, the raw data is first sorted by timestamp. For duplicate data from the same sensor at the same timestamp, the first data entry is retained; for invalid data with values ​​exceeding the preset normal range, the average of the data from preceding and following times is used for replacement; for sensor data with a data loss rate exceeding a preset proportion, it is marked as abnormal and a sensor status check is triggered. After the above processing, cleaned IoT data containing parameters such as temperature, vibration, and humidity is generated. The data format is uniformly JSON, including sensor ID, acquisition timestamp, parameter name, and parameter value fields.

[0019] Step S114: Obtain the original data of the computer-aided design layout of the workshop, divide the original data of the computer-aided design layout of the workshop into processing area layout data, storage area layout data and passage area layout data according to the functional areas of the workshop, add spatial coordinate information to each area layout data, and generate partition layout data with coordinate labels.

[0020] The original data for the workshop computer-aided design layout consisted of 2D drawings in AutoCAD format and 3D model files in Revit format, containing information such as the location, dimensions, and relative layout relationships of equipment within the workshop. First, the original data was converted from 2D drawings to planar layout data in a 3D coordinate system, and then matched and aligned with the 3D model file. Following functional area division rules, with the workshop's central axis as the reference, the left side was designated as the processing area, containing 10 CNC machine tools, 5 milling machines, and 3 grinding machines; the right side was designated as the storage area, containing 8 automated storage and retrieval systems, 2 stacker cranes, and 1 material storage area; the middle area was designated as the aisle area, 3 meters wide, connecting the processing and storage areas, and including 2 turning aisles. When adding spatial coordinate information to the layout data for each area, a geodetic coordinate system was used, with the northwest corner of the workshop as the origin (0, 0, 0). The X-axis extended along the length of the workshop, the Y-axis along the width, and the Z-axis along the height. The coordinates of a CNC machine tool in the machining area are defined as a three-dimensional coordinate range from (X1, Y1, Z1) to (X2, Y2, Z2). The coordinates of the shelves in the storage area are defined as (X3, Y3, Z3) to (X4, Y4, Z4). The coordinates of the aisle area are defined as (X5, Y5, Z5) to (X6, Y6, Z6). The partition layout data with coordinate labels is stored in an XML format file, containing fields such as area name, coordinate range, and a list of included equipment.

[0021] Step S115: Receive workshop equipment status data transmitted by the workshop equipment controller. The workshop equipment status data includes equipment operating status, equipment location information and equipment working parameters. Perform timestamp marking processing on the workshop equipment status data to generate timestamped equipment status data.

[0022] The workshop equipment controllers include PLC controllers for CNC machine tools, motion controllers for stacker cranes, and frequency converters for ventilation systems, all connected to the data receiving module via an industrial bus. Equipment operating status includes four states: "Running," "Stopped," "Fault," and "Maintenance." Equipment location information is real-time coordinates for mobile equipment (such as stacker cranes) and installation position coordinates for fixed equipment (such as CNC machine tools). Equipment operating parameters include spindle speed, feed rate, and depth of cut for CNC machine tools; running speed and lifting height for stacker cranes; and airflow and air pressure for ventilation systems. Received equipment status data is a binary data stream and needs to be parsed into structured data. Timestamp processing uses a network time protocol to ensure millisecond-level time synchronization accuracy for all data. For periodically transmitted equipment status data (such as spindle speed data sent every 100 milliseconds), a timestamp is automatically added upon reception; for non-periodic data (such as equipment fault alarm data), a timestamp is added immediately upon data triggering. Timestamped equipment status data is stored in a relational database, indexed by equipment ID and timestamp for easy subsequent querying and correlation analysis.

[0023] Step S116: Input the preloaded video stream, cleaned IoT data, partition layout data with coordinate identifiers, and device status data with timestamps into the data platform system.

[0024] After the above four types of data are generated, they are input into the data platform system through the data transmission interface. The data transmission interface adopts a dual-channel mechanism: the preloaded video stream data, which is large in volume, is transmitted via UDP protocol to improve speed; the cleaned IoT data, partition layout data with coordinate identifiers, and device status data with timestamps are structured or semi-structured data, and are transmitted via TCP protocol to ensure reliability. The receiving buffer of the data platform system performs format verification on the input data. For data that does not conform to the preset format (such as device status data missing key fields), an error code is returned and retransmission is required; the verified data is stored in the corresponding data partition according to the data type: video stream data is stored in a distributed file system, structured data is stored in a relational database, and semi-structured data is stored in a document database.

[0025] Step S117: Analyze the relationships between various data types through the dependency construction module of the data platform system, and establish a dependency structure that includes data dependency direction and update triggering conditions.

[0026] Step S1171: The dependency construction module of the data platform system performs data type analysis on the preloaded video stream, cleaned IoT data, partition layout data with coordinate identifiers, and device status data with timestamps to determine the attributes and uses of each data type.

[0027] Data type analysis employs metadata parsing technology to read description files for various data types (e.g., metadata for video streams includes resolution, frame rate, and encoding format; metadata for IoT data includes sensor type, parameter units, and sampling frequency). The data type attribute for preloaded video streams is "streaming media data," used to "provide real-time image texture"; the attribute for cleaned IoT data is "time-series sensor data," used to "provide environmental and device operating parameters"; the attribute for partition layout data with coordinate identifiers is "spatial geometric data," used to "construct a hierarchical structure of a 3D scene"; and the attribute for device status data with timestamps is "device status data," used to "drive device model status updates." Analysis results are stored in a data type dictionary.

[0028] Step S1172: Analyze the relationship between the preloaded video stream and the partition layout data with coordinate identifiers. The projection of the preloaded video stream depends on the spatial coordinate information of the partition layout data with coordinate identifiers. Therefore, the data dependency direction is determined to be that the partition layout data with coordinate identifiers points to the preloaded video stream.

[0029] Texture projection of a preloaded video stream requires knowing the spatial coordinates of each region in the 3D scene to determine the projection position and range. Therefore, partition layout data with coordinate identifiers is a prerequisite for preloaded video stream projection. Analysis of the data usage flow reveals that when processing a preloaded video stream, the layered rendering module first needs to access the partition layout data with coordinate identifiers to obtain the region coordinates before performing texture projection calculations. Therefore, the data dependency direction is: partition layout data with coordinate identifiers → preloaded video stream; that is, the partition layout data with coordinate identifiers is the source data, and the preloaded video stream is the dependent data.

[0030] Step S1173: Analyze the correlation between the cleaned IoT data and the scene dynamic adaptation parameter set. The generation of the scene dynamic adaptation parameter set depends on the humidity and temperature information in the cleaned IoT data. Determine the data dependency direction as the cleaned IoT data pointing to the scene dynamic adaptation parameter set.

[0031] The lighting parameters and particle system parameters in the scene dynamic adaptation parameter set need to be adjusted according to the ambient humidity and device temperature, and these parameters are derived from the cleaned IoT data. The processing flow of the scene dynamic adaptation parameter generation module is as follows: input cleaned IoT data → extract humidity and temperature parameters → calculate global humidity and temperature parameters → generate adaptation parameters. Therefore, the data dependency direction is cleaned IoT data → scene dynamic adaptation parameter set, where cleaned IoT data is the source data and the scene dynamic adaptation parameter set is the dependent data.

[0032] Step S1174: Analyze the correlation between the timestamped device status data and the device 3D model status. The update of the device 3D model status depends on the timestamped device status data. Determine the data dependency direction as the timestamped device status data pointing to the device 3D model status.

[0033] The state of the equipment's 3D model (e.g., running, stopped, faulty) needs to be updated based on the actual state of the equipment, which is provided by timestamped equipment state data. The workflow of the finite state machine module is as follows: receive timestamped equipment state data → parse the actual operating state → trigger state transitions → update the state of the equipment's 3D model. Therefore, the data dependency direction is timestamped equipment state data → equipment 3D model state, where the timestamped equipment state data is the source data and the equipment 3D model state is the dependent data.

[0034] Step S1175: Set update trigger conditions for each data dependency. When the spatial coordinates of the partition layout data with coordinate identifiers are adjusted, the texture projection range of the preloaded video stream is updated. When the change in humidity or temperature data in the IoT data after cleaning reaches the preset change threshold, the scene dynamic adaptation parameter set is regenerated. When the actual state of the device in the device status data with timestamps changes, the state of the device 3D model is updated.

[0035] Update trigger conditions are set based on data characteristics and business needs: For spatial coordinate adjustments in partitioned layout data with coordinate identifiers, the trigger condition is a coordinate change exceeding 1 meter; for IoT data after cleaning, the preset threshold for humidity change is 5%, and the preset threshold for temperature change is 2℃; for device status changes in timestamped device status data, the trigger condition is a status code change from "running" to "stopped," "faulted," or a change from "stopped" to "running." These trigger conditions are stored in a dependency rule base. The data platform system monitors source data changes in real time and executes the corresponding dependency data update operation when the trigger condition is met.

[0036] Step S1176: Integrate the data dependency direction with the corresponding update triggering condition to construct a dependency relationship structure with data type as nodes and dependency relationship as edges. The dependency relationship structure is stored in a graphical manner.

[0037] The dependency structure is represented by a directed graph data structure. Nodes represent data types (preloaded video streams, cleaned IoT data, partitioned layout data with coordinate identifiers, timestamped device status data, scene dynamic adaptation parameter sets, and device 3D model status). Directed edges represent the direction of data dependencies, and edge attributes are update trigger conditions. Graphical storage uses both adjacency lists and JSON files. Adjacency lists are used for quick dependency lookups within the data platform system, while JSON files are used for visualization and manual configuration. After the dependency structure is built, it is registered to the dependency management module of the data platform system, supporting dynamic updates and queries.

[0038] Step S118: Start the data subscription module of the data platform system, configure subscription nodes for each type of data, and when a certain type of data is updated, the subscription node will automatically receive the update notification and synchronize the update of the related data.

[0039] The data subscription module of the data platform system adopts a publish-subscribe model, supporting multiple topic subscriptions. Independent subscription topics are created for pre-loaded video streams, cleaned IoT data, partitioned layout data with coordinate identifiers, and device status data with timestamps. Each subscription topic corresponds to multiple subscription nodes. For example, the subscription nodes for pre-loaded video streams include the layered rendering module and the image fusion accuracy detection module of the virtual reality rendering system; the subscription nodes for cleaned IoT data include the scene dynamic adaptation parameter generation module and the device fault early warning module. Subscription nodes are bound to subscription topics through a registration mechanism, during which the notification method for data updates (such as message queue notification or callback function notification) must be specified. When a certain type of data is updated, the publishing module of the data platform system sends the update information (including data ID, update timestamp, and changed fields) to the corresponding subscription topic. The subscription node listens for topic messages and receives the update notification, then pulls the latest data from the data platform system according to the notification content, completing the synchronous update of related data.

[0040] Step S119: Receive data requests for subsequent rendering steps through the request forwarding module of the data platform system, retrieve the corresponding data source according to the dependency structure, obtain the data, and return it to the request node.

[0041] Subsequent rendering steps (such as layered virtual-real fusion rendering operations) send data requests to the data platform system via a data request interface during execution. The data request includes the requested data type, request parameters (such as device ID, time range, and region name), and data format requirements. Upon receiving the request, the request forwarding module parses the request parameters and queries the dependency structure to determine the data source for the required data. For example, when the virtual reality rendering system requests video stream data for a processing area, the request forwarding module retrieves the processing area coordinates of the partition layout data with coordinate identifiers that the data depends on based on the dependency structure. Therefore, it first queries the partition layout data with coordinate identifiers to obtain the coordinate range of the processing area, and then filters the video stream data for the corresponding region from the pre-loaded video stream data source based on the coordinate range. During data acquisition, if the data source is a local cache (such as frequently requested partition layout data), it is returned directly from the cache; if it is a remote database, the data is queried through the database access interface. The acquired data is converted according to the request format requirements and returned to the request node to ensure that the request node can use it directly.

[0042] Step S1110: Integrate all the data after dependency association and subscription synchronization processing to generate a multi-source data collaborative access dataset.

[0043] After the aforementioned dependency association and subscription synchronization processing, various data types have formed an organic whole. The data integration process includes data consistency verification, redundant data removal, and data index construction. Data consistency verification ensures accurate data association by comparing the consistency of related fields in different data types (such as the device ID in device status data and video stream data); redundant data removal deletes duplicate storage or data that has exceeded its retention period (such as historical video stream data older than 7 days); data index construction establishes multi-dimensional indexes based on data type, timestamp, region name, and other dimensions to improve data query efficiency. The integrated multi-source data collaborative access dataset is stored in the form of a data warehouse, supporting batch queries, real-time retrieval, and correlation analysis.

[0044] Step S120: Based on the real-time video stream characteristics of the workshop and the workshop IoT operation data in the multi-source data collaborative access dataset, perform intelligent adaptation processing on the scene parameters of the preset workshop virtual scene to generate a scene dynamic adaptation parameter set.

[0045] The preset virtual workshop scene is a 3D model constructed based on partitioned layout data with coordinate markers, containing basic scene parameters. In the virtual scene rendering of the hydropower equipment production workshop, to ensure the virtual scene realistically reflects changes in the actual workshop environment, scene parameters need to be intelligently adapted based on real-time video stream characteristics and IoT operational data. For example, when workshop lighting conditions weaken due to weather changes, the virtual scene's lighting parameters need to be adjusted to match the actual brightness; when equipment operating temperature rises, the material parameters of the equipment's 3D model need to be adjusted to simulate thermal radiation effects. Intelligent adaptation processing establishes a mapping relationship between image features, IoT data, and scene parameters, enabling dynamic adjustment of scene parameters and generating a dynamic scene adaptation parameter set.

[0046] Step S121: Extract the preloaded video stream from the multi-source data collaborative access dataset, select multiple consecutive video frames to form a video frame sequence, perform color feature extraction on each video frame, and obtain the hue and color temperature features of each video frame.

[0047] When extracting preloaded video streams from a multi-source collaborative access dataset, the extraction area needs to be determined based on the current rendering requirements. For example, if the current rendering requirement is for a virtual scene in a processing area, then the preloaded video stream corresponding to that processing area should be extracted. The number of consecutive video frames selected is determined by the frame rate of the video stream. If the frame rate is 30 frames per second, 90 video frames within 3 seconds are selected to form a video frame sequence, ensuring that the sequence can reflect short-term changes in image features. Color feature extraction uses an HSV color space-based extraction method, converting each video frame from RGB color space to HSV space. The hue feature corresponds to the H channel value in the HSV space, and the color temperature feature is obtained by converting the H channel value to a preset color temperature mapping table. For each video frame, the average value of the H channel is calculated as the hue feature of that frame. The corresponding color temperature feature is obtained by querying the color temperature mapping table based on the hue feature and stored in a two-dimensional array (hue feature value, color temperature feature value).

[0048] Step S122: Calculate the average hue feature of all video frames in the video frame sequence to obtain the global hue parameter; calculate the average color temperature feature of all video frames to obtain the global color temperature parameter.

[0049] The average hue feature of a video frame sequence is calculated by summing the hue feature values ​​of all video frames in the sequence and then dividing by the number of video frames. If there are abnormal frames in the video frame sequence (such as frames whose hue feature values ​​exceed the normal range due to sudden changes in lighting), outlier removal is required before calculation. The removal rule is: if the hue feature value of a frame deviates from the sequence average by more than three times the standard deviation, it is considered an abnormal frame and excluded. The global hue parameter is the average hue feature value after removing abnormal frames, reflecting the overall hue tendency of the video frame sequence. The calculation process for the global color temperature parameter is similar: summing the color temperature feature values ​​of all video frames and dividing by the number of valid video frames yields the global color temperature parameter, which reflects the overall color temperature level.

[0050] Step S123: Extract the cleaned IoT data from the multi-source data collaborative access dataset, filter out the workshop environmental humidity data and equipment operating temperature data, calculate the average value of the workshop environmental humidity data to obtain the global humidity parameter; calculate the average value of the equipment operating temperature data to obtain the global temperature parameter.

[0051] The IoT data after cleaning contains various sensor parameters, which need to be filtered to extract workshop environmental humidity data and equipment operating temperature data according to the requirements of scene parameter adaptation. Workshop environmental humidity data comes from humidity sensors in the storage area, aisle area, and central control room, while equipment operating temperature data comes from spindle temperature sensors of CNC machine tools in the processing area and oil temperature sensors of the main transformer, etc. During the filtering process, data is filtered by sensor ID and parameter name, retaining data with parameter names of "environmental humidity" and "operating temperature". When calculating the global humidity parameter, all environmental humidity data are aligned by timestamp, the average humidity value at the same timestamp is taken, and then the average values ​​of different timestamps are averaged twice to obtain the global humidity parameter reflecting the overall environmental humidity of the workshop. The calculation method for the global temperature parameter is similar: the equipment operating temperature data is grouped by equipment type, the average temperature of each group is calculated, and then different weights are assigned according to the importance of the equipment (e.g., spindle temperature has a higher weight than guide rail temperature), and a weighted average is taken to obtain the global temperature parameter.

[0052] Step S124: Obtain the basic scene parameters of the preset workshop virtual scene. The basic scene parameters include lighting parameters, particle system parameters, model material parameters, and viewpoint parameters. The lighting parameters include lighting color parameters and lighting intensity parameters. The particle system parameters include particle type parameters and particle quantity parameters. The model material parameters include material reflection parameters and material light transmission parameters. The viewpoint parameters include viewpoint position parameters and viewpoint switching frequency parameters.

[0053] The basic parameters of the preset workshop virtual scene are stored in an XML configuration file and read through the parameter parsing module. The lighting parameters include: lighting color parameter using the RGB color model with a value range of 0-255, and lighting intensity parameter with a value range of 0-1000 candela; particle system parameters include four particle types: "no particles," "dust particles," "smoke particles," and "raindrop particles," with the particle quantity parameter being an integer; material parameters include: material reflection parameter reflecting the material's ability to reflect light with a value range of 0-1, and material transmittance parameter reflecting the material's light transmittance with a value range of 0-1; and viewpoint parameters include: viewpoint position parameter in three-dimensional coordinates (X, Y, Z), and viewpoint switching frequency parameter in the number of switches per minute. The initial values ​​of the basic scene parameters are set based on the workshop design drawings and historical rendering experience; for example, the initial lighting intensity parameter for the processing area is set to 800 candela, and the model material reflection parameter is set to 0.6.

[0054] Step S125: Establish a mapping relationship between global hue parameters and lighting color parameters, determine the corresponding target lighting color parameters based on the global hue parameters, and update the lighting color parameters in the lighting parameters to the target lighting color parameters.

[0055] Step S1251: Collect the tonal characteristics of video frames and the corresponding optimal lighting color parameter samples under different time periods and environments in the workshop. The optimal lighting color parameter samples include multiple tonal ranges and corresponding lighting color values.

[0056] Data collection was conducted over the past year, covering workshop environments across different seasons (spring, summer, autumn, and winter), weather conditions (sunny, cloudy, rainy, and snowy), and time periods (morning, noon, evening, and night). For each time period, 100 sets of video frame tonal features and optimal lighting color parameters were collected, totaling 100 × 4 × 4 × 4 = 64,000 samples. Video frame tonal features were obtained from the real-time video stream using a color extraction tool. Optimal lighting color parameters were determined by professional rendering engineers based on visual effect scoring, using a 5-point scale. Samples with a score ≥ 4.5 were selected as valid samples. Sample data was stored in a sample database, including fields such as sample ID, collection time, tonal feature value, RGB lighting color value, and environment description.

[0057] Step S1252: Classify the collected sample data and divide it into multiple hue intervals according to the numerical range of hue features. Each hue interval corresponds to one or more lighting color parameter samples.

[0058] The hue feature values ​​range from 0 to 360 (H channel values ​​in the HSV color space), and are divided into multiple hue intervals using an equally spaced method, such as [0, 15), [15, 30), ..., [345, 360], for a total of 24 intervals. The sample data are sorted according to their hue feature values, and each sample is assigned to its corresponding hue interval. Each hue interval contains multiple lighting color parameter samples; for example, the [0, 15) interval corresponds to red lighting color samples, and the [180, 195) interval corresponds to cyan lighting color samples. After classification, a hue interval-sample list lookup table is generated and stored in the classification database.

[0059] Step S1253: Calculate the average value of the lighting color parameter samples in each hue range, and determine the average value as the standard lighting color parameter corresponding to that hue range.

[0060] For each hue interval, iterate through the RGB values ​​of all lighting color parameter samples within the interval, and calculate the average values ​​of the red (R), green (G), and blue (B) components respectively. For example, if a hue interval has N samples, the average value of R is (R1+R2+...+RN) / N, and the average values ​​of G and B are calculated in the same way. (R_avg, G_avg, B_avg) are then defined as the standard lighting color parameters for that hue interval and stored in the standard parameter table.

[0061] Step S1254: Organize the hue range and the corresponding standard illumination color parameters into a mapping table, wherein the mapping table includes the start value and end value of the hue range and the corresponding standard illumination color parameters.

[0062] The mapping table uses a two-dimensional array structure, with each row representing a hue range and containing three fields: start value (H_start), end value (H_end), and standard lighting color parameters (R_std, G_std, B_std). For example: [0, 15), (255, 100, 100) [15, 30), (255, 150, 100) ... [345, 360], (255, 100, 200) The mapping table is stored as a CSV file for easy querying and retrieval later.

[0063] Step S1255: Find the hue range to which the global hue parameter belongs from the mapping table, and obtain the standard lighting color parameter corresponding to the hue range.

[0064] The global hue parameter is the average hue feature H_avg calculated in step S122. Based on the value of H_avg, the corresponding hue interval is searched in the mapping table (e.g., if H_avg=20, it belongs to the interval [15, 30)), and the standard lighting color parameters (R_std, G_std, B_std) of that interval are read. The search process uses a binary search algorithm to improve query efficiency.

[0065] Step S1256: Fine-tune the acquired standard lighting color parameters. Combine the regional functional characteristics in the computer-aided design layout data of the workshop. If the area to which the global hue parameter belongs is the processing area, increase the proportion of the red component in the standard lighting color parameters to meet the requirements for the detailed presentation of equipment in the processing area. If the area to which it belongs is the storage area, increase the proportion of the blue component in the standard lighting color parameters to meet the requirements for enhancing the sense of space in the storage area.

[0066] The functional characteristics of each area in the computer-aided design layout data of the workshop include information such as the purpose of each area, equipment type, and lighting requirements. The processing area requires high detail in equipment presentation, necessitating an enhancement of the red component to highlight the texture of the metal processing surface. The fine-tuning rule is: R_new = R_std + ΔR, where ΔR is the increment of the red component (set according to the area characteristics). The storage area requires a high degree of spatial awareness, necessitating an enhancement of the blue component to expand the visual sense of space. The fine-tuning rule is: B_new = B_std + ΔB, where ΔB is the increment of the blue component. The fine-tuned lighting color parameters are (R_new, G_std, B_new) or (R_std, G_std, B_new).

[0067] Step S1257: Determine the fine-tuned standard lighting color parameters as the target lighting color parameters.

[0068] The fine-tuned standard lighting color parameters are tested for visual effects. If they meet the functional characteristics of the area, they are determined as the target lighting color parameters; if not, the values ​​of ΔR or ΔB are readjusted until the requirements are met. The target lighting color parameters are stored in RGB format, with a value range of 0-255.

[0069] Step S1258: Extract lighting parameters from the basic scene parameters of the preset workshop virtual scene, update the lighting color parameters in the lighting parameters to the target lighting color parameters, and complete the update of the lighting color parameters.

[0070] The basic parameters of the preset workshop virtual scene are stored in an XML configuration file. The lighting parameters include lighting color parameters (R_base, G_base, B_base) and lighting intensity parameters. The parameter parsing module reads the XML file, locates the lighting color parameter field, replaces its value with the target lighting color parameters (R_new, G_new, B_new), and saves the updated XML file, thus completing the update of the lighting color parameters.

[0071] Step S126: Establish a mapping relationship between global color temperature parameters and light intensity parameters, determine the corresponding target light intensity parameters based on the global color temperature parameters, update the light intensity parameters in the light parameters to the target light intensity parameters, and generate adapted light parameters.

[0072] The global color temperature parameter reflects the warmth or coolness of a video image and has a non-linear mapping relationship with the illumination intensity parameter. Experimental data shows that when the global color temperature parameter is low (the image is warmer), the illumination intensity needs to be reduced to avoid overexposure; when the global color temperature parameter is high (the image is cooler), the illumination intensity needs to be increased to enhance the image brightness. To establish the mapping relationship, a polynomial fitting method was used to fit the experimental data, obtaining a functional relationship between the global color temperature parameter and the illumination intensity parameter. Based on this functional relationship, the target illumination intensity parameter was calculated by inputting the global color temperature parameter. The target illumination color parameter and the target illumination intensity parameter were integrated to generate the adapted illumination parameters, which were then stored as structured data.

[0073] Step S127: If the global humidity parameter reaches the preset humidity trigger value, set the particle type parameter in the particle system parameters to raindrop particle type, determine the target particle quantity parameter according to the correspondence between the global humidity parameter and the particle quantity parameter, and update the particle quantity parameter to the target particle quantity parameter; if the global humidity parameter does not reach the preset humidity trigger value, keep the particle type parameter as no particle type and the particle quantity parameter as zero, and generate the adapted particle system parameters.

[0074] The preset humidity trigger value is set according to the workshop's moisture-proof standards. When the global humidity parameter exceeds this value, it indicates that the workshop environment has high humidity and there may be a risk of condensation. Raindrop particle effects need to be displayed in the virtual scene to alert operators. The global humidity parameter and particle quantity parameter are positively correlated; the higher the humidity, the more particles. The correlation is defined by a piecewise function: when the global humidity parameter is between the trigger value and trigger value +10%, the particle quantity parameter is N1; when it is between trigger value +10% and trigger value +20%, the particle quantity parameter is N2 (N2>N1), and so on. If the global humidity parameter does not reach the trigger value, the particle system does not start, the particle type parameter is "no particles," and the particle quantity parameter is zero. The generated adapted particle system parameters include particle type and particle quantity information.

[0075] Step S128: Establish the mapping relationship between global temperature parameters and model material parameters. Determine the corresponding target material reflection parameters and target material light transmission parameters based on the global temperature parameters. Update the material reflection parameters in the model material parameters to the target material reflection parameters and the material light transmission parameters to the target material light transmission parameters, and generate the adapted model material parameters.

[0076] The global temperature parameter reflects the operating heat of the equipment. The reflection and transmittance parameters in the model material parameters need to be adjusted with temperature changes to simulate the thermal radiation effect of the equipment. When the global temperature parameter increases, the surface temperature of the equipment rises, the material reflection parameter decreases (the surface gloss decreases), and the transmittance parameter increases (the translucent effect caused by thermal radiation). When establishing the mapping relationship, the reflectivity and transmittance data of the equipment material at different temperatures are obtained through thermal imaging experiments, and an interpolation algorithm is used to generate the mapping curve. Based on the global temperature parameter, the mapping curve is queried to obtain the target material reflection and transmittance parameters, which are then used to replace the original values ​​in the model material parameters to generate the adapted model material parameters.

[0077] Step S129: Based on the coordinate-labeled partition layout data in the multi-source data collaborative access dataset, determine the key observation points of each functional area, set the view position parameter in the view parameters to the coordinates of the key observation points, determine the view switching frequency parameter based on the number of key observation points, and generate the adapted view parameters.

[0078] The zone layout data with coordinate identifiers includes the spatial coordinates and equipment distribution information of each functional area. The principles for determining key observation points are: coverage of all critical equipment within the area, no visual obstruction, and ease of monitoring by operators. For example, key observation points in the machining area are set directly in front of the CNC machine tool's operating surface and to the side of the spindle box; key observation points in the switch station area are set directly above the circuit breaker and at the busbar connection. The three-dimensional coordinates of each key observation point are obtained using a spatial coordinate calculation tool and used as the viewpoint position parameter. The viewpoint switching frequency parameter is determined based on the number and importance of key observation points; the more points, the higher the switching frequency. Important areas (such as the main transformer area) have higher switching priority than other areas, and their switching frequency percentage can be appropriately increased. The generated adapted viewpoint parameters include a list of viewpoint position coordinates and switching frequency rules.

[0079] Step S1210: Integrate the adapted lighting parameters, adapted particle system parameters, adapted model material parameters, and adapted viewpoint parameters to generate a scene dynamic adaptation parameter set.

[0080] After the adapted lighting parameters, particle system parameters, model material parameters, and viewpoint parameters are generated, parameter validation is required to ensure that the values ​​of each parameter meet the requirements of the virtual reality rendering system. For example, the lighting intensity parameter must be within the range of 0-1000 candela, and the particle number parameter must be a non-negative integer. After successful validation, the four types of parameters are integrated according to the scene parameter classification standards to generate a scene dynamic adaptation parameter set. The parameter set is stored in JSON format, containing fields such as parameter name, parameter value, update timestamp, and effective area, facilitating subsequent reading and parsing by the rendering module.

[0081] Step S130: Call the layered rendering module of the virtual reality rendering system, combine the workshop computer-aided design layout data in the multi-source data collaborative access dataset with the scene dynamic adaptation parameter set, and perform layered virtual-real fusion rendering operation on the workshop real-time video stream and the preset workshop virtual scene to generate the initial workshop virtual scene rendering result.

[0082] The layered rendering module of the virtual reality rendering system supports region-based layered rendering, enabling the decomposition of complex scenes into multiple layers for parallel processing. In the virtual scene rendering of a hydropower equipment production workshop, based on the functional area divisions in the workshop's computer-aided design layout data, the virtual scene is divided into processing area layers, storage area layers, passageway area layers, and assembly and debugging area layers. Each layer's rendering requires calling the corresponding scene dynamic adaptation parameters (e.g., the processing area layer uses the adapted lighting parameters for the processing area), and integrating real-time video stream data from the workshop. The layered virtual-real fusion rendering operation includes steps such as video texture projection, adding lighting effects, generating particle effects, adjusting material properties, and image compositing. Multi-threaded parallel processing improves rendering efficiency, generating the initial workshop virtual scene rendering result.

[0083] Step S131: Extract the partition layout data with coordinate labels from the multi-source data collaborative access dataset, and input it into the layered rendering module of the virtual reality rendering system. The layered rendering module constructs the corresponding three-dimensional scene layered structure according to the division of processing area, storage area and channel area.

[0084] The partitioned layout data with coordinate identifiers contains the spatial coordinates of each functional area and equipment model information. After being input into the layered rendering module, the module first parses the area division information in the data and creates three sub-scene nodes: processing area layer, storage area layer, and passage area layer. Each sub-scene node is initialized according to the spatial coordinate range of the area, setting the layer's bounding box, rendering priority (processing area layer has the highest priority), and visibility attributes. The 3D scene layered structure adopts a tree organization, with the root node representing the entire workshop virtual scene, child nodes representing each functional area layer, and leaf nodes representing the equipment models within each area. After the layered structure is built, it is stored in the scene graph database, supporting dynamic loading and unloading.

[0085] Step S132: Extract the preloaded video stream from the multi-source data collaborative access dataset, perform texture processing on each video frame in the preloaded video stream, and generate video texture data adapted to the layered structure of the three-dimensional scene.

[0086] The preloaded video stream contains real-time video footage from multiple regions, requiring texture processing to adapt to the layered structure of the 3D scene. Texture processing includes steps such as video frame cropping, resolution adjustment, and format conversion. Based on the bounding box coordinates of each region layer in the 3D scene's layered structure, video frames are cropped, retaining the portion of the video footage corresponding to that region layer. Resolution adjustment is performed according to the rendering precision requirements of the region layer. For regions requiring high rendering precision, the video texture resolution is adjusted to the original resolution; for regions with lower precision requirements, the resolution can be reduced to decrease computation. Format conversion converts the RGB format of the video frames to the RGBA format supported by the rendering system, adding an alpha channel to achieve a transparent overlay effect. The processed video texture data is categorized and stored according to region layers, with each region layer corresponding to a set of video texture data.

[0087] Step S133: Call the projector management module of the virtual reality rendering system, and configure a corresponding projector object for each three-dimensional scene layer according to the adapted viewpoint parameters in the scene dynamic adaptation parameter set. The projector object is used to control the projection range and projection accuracy of video texture data in the three-dimensional scene layer.

[0088] The adapted viewing angle parameters include the coordinates of key observation points and the viewing angle switching frequency for each functional area. The projector management module determines the position and orientation of the projector object based on the key observation point coordinates, ensuring that the projector's projection direction is consistent with the viewing angle. Each 3D scene layer is configured with at least one projector object. Due to the dense equipment and complex details in the processing area layer, multiple projector objects are configured (e.g., one projector per CNC machine tool). The storage area layer and the passageway layer can each be configured with a single projector object. The projection range of the projector object is set based on the bounding box coordinates of the area layer, and the projection accuracy is determined by the viewing angle switching frequency in the adapted viewing angle parameters; areas with higher switching frequencies require higher projection accuracy. After creation, projector objects are stored in a projector object pool, supporting dynamic scheduling and reuse.

[0089] For example, the method may further include: step S1331: extracting the adapted view parameters from the scene dynamic adaptation parameter set, and obtaining the view position parameters and view switching frequency parameters in the adapted view parameters.

[0090] The scene dynamic adaptation parameter set is a JSON format file. After adaptation, the view parameters are stored in the "viewParameters" field, which includes "position" (view position parameter, 3D coordinates) and "switchFrequency" (view switching frequency parameter, times / minute). This field is read using a JSON parsing library to extract the view position parameters (X_view, Y_view, Z_view) and the view switching frequency parameter F_switch, and then stored in a memory variable.

[0091] Step S1332: Determine the observation view range of each 3D scene layer based on the view position parameters. The observation view range of the processing area covers all processing equipment, the observation view range of the storage area covers the entire shelf, and the observation view range of the aisle area covers the aisle entrance and exit.

[0092] The viewpoint position parameters (X_view, Y_view, Z_view) represent the coordinates of the observation point. The viewing angle range is determined by calculating the frustum between the observation point and the region boundary. The horizontal and vertical viewing angles of the frustum are set according to the resolution of the display device, such as a horizontal viewing angle of 90 degrees and a vertical viewing angle of 60 degrees. The viewing angle range of the processing area must cover all CNC machine tools, milling machines, grinding machines, and other processing equipment. By calculating the intersection relationship between the boundary sphere of the equipment model and the frustum, the size of the frustum is adjusted to ensure that all equipment is within the frustum. The storage area must cover the entire shelving, including the goods on the top and bottom shelves. The aisle area must cover entrances, exits, and intermediate turns to ensure no blind spots. The viewing angle range is represented by boundary coordinates (X_min, X_max, Y_min, Y_max, Z_min, Z_max).

[0093] Step S1333: Based on the viewing angle range of each 3D scene layer, calculate the projection angle and projection distance of the projector object. The projection angle ensures that the video texture data completely covers the 3D scene layer surface within the viewing angle range, and the projection distance is determined according to the spatial size of the 3D scene layer to avoid distortion of the projected image.

[0094] The projection angle includes the horizontal projection angle and the vertical projection angle, calculated as follows: Horizontal projection angle = 2 × arctan((X_max - X_min) / 2 / D), Vertical projection angle = 2 × arctan((Z_max - Z_min) / 2 / D), where D is the projection distance, (X_max - X_min) is the horizontal width of the viewing angle range, and (Z_max - Z_min) is the vertical height. The projection distance D is determined based on the spatial dimensions of the 3D scene layers (such as the length L of the processing area), typically taken as D = k × L, where k is a scaling factor (set empirically, such as k = 1.5) to ensure that the projected image scale matches the scene size and avoids distortion.

[0095] Step S1334: Based on the projection angle and projection distance, assign corresponding projector objects to each 3D scene layer. Due to the high density of equipment in the processing area, multiple projector objects are configured to meet the projection accuracy requirements. The storage area and the passage area are configured with one or more projector objects according to the size of the space.

[0096] The processing area is densely populated with equipment, and the projection accuracy of a single projector object is insufficient to meet the detailed requirements of all devices. Therefore, projector objects are divided according to equipment clusters, with one projector object configured for each cluster (e.g., one projector object for every three CNC machine tools). If the storage area has a large space (length exceeding a preset threshold), two projector objects are configured (one for projecting onto the left and one for projecting onto the right half of the shelving). The aisle area is long and narrow, and one projector object is sufficient to cover it. The number of projector objects is determined comprehensively based on the area, equipment density, and projection accuracy requirements.

[0097] Step S1335: Set projection range parameters for each projector object. The projection range parameters are determined based on the spatial coordinate information of the three-dimensional scene layers, and the boundary coordinates of the projection area are defined in pixels.

[0098] The spatial coordinate information of the 3D scene layering includes the bounding box coordinates of the region (X_min, X_max, Y_min, Y_max, Z_min, Z_max), which are converted into pixel coordinates in the screen coordinate system (U_min, U_max, V_min, V_max) using a coordinate transformation algorithm, where U is the horizontal pixel coordinate and V is the vertical pixel coordinate. The projection range parameter (U_min, U_max, V_min, V_max) defines the display area of ​​the video texture on the screen, ensuring that the texture projection does not exceed the boundaries of the 3D scene layering.

[0099] Step S1336: Set the projection accuracy parameters of the projector objects. Set the projection accuracy parameters of the projector objects in the processing area to a level that meets the requirements for detail rendering of the equipment in the processing area. Set the projection accuracy parameters of the projector objects in the storage area and the passage area to a level that meets the observation requirements of the corresponding areas according to the observation needs.

[0100] Projection accuracy parameters are graded from 1 to 5, with 1 being the lowest and 5 the highest. For processing areas requiring high detail rendering, projection accuracy parameters are set to 4-5, corresponding to texture sampling frequencies of 2048×1536-4096×3072. For warehouse areas, where the focus is on observing the storage status of goods, projection accuracy parameters are set to 2-3, with sampling frequencies of 1024×768-1536×1152. For aisle areas, where the focus is on observing the flow of personnel and materials, projection accuracy parameters are set to 2, with a sampling frequency of 1024×768. The correspondence between accuracy levels and sampling frequencies is stored in the projector configuration table.

[0101] Step S1337: Configure a control signal receiving unit for each projector object, the control signal receiving unit being used to receive adjustment signals for projection range and projection accuracy in subsequent steps.

[0102] The control signal receiving unit uses a Socket communication interface, supports the TCP / IP protocol, and listens on a preset port (e.g., port number 12345). The unit includes a signal parsing module, capable of parsing the parameter type (projection range or projection accuracy), adjustment value, and effective time information in the adjustment signal. After startup, the control signal receiving unit registers with the projector management module's device list, receives adjustment signals forwarded by the module, and updates the projection range and projection accuracy parameters of the projector object in real time.

[0103] Step S1338: Associate and store the configured projector object with the corresponding 3D scene layer to generate a projector object configuration table. The configuration table includes 3D scene layer identifier, projector object identifier, projection range parameters, projection accuracy parameters, and control signal receiving unit information.

[0104] The 3D scene layer identifiers are unique IDs for the processing area layer, storage area layer, and passage area layer; the projector object identifier is a unique number for each projector; the projection range parameters are (U_min, U_max, V_min, V_max); the projection accuracy parameters are numerical values ​​in a grade system; the control signal receiving unit information includes the IP address and port number. The projector object configuration table is stored using a relational database table structure, named "projector_config", and contains the fields "layer_id", "projector_id", "projection_range", "precision_level", "control_ip", and "control_port".

[0105] Step S1339: Input the projector object configuration table into the projector management module of the virtual reality rendering system. The projector management module initializes each projector object according to the configuration table and performs a functional test on each projector object. After the test is passed, the initialization is completed.

[0106] The projector management module reads the projector object configuration table, creates a projector object instance based on the "projector_id", sets the instance's projection range and projection accuracy parameters, and starts the control signal receiving unit (binding the IP address and port number). Functional tests include projection range testing (checking if the texture is displayed within the set range), projection accuracy testing (checking if the texture resolution meets the required level), and control signal receiving testing (sending test adjustment signals and checking if the parameters are updated). Projector objects that pass the tests are marked as "available," while those that fail are marked as "faulty" and error logs are recorded, completing the initialization process.

[0107] Step S134: Input the video texture data into the corresponding projector object. The projector object projects the video texture data onto the corresponding three-dimensional scene layer surface according to the spatial coordinate information of the three-dimensional scene layer, generating layered scene data after texture projection.

[0108] Video texture data is input to the corresponding projector object via the texture binding interface. The projector object calculates the transformation matrix (including translation, rotation, and scaling transformations) for texture projection based on the spatial coordinate information of the 3D scene layers. The calculation of the transformation matrix is ​​based on the camera calibration parameters and the positional relationship of the 3D scene layers, ensuring that the video texture can be accurately projected onto the corresponding position on the surface of the region layer (e.g., projecting the video texture of a CNC machine tool onto the surface of the machine tool in a 3D model). During projection, the projector object adjusts the UV coordinates of the texture in real time to resolve texture stretching or deformation issues. After texture projection is complete, layered scene data containing the video texture is generated and stored in video memory to accelerate subsequent rendering steps.

[0109] Step S135: Input the adapted lighting parameters from the scene dynamic adaptation parameter set into the lighting rendering module of the virtual reality rendering system. The lighting rendering module adds lighting effects to the layered scene data after texture projection based on the target lighting color parameters and target lighting intensity parameters in the adapted lighting parameters, generating layered scene data with lighting.

[0110] The adapted lighting parameters include the target lighting color and target lighting intensity. After being input into the lighting rendering module, the module first creates lighting objects such as parallel lights, point lights, and spotlights, and sets the color (target lighting color parameter), intensity (target lighting intensity parameter), position, and direction of the lighting objects. For the processing area layer, spotlights are added to highlight equipment details; for the storage area layer, parallel lights are added to simulate ambient light. During the addition of lighting effects, the Phong lighting model is used to calculate the diffuse reflection, specular reflection, and ambient light components of the object surface, combined with the color information of the video texture, to generate layered scene data with lighting. The lighting calculation supports real-time shadow effects, generating shadow textures through shadow mapping technology and overlaying them onto the layered scene data.

[0111] Step S136: Input the adapted particle system parameters from the scene dynamic adaptation parameter set into the particle rendering module of the virtual reality rendering system. The particle rendering module generates particle effects in the spatial region corresponding to the layered scene data with illumination based on the particle type parameters and particle quantity parameters in the adapted particle system parameters, thereby generating layered scene data with particle effects.

[0112] After adaptation, the particle type and quantity in the particle system parameters determine the shape and density of the particle effects. The particle rendering module creates a corresponding particle emitter based on the particle type parameter, such as a raindrop emitter for the "raindrop particle" type. The emitter's position (determined by the region coordinates in the partitioned layout data with coordinate identifiers), emission direction (vertically downwards), particle speed, and lifespan are set. The particle quantity parameter controls the emitter's emission rate; a higher quantity results in a higher emission rate. After the particle effects are generated, they are depth-tested and blended with layered scene data with lighting to ensure that the particles can be correctly occluded or occluded, generating layered scene data with particle effects.

[0113] Step S137: Input the adapted model material parameters from the scene dynamic adaptation parameter set into the material rendering module of the virtual reality rendering system. The material rendering module adjusts the surface material properties of the device 3D model in the layered scene data with particle effects according to the target material reflection parameters and target material light transmission parameters in the adapted model material parameters, and generates the layered scene data after material adjustment.

[0114] After adaptation, the model's material parameters include target reflection and target light transmission parameters. These are input into the material rendering module, which iterates through all 3D models of devices in the layered scene data with particle effects, reads the model's material attribute files, and updates the reflection and light transmission parameters to the target values. For devices made of metal (such as CNC machine tool spindles), the adjustment range for reflection parameters is larger; for devices made of non-metallic materials (such as plastic casings), the adjustment range is smaller. After the material attributes are adjusted, the lighting response of the model surface is recalculated to ensure that the material changes match the lighting effects, generating the layered scene data after material adjustment.

[0115] Step S138: Call the viewpoint management module of the virtual reality rendering system, control the switching of the observation viewpoint between the three-dimensional scene layers according to the adapted viewpoint parameters in the scene dynamic adaptation parameter set, and perform image compositing processing on the layered scene data after material adjustment under each viewpoint.

[0116] The adapted viewpoint parameters include the coordinates of key observation points and the viewpoint switching frequency. The viewpoint management module creates a virtual camera based on the coordinates of the key observation points and sets parameters such as the camera's position, orientation, and field of view. The viewpoint switching frequency parameter controls the switching interval between different observation points of the virtual camera, such as switching the viewpoint every 10 seconds. In the image compositing process, the adjusted material data of each 3D scene layer are superimposed according to rendering priority, with the processing area layer at the top, the storage area layer next, and the channel area layer at the bottom. During compositing, anti-aliasing algorithms (such as FXAA) are applied to reduce image jaggedness, and tone mapping algorithms are used to adjust image contrast and brightness to ensure consistent visual effects in the composite image.

[0117] Step S139: Perform anti-aliasing and resolution adaptation processing on the composited image, integrate the continuous image frames according to the preset frame rate, and generate the initial workshop virtual scene rendering result.

[0118] The composited image may have jagged edges, affecting visual quality, necessitating anti-aliasing. Anti-aliasing employs a multi-sampling algorithm, sampling each pixel multiple times and calculating the average of the sampled colors as the final pixel color. Resolution adaptation is performed based on the display device's resolution requirements, adjusting the composited image resolution to match the device's physical resolution to avoid stretching or compression distortion. The preset frame rate is determined based on the virtual reality rendering system's performance and the display device's refresh rate, typically set to 60 frames per second. Following the preset frame rate, consecutive anti-aliased and resolution-adapted frames are integrated into a video stream, stored in the rendering result buffer, and used to generate the initial workshop virtual scene rendering result.

[0119] Step S140: Collect real-time feedback information of the initial workshop virtual scene rendering result. The real-time feedback information includes image blending accuracy information and scene response speed information. Based on the real-time feedback information, perform iterative adjustment of the scene dynamic adaptation parameter set to generate an optimized scene dynamic adaptation parameter set.

[0120] After the initial virtual scene rendering of the workshop is generated, its quality needs to be evaluated through a real-time feedback mechanism. If problems such as insufficient image blending accuracy or slow response speed exist, the scene dynamic adaptation parameter set needs to be adjusted. Real-time feedback information is collected by a detection module deployed at the output of the rendering system. Image blending accuracy information is obtained by comparing the edge matching degree between video textures and 3D models, and scene response speed information is obtained by measuring the time interval between data input and image output. Based on the feedback information, the direction of parameter adjustment is determined (such as improving projection accuracy to improve blending accuracy, and reducing the number of particles to improve response speed). Through multiple iterations of adjustments, an optimized scene dynamic adaptation parameter set is generated.

[0121] Step S141: Deploy the feedback acquisition module of the virtual reality rendering system, wherein the feedback acquisition module includes an image sampling unit and a speed detection unit.

[0122] The feedback acquisition module is deployed at the end of the rendering pipeline in the virtual reality rendering system and is directly connected to the rendering result buffer. The image sampling unit uses frame capture technology to capture image frames from the rendering result buffer at a preset sampling frequency (e.g., 30 times / second) and store them in the sampling cache. The speed detection unit records the timestamps of each stage of data processing using a timer, including data input time, rendering start time, rendering end time, and image output time. The module also includes a data preprocessing unit to perform format conversion and noise filtering on the acquired image frames and timestamp data to ensure the accuracy of the feedback information.

[0123] Step S142: The image sampling unit performs continuous sampling on the initial workshop virtual scene rendering result to obtain multiple sampled images. It performs edge comparison between the video texture area and the three-dimensional scene area in each sampled image to generate edge matching degree data. The edge matching degree data is used to describe the image fusion accuracy and constitutes the image fusion accuracy information.

[0124] The image sampling unit continuously samples the initial rendering result for 10 seconds with a 100-millisecond interval, acquiring a total of 100 sampled frames. For each sampled frame, the video texture region and the 3D scene region are separated using an image segmentation algorithm, and the edge contours of the regions are extracted (using the Canny edge detection operator). Edge matching uses the Hausdorff distance algorithm to calculate the average distance between the video texture edge and the 3D scene edge; the smaller the distance, the higher the edge matching degree. The edge matching degree data ranges from 0 to 1, where 1 represents a perfect match and 0 represents a complete mismatch. The average edge matching degree data of the 100 sampled frames is used as a quantitative indicator of image fusion accuracy.

[0125] Step S143: The speed detection unit records the time interval from the input of the multi-source data collaborative access dataset to the output of the initial workshop virtual scene rendering result, and at the same time records the time interval of screen update when the viewpoint is switched. The time interval data constitutes the scene response speed information.

[0126] The speed detection unit records the initial timestamp T1 when the multi-source data collaborative access dataset is input into the data platform system; records the timestamp T2 when the virtual reality rendering system begins to execute layered rendering operations; and records the timestamp T3 when the initial workshop virtual scene rendering result is output to the display device. The time interval from data input to screen output is T3-T1. When switching perspectives, the unit records the timestamp T4 when the perspective switching command is issued and the timestamp T5 when the screen update is completed, with the screen update time interval being T5-T4. The time interval data is statistically analyzed according to the sampling period (e.g., 1 minute) to calculate the average time interval, maximum time interval, and minimum time interval, thus forming the scene response speed information.

[0127] Step S144: Analyze the edge matching degree data in the image fusion accuracy information. If the edge matching degree data is lower than the preset matching degree threshold, determine that the adapted viewing angle parameters and adapted projector control parameters in the scene dynamic adaptation parameter set need to be adjusted.

[0128] The preset matching threshold is set according to industry standards and user experience requirements, with a value of 0.8 (meaning the edge matching degree must reach above 80%). Trend analysis of the edge matching degree data is performed; if the average matching degree for five consecutive sampling periods is below 0.8 and the fluctuation range is less than 0.05, the image fusion accuracy is deemed insufficient. By locating low-matching degree areas, it was found that the problem is mainly concentrated at the edges of the CNC machine tools in the processing area and the video texture. Therefore, it was determined that the post-adaptation viewing angle parameters (optimizing the viewing angle to reduce occlusion) and the post-adaptation projector control parameters (improving projection accuracy to optimize edge alignment) need to be adjusted.

[0129] Step S145: Adjust the view position parameter in the adapted view parameters, fine-tune the view position parameter to the coordinate position where the edge matching degree data meets the preset requirements; adjust the projection accuracy parameter of the projector object, improve the projection accuracy of the video texture data on the layered surface of the three-dimensional scene to the preset standard, and generate the adjusted view parameters and the adjusted projector parameters.

[0130] The initial value of the viewpoint position parameter in the adapted viewpoint parameters is (X0, Y0, Z0). By traversing the surrounding coordinate points (X0±ΔX, Y0±ΔY, Z0±ΔZ), the edge matching degree data of each point is calculated, and the coordinate point with the highest matching degree is selected as the new viewpoint position parameter. ΔX, ΔY, and ΔZ are the fine-tuning step sizes, set according to the scene scale (e.g., 0.5 meters). The projection accuracy parameter of the projector object is achieved by adjusting the texture sampling frequency, initially at 1024×768, gradually increasing to 2048×1536, while enabling supersampling anti-aliasing technology to improve edge smoothness. The adjusted viewpoint parameters and projector parameters are stored as "Adjusted Viewpoint Parameters" and "Adjusted Projector Parameters," respectively.

[0131] Step S146: Analyze the time interval data in the scene response speed information. If the time interval data exceeds the preset speed threshold, determine that the adapted particle system parameters and adapted lighting parameters in the scene dynamic adaptation parameter set need to be adjusted.

[0132] The preset speed thresholds are set based on the display device's refresh rate and user interaction requirements. The time interval threshold from data input to screen output is 100 milliseconds, and the time interval threshold for screen updates during viewpoint switching is 50 milliseconds. Statistical analysis of the time interval data shows that if the average time interval exceeds the threshold and the 90th percentile time interval also exceeds the threshold, the scene response speed is deemed insufficient. Performance analysis tools revealed that particle rendering and lighting calculations consumed a significant amount of GPU resources. Therefore, it was determined that the adapted particle system parameters (reducing the number of particles to reduce computational load) and adapted lighting parameters (simplifying the lighting model to reduce computational complexity) needed adjustment.

[0133] Step S147: Reduce the particle number parameter in the adapted particle system parameters to a preset range to reduce the computational load of particle rendering; adjust the light intensity parameter in the adapted lighting parameters according to a preset adjustment ratio to reduce the complexity of lighting calculations, and generate the adjusted particle system parameters and adjusted lighting parameters.

[0134] The initial value of the particle number parameter in the adapted particle system parameters is N, with a preset range of [N_min, N_max], where N_min is the minimum number of particles to ensure visual quality. Based on the relationship curve between the time interval data and particle number in the scene response speed information, the particle number parameter is reduced from N to N' (N' is within the range of [N_min, N_max]), while particle effects in some non-critical areas (such as dust particles in the channel area) are turned off. The light intensity parameter in the adapted lighting parameters is adjusted according to a preset adjustment ratio (such as a 10% reduction), while reducing the number of point lights and spotlights, retaining the lighting effects in critical areas. The adjusted parameters are stored as the adjusted particle system parameters and the adjusted lighting parameters, respectively.

[0135] Step S148: Substitute the adjusted viewpoint parameters, adjusted projector parameters, adjusted particle system parameters, and adjusted lighting parameters into the original scene dynamic adaptation parameter set, and replace the corresponding original parameters.

[0136] The original scene dynamic adaptation parameter set is a JSON file containing fields such as adapted lighting parameters, adapted particle system parameters, adapted model material parameters, and adapted viewpoint parameters. Replace the viewpoint position parameter in the original adapted viewpoint parameters with the adjusted viewpoint parameters; add the adjusted projector parameters to the adapted projector control parameter field (this field is not present in the original parameter set and needs to be added); replace the particle quantity parameter in the original adapted particle system parameters with the adjusted particle system parameters; and replace the light intensity parameter in the original adapted lighting parameters with the adjusted lighting parameters. During the replacement process, ensure that the parameter format is consistent with the original parameter set, and add field descriptions and default values ​​for newly added fields.

[0137] Step S149: Perform a rendering pre-test on the replaced parameter set, and collect the image fusion accuracy information and scene response speed information during the pre-test process. If the image fusion accuracy information of the pre-test meets the preset matching degree threshold and the scene response speed information meets the preset speed threshold, then the replaced parameter set is determined as the optimized scene dynamic adaptation parameter set.

[0138] The rendering pre-test was conducted in a separate test environment using the same hardware and software configuration as the actual rendering system. The pre-test lasted 5 minutes, simulating real-world data input (such as real-time video streams and IoT data) and user interactions (such as viewpoint switching and zooming). Image blending accuracy and scene response speed information were collected during the pre-test. If the average edge matching score was greater than a preset matching score threshold (0.8) and the average time interval was less than a preset speed threshold (100 milliseconds), the parameter adjustment was deemed effective, and the replaced parameter set became the optimized scene dynamic adaptation parameter set.

[0139] Step S1410: If the pre-test results do not meet the threshold requirements, repeat the parameter adjustment and pre-test steps until the threshold requirements are met, and finally generate the optimized scene dynamic adaptation parameter set.

[0140] If the image blending accuracy or scene response speed information in the pre-test results does not meet the threshold requirements, the reasons for the failure need to be analyzed, and the direction and magnitude of parameter adjustments need to be redefined. For example, if the blending accuracy is still insufficient, the projector projection accuracy needs to be further improved or the viewing angle position fine-tuned; if the response speed is still too slow, the number of particles needs to be further reduced or the lighting calculations simplified. After each parameter adjustment, the rendering pre-test is re-executed, and the test results are recorded. Through multiple iterations (usually 3-5 times), until the pre-test results meet the threshold requirements, the optimized scene dynamic adaptation parameter set is finally generated.

[0141] Step S150: Substitute the optimized scene dynamic adaptation parameter set into the layered virtual-real fusion rendering operation to generate the final workshop virtual scene rendering result, and based on the final workshop virtual scene rendering result and workshop equipment status data, drive the workshop equipment 3D model to perform state synchronization update through a finite state machine.

[0142] After the optimized scene dynamic adaptation parameter set is generated, it is re-substituted into the layered virtual-real fusion rendering operation. At this point, the rendering result's image fusion accuracy and scene response speed both meet the requirements, generating the final workshop virtual scene rendering result. To ensure that the equipment model state in the virtual scene is consistent with the actual equipment, it is necessary to drive the equipment 3D model to update its state based on workshop equipment state data (such as operating status and working parameters). Finite state machine, as the core mechanism of state-driven operation, defines multiple states of the equipment model (such as running, stopped, and faulty) and state transition conditions, triggering corresponding state transitions based on equipment state data to achieve synchronous update of the equipment 3D model's state.

[0143] Step S151: Input the optimized lighting parameters, optimized particle system parameters, optimized model material parameters, and optimized viewpoint parameters from the optimized scene dynamic adaptation parameter set into the corresponding rendering modules of the virtual reality rendering system.

[0144] The optimized scene dynamic adaptation parameter set contains various optimized parameters, which are input to the corresponding rendering modules through the parameter distribution interface: optimized lighting parameters are input to the lighting rendering module, optimized particle system parameters are input to the particle rendering module, optimized model material parameters are input to the material rendering module, and optimized viewpoint parameters are input to the viewpoint management module. During the parameter input process, each rendering module performs validity checks on the parameters (such as value range checks and format checks). After the checks pass, the module updates its internal parameter cache and immediately applies it to subsequent rendering operations.

[0145] Step S152: Following the same layered rendering process as generating the initial workshop virtual scene rendering result, call the layered rendering module, projector management module, lighting rendering module, particle rendering module, material rendering module and view management module to perform layered virtual and real fusion rendering operation on the preloaded video stream and the partitioned layout data with coordinate labels in the multi-source data collaborative access dataset, and generate the final workshop virtual scene rendering result.

[0146] The layered rendering process is consistent with the initial rendering, including steps such as constructing the 3D scene layer structure, video texture processing, projector configuration, texture projection, adding lighting, generating particle effects, adjusting materials, switching perspectives, and compositing the image. The difference lies in that each rendering module uses optimized parameters; for example, the lighting rendering module uses optimized lighting parameters, and the particle rendering module uses optimized particle system parameters. During the rendering process, the real-time feedback acquisition module continues to work, triggering emergency adjustments if parameter anomalies are detected. The final generated virtual workshop scene rendering result is stored in the final rendering result buffer and simultaneously output to the display device for operators to view.

[0147] Step S153: Extract the current rendering status information of the equipment 3D model from the final workshop virtual scene rendering result. The current rendering status information includes the position information, posture information and surface status information of the equipment 3D model.

[0148] The final virtual scene rendering result of the workshop contains complete 3D scene data. The current rendering state information of the equipment's 3D model is extracted through a scene query interface. Position information includes the model's coordinates (X, Y, Z) in the world coordinate system; pose information includes the model's rotation angles (rotation angles around the X, Y, and Z axes); surface state information includes material color, texture ID, and luminous intensity. During extraction, the corresponding model node is searched from the scene graph database based on the device ID. The node's transformation matrix and material properties are read, and the current rendering state information is parsed and stored as structured data.

[0149] Step S154: Extract time-stamped device status data from the multi-source data collaborative access dataset. The time-stamped device status data includes the actual location information, actual posture information, and actual working status information of the device.

[0150] Timestamped device status data is stored in a relational database. Queries are performed using the device ID and a timestamp range to retrieve the latest device status data. Actual location information includes real-time coordinates for mobile devices and installation location coordinates for fixed devices. Actual posture information includes the movement angles of device components (such as valve opening degree or robotic arm joint angles). Actual operating status information includes status codes such as "Running," "Stopped," "Fault," and "Maintenance," along with corresponding fault codes and maintenance records. The retrieved data is categorized by device ID and stored in the device status cache.

[0151] Step S155: Construct a device state mapping relationship, associate the current rendering state information of the device 3D model with the actual state information of the device, and determine the device 3D model state parameters that need to be adjusted.

[0152] The equipment status mapping is achieved by establishing a status parameter lookup table. This table contains the correspondence between the status parameters of the equipment's 3D model (such as position, orientation, and surface color) and the actual status information of the equipment (such as actual position, actual orientation, and operating status code). For example, when the actual operating status information is "fault," the corresponding surface status parameter is "flashing red"; when the valve opening in the actual orientation information is 60%, the corresponding model orientation parameter is a 60-degree rotation around the Y-axis. By comparing the current rendered status information with the actual status information, inconsistent parameter items are identified, and the equipment's 3D model status parameters that need adjustment are determined (such as position parameters with position deviations exceeding a threshold, orientation parameters with inconsistent orientations, and surface status parameters with mismatched status codes).

[0153] Step S156: Deploy a finite state machine module, which includes a device state definition unit, a state transition condition setting unit, and a state driving unit. The device state definition unit defines multiple states of the device's 3D model based on the actual working state information of the device, including running state, stopped state, and maintenance state. The state transition condition setting unit sets transition conditions between different states based on changes in timestamped device state data. When the actual state information of the device meets the transition conditions, the state transition of the device's 3D model is triggered. The state driving unit generates a state adjustment instruction for the device's 3D model by combining the state and transition conditions defined by the finite state machine module with the device's 3D model state parameters that need to be adjusted as required.

[0154] For example, step S1561: Extract the current rendering position, current rendering pose and current surface state of the device from the state parameters of the device's 3D model that need to be adjusted.

[0155] The state parameters of the device's 3D model that need adjustment are stored in structured data, including fields such as device ID, state parameter type, current value, and target value. The current rendering position (X_render, Y_render, Z_render), current rendering pose (Rx_render, Ry_render, Rz_render), and current surface state (color_render, texture_render, glow_render) are extracted using the SQL query "SELECT position, pose, surface_state FROM device_adjust_params WHERE device_id='XXX'". Here, color_render represents the RGB color value, texture_render is the texture ID, and glow_render is the luminance intensity. The extracted data is stored in the state adjustment buffer.

[0156] Step S1562: Determine the current rendering state of the device's 3D model from the states defined by the finite state machine module. If the current rendering position deviates from the actual position by a preset position deviation value, or the current rendering posture deviates from the actual posture by a preset posture deviation value, or the current surface state does not match the actual working state, then it is determined that a state transition is required.

[0157] The states defined by the finite state machine module are stored in a state enumeration table, which includes four states: "Running," "Stopped," "Fault," and "Maintenance." By comparing the current surface state with the state characteristics in the state enumeration table (such as the "Fault" state characteristic being a flashing red color), the current rendering state S_current of the device's 3D model is determined. Calculate the Euclidean distance between the current rendering position and the actual position: D_pos = sqrt((X_render - X_actual)^2 + (Y_render - Y_actual)^2 + (Z_render - Z_actual)^2). If D_pos > the preset position deviation value (e.g., 0.1 meters), the position deviation exceeds the limit. Calculate the angle difference between the current rendering posture and the actual posture: ΔRx = |Rx_render - Rx_actual|, ΔRy = |Ry_render - Ry_actual|, ΔRz = |Rz_render - Rz_actual|. If any angle difference > the preset posture deviation value (e.g., 1 degree), the posture deviation exceeds the limit. If the color, texture, or luminous intensity of the current surface state does not match the standard state characteristics corresponding to the actual working state (e.g., the actual working state is "fault" but the current surface state is green), the surface state does not match. Meeting any of these conditions determines that a state transition is required.

[0158] Step S1563: Locate the state transition conditions set in the finite state machine module, and determine the target state to which the current rendering state needs to transition. If the actual working state of the device is running and the current rendering state is stopped, then the target state is running; if the actual working state of the device is maintenance and the current rendering state is running, then the target state is maintenance.

[0159] The state transition conditions of the finite state machine module are stored in a transition rule base, which contains three fields: current state, trigger condition, and target state. For example, the rule "Current state = Stopped, Trigger condition = Actual working state = Running, Target state = Running"; and the rule "Current state = Running, Trigger condition = Actual working state = Maintenance, Target state = Maintenance". The transition rule base is queried using the current rendering state S_current and the actual device working state S_actual to match the corresponding trigger condition and determine the target state S_target.

[0160] Step S1564: Based on the target state, determine the specific parameters that need to be adjusted in the 3D model of the equipment. If the target state is the running state, the parameters that need to be adjusted include the position parameters, attitude parameters, and surface state parameters. The position parameters are adjusted to the actual position of the equipment, the attitude parameters are adjusted to the running attitude of the equipment, and the surface state parameters are set to the color corresponding to the running state of the equipment. If the target state is the maintenance state, the parameters that need to be adjusted include the attitude parameters and surface state parameters. The attitude parameters are adjusted to the maintenance attitude of the equipment, and the surface state parameters are set to the color corresponding to the maintenance state of the equipment.

[0161] The correspondence between target states and the parameters that need adjustment is stored in a state-parameter lookup table. When the target state is "Running," the parameters in the lookup table are "Position, Attitude, Surface Color," with corresponding values ​​for the actual device position (X_actual, Y_actual, Z_actual), device running attitude (Rx_run, Ry_run, Rz_run), and running status color (R_run, G_run, B_run). When the target state is "Maintenance," the parameters are "Attitude, Surface Color," with corresponding values ​​for the device maintenance attitude (Rx_maintain, Ry_maintain, Rz_maintain) and maintenance status color (R_maintain, G_maintain, B_maintain). The specific parameters that need adjustment and their target values ​​are determined by querying the lookup table based on the target state S_target.

[0162] Step S1565: Calculate the adjustment amount for each parameter that needs to be adjusted. The adjustment amount for the position parameter is the difference between the actual position of the device and the current rendering position. The adjustment amount for the pose parameter is the difference between the actual pose of the device and the current rendering pose. The adjustment amount for the surface state parameter is the difference between the target surface state and the current surface state.

[0163] The position parameter adjustment amounts are ΔX = X_actual - X_render, ΔY = Y_actual - Y_render, ΔZ = Z_actual - Z_render; the attitude parameter adjustment amounts are ΔRx = Rx_target - Rx_render, ΔRy = Ry_target - Ry_render, ΔRz = Rz_target - Rz_render; and the surface state parameter adjustment amounts are ΔR = R_target - R_render, ΔG = G_target - G_render, ΔB = B_target - B_render. The units of the adjustment amounts are consistent with the parameter items (position in meters, attitude in degrees, color from 0-255).

[0164] Step S1566: Generate the parameter part of the status adjustment instruction based on the adjustment amount. The parameter part includes parameter item identifier, adjustment amount, and adjustment order.

[0165] The parameter part of the state adjustment instruction adopts a key-value pair format, with parameter items identified as "position", "attitude", and "surface color". The adjustment amount is ΔX, ΔY, ΔZ, etc. calculated in step S1565. The adjustment order is the execution order of the parameter items (e.g., adjust position first, then attitude, and finally surface color). For example, the parameter part is "{\"params\":[{\"id\":\"position\",\"delta\":[ΔX, ΔY, ΔZ]},{\"id\":\"attitude\",\"delta\":[ΔRx, ΔRy, ΔRz]},{\"id\":\"surface_color\",\"delta\":[ΔR, ΔG, ΔB]}],\"order\":[\"position\",\"attitude\",\"surface_color\"]}.

[0166] Step S1567: Add an instruction identifier and a device identifier to the status adjustment instruction. The instruction identifier is used to distinguish different types of adjustment instructions, and the device identifier is used to specify the 3D model of the device receiving the adjustment instruction.

[0167] The command identifier is "state_adjust" to distinguish it from other types of commands (such as viewpoint switching commands); the device identifier is the unique ID of the device's 3D model (such as "device_001"). The format of the added state adjustment command is "{\"cmd_id\":\"state_adjust\",\"device_id\":\"device_001\",\"params\":[...],\"order\":[...]}".

[0168] Step S1568: Integrate the instruction identifier, device identifier, and parameter parts to generate a complete device 3D model state adjustment instruction. Perform format verification on the generated state adjustment instruction to check whether the instruction format meets the receiving requirements of the device model driver module of the virtual reality rendering system. If it does not meet the requirements, readjust the instruction format until the verification passes.

[0169] The complete equipment 3D model status adjustment instructions are in JSON format strings. These are validated using a JSON format validation tool to check for syntax errors (such as mismatched brackets or missing fields) and data type errors (such as adjusting values ​​that should be numbers but are actually strings). If validation fails, the instruction format is corrected based on the error message (such as adding missing fields or converting data types), and validation is repeated until it passes. Successfully validated status adjustment instructions are stored in the instruction buffer, awaiting transmission.

[0170] Step S157: Input the state adjustment command into the device model driving module of the virtual reality rendering system. The device model driving module updates the position information, posture information and surface state information of the device 3D model according to the state adjustment command, so as to realize the synchronous update of the state of the device 3D model and the actual device.

[0171] The state adjustment command is a JSON-formatted control message sent to the device model driver module of the virtual reality rendering system via the communication module. The module parses the device ID, state parameter adjustment values, and adjustment methods from the command, locating the corresponding 3D device model node from the scene graph database. Updates to position and pose information are achieved by modifying the transformation matrix of the model node, supporting both smooth transitions (position / pose changes completed within a specified time using interpolation algorithms) and immediate switching. Updates to surface state information are achieved by modifying the model's material properties (such as color, texture, and luminous intensity) and animation controllers (such as starting / stopping rotation animations and triggering blinking effects). After the update is complete, the device model driver module returns a state update confirmation message, and the finite state machine module records the state transition history, completing the state synchronization update between the device 3D model and the actual device.

[0172] In one exemplary embodiment, a virtual reality-based workshop virtual scene rendering system is provided. This system can be a terminal, server, etc., and its internal structure diagram can be as follows: Figure 2 As shown, this virtual reality-based workshop virtual scene rendering system includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, near-field communication, or other technologies. When the computer program is executed by the processor, it implements a virtual reality-based workshop virtual scene rendering method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, or a button, trackball, or touchpad set on the shell of a virtual reality-based workshop virtual scene rendering system, or an external keyboard, touchpad, or mouse, etc.

[0173] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for rendering a virtual workshop scene based on virtual reality, characterized in that, The method includes: The workshop multi-source data collaborative access processing is performed. The dependency structure between multi-source data is established through the data middleware system, and a multi-source data collaborative access dataset is generated. The workshop multi-source data includes workshop real-time video stream, workshop IoT operation data, workshop computer-aided design layout data, and workshop equipment status data. Based on the real-time video stream characteristics of the workshop and the workshop IoT operation data in the multi-source data collaborative access dataset, intelligent adaptation processing is performed on the scene parameters of the preset workshop virtual scene to generate a dynamic scene adaptation parameter set. The layered rendering module of the virtual reality rendering system is invoked. The workshop computer-aided design layout data in the multi-source data collaborative access dataset and the scene dynamic adaptation parameter set are combined to perform a layered virtual-real fusion rendering operation on the real-time video stream of the workshop and the preset workshop virtual scene to generate the initial workshop virtual scene rendering result. Real-time feedback information of the initial workshop virtual scene rendering result is collected. The real-time feedback information includes image fusion accuracy information and scene response speed information. Based on the real-time feedback information, the scene dynamic adaptation parameter set is iteratively adjusted to generate an optimized scene dynamic adaptation parameter set. The optimized scene dynamic adaptation parameter set is substituted into the layered virtual-real fusion rendering operation to generate the final workshop virtual scene rendering result. Based on the final workshop virtual scene rendering result and workshop equipment status data, the workshop equipment 3D model is driven by a finite state machine to perform synchronous status updates.

2. The workshop virtual scene rendering method based on virtual reality according to claim 1, characterized in that, The multi-source data collaborative access processing in the execution workshop establishes a dependency structure between multi-source data through a data platform system, generating a multi-source data collaborative access dataset, including: The system calls a video access interface to receive raw video streams transmitted from multiple camera devices in the workshop. The video access interface supports real-time streaming protocols and proprietary protocols of multiple mainstream vendors' software development kits. It performs protocol conversion processing on raw video streams with different protocols to generate a unified protocol video stream. Preloading processing is performed on the unified protocol video stream. The number of preloaded video frames is determined based on historical rendering requirement data. The corresponding number of video frames are read in advance and stored in the cache area to generate a preloaded video stream. Deploy an IoT data receiving gateway to receive raw IoT data transmitted from temperature sensors, vibration sensors, and humidity sensors in the workshop. Perform data cleaning processing on the raw IoT data to remove duplicate and invalid data and generate cleaned IoT data. Obtain the original data of the computer-aided design layout of the workshop, divide the original data of the computer-aided design layout of the workshop into processing area layout data, storage area layout data and passage area layout data according to the functional areas of the workshop, add spatial coordinate information to each area layout data, and generate partition layout data with coordinate labels. The system receives workshop equipment status data transmitted from the workshop equipment controller. The workshop equipment status data includes equipment operating status, equipment location information, and equipment operating parameters. The system performs timestamp marking processing on the workshop equipment status data to generate timestamped equipment status data. The preloaded video stream, cleaned IoT data, partition layout data with coordinate identifiers, and device status data with timestamps are input into the data platform system. By using the dependency building module of the data middle platform system, the relationships between various data types are analyzed, and a dependency structure including data dependency direction and update triggering conditions is established. Start the data subscription module of the data platform system, configure subscription nodes for each data type, and when a certain type of data is updated, the subscription node will automatically receive the update notification and synchronize the update of the related data. The request forwarding module of the data platform system receives data requests for subsequent rendering steps, retrieves the corresponding data source according to the dependency structure, obtains the data, and returns it to the request node. All data, after being processed through dependency association and subscription synchronization, are integrated to generate a multi-source data collaborative access dataset.

3. The workshop virtual scene rendering method based on virtual reality according to claim 1, characterized in that, The process involves using the real-time video stream features of the workshop and the workshop IoT operation data from the multi-source data collaborative access dataset to perform intelligent adaptation processing on the scene parameters of a preset workshop virtual scene, generating a dynamic scene adaptation parameter set, including: The preloaded video stream is extracted from the multi-source data collaborative access dataset, and a video frame sequence is formed by selecting multiple consecutive video frames. Color feature extraction is performed on each video frame to obtain the hue and color temperature features of each video frame. Calculate the average hue feature of all video frames in the video frame sequence to obtain the global hue parameter; calculate the average color temperature feature of all video frames to obtain the global color temperature parameter; The cleaned IoT data is extracted from the multi-source data collaborative access dataset. The workshop environmental humidity data and equipment operating temperature data are filtered out. The average value of the workshop environmental humidity data is calculated to obtain the global humidity parameter. The average value of the equipment operating temperature data is calculated to obtain the global temperature parameter. Obtain the basic parameters of the preset workshop virtual scene. The basic parameters of the scene include lighting parameters, particle system parameters, model material parameters and view parameters. The lighting parameters include lighting color parameters and lighting intensity parameters. The particle system parameters include particle type parameters and particle quantity parameters. The model material parameters include material reflection parameters and material light transmission parameters. The view parameters include view position parameters and view switching frequency parameters. Establish a mapping relationship between global hue parameters and lighting color parameters, determine the corresponding target lighting color parameters based on the global hue parameters, and update the lighting color parameters in the lighting parameters to the target lighting color parameters; Establish a mapping relationship between global color temperature parameters and light intensity parameters, determine the corresponding target light intensity parameters based on the global color temperature parameters, update the light intensity parameters in the light parameters to the target light intensity parameters, and generate adapted light parameters. If the global humidity parameter reaches the preset humidity trigger value, the particle type parameter in the particle system parameter is set to raindrop particle type. The target particle number parameter is determined according to the correspondence between the global humidity parameter and the particle number parameter, and the particle number parameter is updated to the target particle number parameter. If the global humidity parameter does not reach the preset humidity trigger value, the particle type parameter is kept as no particle type, the particle number parameter is zero, and the adapted particle system parameter is generated. Establish a mapping relationship between global temperature parameters and model material parameters. Determine the corresponding target material reflection parameters and target material light transmission parameters based on the global temperature parameters. Update the material reflection parameters in the model material parameters to the target material reflection parameters and the material light transmission parameters to the target material light transmission parameters to generate adapted model material parameters. Based on the coordinate-labeled partition layout data in the multi-source data collaborative access dataset, the key observation points of each functional area are determined, the view position parameter in the view parameters is set as the coordinates of the key observation points, the view switching frequency parameter is determined according to the number of key observation points, and the adapted view parameters are generated. The adapted lighting parameters, adapted particle system parameters, adapted model material parameters, and adapted viewpoint parameters are integrated to generate a scene dynamic adaptation parameter set.

4. The workshop virtual scene rendering method based on virtual reality according to claim 1, characterized in that, The layered rendering module of the virtual reality rendering system, combining the workshop computer-aided design layout data in the multi-source data collaborative access dataset with the scene dynamic adaptation parameter set, performs a layered virtual-real fusion rendering operation on the real-time workshop video stream and the preset workshop virtual scene to generate an initial workshop virtual scene rendering result, including: The coordinate-labeled partition layout data is extracted from the multi-source data collaborative access dataset and input into the layered rendering module of the virtual reality rendering system. The layered rendering module constructs the corresponding three-dimensional scene layered structure according to the division of processing area, storage area and channel area. Extract a preloaded video stream from the multi-source data collaborative access dataset, perform texture processing on each video frame of the preloaded video stream, and generate video texture data that adapts to the layered structure of the three-dimensional scene. The projector management module of the virtual reality rendering system is invoked. Based on the adapted viewpoint parameters in the scene dynamic adaptation parameter set, a corresponding projector object is configured for each 3D scene layer. The projector object is used to control the projection range and projection accuracy of video texture data in the 3D scene layer. The video texture data is input into the corresponding projector object. The projector object projects the video texture data onto the corresponding three-dimensional scene layer surface according to the spatial coordinate information of the three-dimensional scene layer, generating layered scene data after texture projection. The adapted lighting parameters from the scene dynamic adaptation parameter set are input into the lighting rendering module of the virtual reality rendering system. The lighting rendering module adds lighting effects to the layered scene data after texture projection based on the target lighting color parameters and target lighting intensity parameters in the adapted lighting parameters, generating layered scene data with lighting. The adapted particle system parameters from the scene dynamic adaptation parameter set are input into the particle rendering module of the virtual reality rendering system. The particle rendering module generates particle effects in the spatial region corresponding to the layered scene data with illumination based on the particle type parameters and particle quantity parameters in the adapted particle system parameters, thus generating layered scene data with particle effects. The adapted model material parameters from the scene dynamic adaptation parameter set are input into the material rendering module of the virtual reality rendering system. The material rendering module adjusts the surface material properties of the device 3D model in the layered scene data with particle effects based on the target material reflection parameters and target material light transmission parameters in the adapted model material parameters, and generates layered scene data with adjusted materials. The viewpoint management module of the virtual reality rendering system is invoked to control the switching of the observation viewpoint between the three-dimensional scene layers according to the adapted viewpoint parameters in the scene dynamic adaptation parameter set. At the same time, the layered scene data with material adjustment under each viewpoint is processed for image compositing. Anti-aliasing and resolution adaptation are performed on the composited image, and consecutive frames are integrated according to the preset frame rate to generate the initial workshop virtual scene rendering result.

5. The workshop virtual scene rendering method based on virtual reality according to claim 1, characterized in that, The process involves collecting real-time feedback information on the initial virtual scene rendering results of the workshop. This real-time feedback information includes image blending accuracy information and scene response speed information. Based on this real-time feedback information, iterative adjustments are made to the scene dynamic adaptation parameter set to generate an optimized scene dynamic adaptation parameter set, including: A feedback acquisition module for deploying a virtual reality rendering system, the feedback acquisition module including an image sampling unit and a speed detection unit; The image sampling unit performs continuous sampling on the initial workshop virtual scene rendering result to obtain multiple sampled images. It performs edge comparison between the video texture area and the three-dimensional scene area in each sampled image to generate edge matching degree data. The edge matching degree data is used to describe the image fusion accuracy and constitutes the image fusion accuracy information. The speed detection unit records the time interval from the input of the multi-source data collaborative access dataset to the output of the initial workshop virtual scene rendering result, and also records the time interval of screen update when the viewpoint is switched. The time interval data constitutes the scene response speed information. Analyze the edge matching degree data in the image fusion accuracy information. If the edge matching degree data is lower than the preset matching degree threshold, it is determined that the adapted viewing angle parameters and the adapted projector control parameters in the scene dynamic adaptation parameter set need to be adjusted. Adjust the view position parameter in the adapted view parameters to fine-tune the view position parameter to the coordinate position where the edge matching degree data meets the preset requirements; adjust the projection accuracy parameter of the projector object to improve the projection accuracy of the video texture data on the layered surface of the three-dimensional scene to the preset standard, and generate the adjusted view parameters and the adjusted projector parameters. If the time interval data in the scene response speed information is analyzed, and the time interval data exceeds the preset speed threshold, it is determined that the adapted particle system parameters and adapted lighting parameters in the scene dynamic adaptation parameter set need to be adjusted. Reduce the particle number parameter in the adapted particle system parameters to a preset range to reduce the computational load of particle rendering; adjust the light intensity parameter in the adapted lighting parameters according to a preset adjustment ratio to reduce the complexity of lighting calculations, and generate the adjusted particle system parameters and adjusted lighting parameters. Substitute the adjusted viewpoint parameters, adjusted projector parameters, adjusted particle system parameters, and adjusted lighting parameters into the original scene dynamic adaptation parameter set, and replace the corresponding original parameters. Rendering pre-test is performed on the replaced parameter set, and the image fusion accuracy information and scene response speed information are collected during the pre-test process. If the image fusion accuracy information of the pre-test meets the preset matching degree threshold and the scene response speed information meets the preset speed threshold, then the replaced parameter set is determined as the optimized scene dynamic adaptation parameter set. If the pre-test results do not meet the threshold requirements, repeat the parameter adjustment and pre-test steps until the threshold requirements are met, and finally generate the optimized scene dynamic adaptation parameter set.

6. The workshop virtual scene rendering method based on virtual reality according to claim 1, characterized in that, The step of substituting the optimized scene dynamic adaptation parameter set into the layered virtual-real fusion rendering operation to generate the final workshop virtual scene rendering result, and based on the final workshop virtual scene rendering result and workshop equipment status data, driving the workshop equipment 3D model to perform state synchronization update through a finite state machine, includes: The optimized lighting parameters, optimized particle system parameters, optimized model material parameters, and optimized viewpoint parameters from the optimized scene dynamic adaptation parameter set are respectively input into the corresponding rendering modules of the virtual reality rendering system. Following the same layered rendering process as the initial workshop virtual scene rendering result, the layered rendering module, projector management module, lighting rendering module, particle rendering module, material rendering module and view management module are called to perform layered virtual and real fusion rendering operation on the preloaded video stream and the partitioned layout data with coordinate labels in the multi-source data collaborative access dataset, and generate the final workshop virtual scene rendering result. Extract the current rendering state information of the equipment 3D model from the final workshop virtual scene rendering result. The current rendering state information includes the position information, posture information and surface state information of the equipment 3D model. Extract time-stamped device status data from the multi-source data collaborative access dataset. The time-stamped device status data includes the actual location information, actual posture information and actual working status information of the device. Construct a device state mapping relationship, associate the current rendering state information of the device's 3D model with the actual state information of the device, and determine the device 3D model state parameters that need to be adjusted; A finite state machine module is deployed, comprising a device state definition unit, a state transition condition setting unit, and a state driving unit. The device state definition unit defines multiple states of the device's 3D model based on the device's actual operating state information, including running state, stopped state, and maintenance state. The state transition condition setting unit sets transition conditions between different states based on changes in timestamped device state data. When the actual device state information meets the transition conditions, a state transition of the device's 3D model is triggered. The state driving unit generates state adjustment instructions for the device's 3D model based on the device's 3D model state parameters that need adjustment, combined with the states and transition conditions defined by the finite state machine module. The state adjustment command is input into the device model driving module of the virtual reality rendering system. The device model driving module updates the position information, posture information and surface state information of the device's 3D model according to the state adjustment command, so as to realize the synchronous update of the state of the device's 3D model and the actual device.

7. The workshop virtual scene rendering method based on virtual reality according to claim 2, characterized in that, The dependency relationship building module of the data platform system analyzes the relationships between various data types and establishes a dependency relationship structure that includes data dependency directions and update triggering conditions, including: The dependency construction module of the data platform system performs data type analysis on preloaded video streams, cleaned IoT data, partition layout data with coordinate identifiers, and device status data with timestamps to determine the attributes and uses of each data type. Analyzing the relationship between the preloaded video stream and the partition layout data with coordinate identifiers, the projection of the preloaded video stream depends on the spatial coordinate information of the partition layout data with coordinate identifiers. Therefore, the data dependency direction is determined to be that the partition layout data with coordinate identifiers points to the preloaded video stream. The relationship between the cleaned IoT data and the scene dynamic adaptation parameter set was analyzed. The generation of the scene dynamic adaptation parameter set depends on the humidity and temperature information in the cleaned IoT data. The data dependency direction is determined to be that the cleaned IoT data points to the scene dynamic adaptation parameter set. The relationship between timestamped device status data and device 3D model status is analyzed. The update of device 3D model status depends on timestamped device status data. The data dependency direction is determined to be that timestamped device status data points to device 3D model status. Set update trigger conditions for each data dependency. When the spatial coordinates of the partition layout data with coordinate identifiers are adjusted, the texture projection range of the preloaded video stream is updated. When the change in humidity or temperature data in the IoT data after cleaning reaches the preset change threshold, the scene dynamic adaptation parameter set is regenerated. When the actual state of the device in the device status data with timestamps changes, the state of the device's 3D model is updated. The data dependency direction is integrated with the corresponding update triggering condition to construct a dependency relationship structure with data type as nodes and dependency relationship as edges. The dependency relationship structure is stored in a graphical manner.

8. The workshop virtual scene rendering method based on virtual reality according to claim 3, characterized in that, The process of establishing a mapping relationship between global hue parameters and illumination color parameters, determining the corresponding target illumination color parameters based on the global hue parameters, and updating the illumination color parameters in the illumination parameters to the target illumination color parameters includes: Collect video frame hue features and corresponding optimal lighting color parameter samples from different time periods and environments in the workshop. The optimal lighting color parameter samples include multiple hue ranges and corresponding lighting color values. The collected sample data is classified and divided into multiple hue intervals according to the numerical range of hue features. Each hue interval corresponds to one or more lighting color parameter samples. Calculate the average value of the lighting color parameter samples within each hue range, and determine the average value as the standard lighting color parameter corresponding to that hue range; A mapping table is created by organizing the hue ranges and their corresponding standard lighting color parameters. The mapping table includes the start and end values ​​of the hue ranges and their corresponding standard lighting color parameters. Find the hue range to which the global hue parameter belongs from the mapping table, and obtain the standard lighting color parameter corresponding to that hue range; Fine-tune the acquired standard lighting color parameters, and combine them with the regional functional characteristics in the computer-aided design layout data of the workshop. If the area to which the global hue parameter belongs is the processing area, increase the proportion of the red component in the standard lighting color parameter to meet the requirements for the detailed presentation of equipment in the processing area. If the area to which it belongs is the storage area, increase the proportion of the blue component in the standard lighting color parameter to meet the requirements for enhancing the sense of space in the storage area. The fine-tuned standard lighting color parameters are determined as the target lighting color parameters; The lighting parameters are extracted from the basic scene parameters of the preset virtual workshop scene, and the lighting color parameters are replaced with the target lighting color parameters to complete the update of the lighting color parameters.

9. A workshop virtual scene rendering system based on virtual reality, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the virtual reality-based workshop virtual scene rendering method according to any one of claims 1 to 8 by executing the machine-executable instructions.

10. A computer program product, characterized in that, The computer program product includes machine-executable instructions stored in a computer-readable storage medium. The processor of the virtual reality-based workshop virtual scene rendering system reads the machine-executable instructions from the computer-readable storage medium and executes the machine-executable instructions, causing the virtual reality-based workshop virtual scene rendering system to perform the virtual reality-based workshop virtual scene rendering method as described in any one of claims 1 to 8.

Images