1065 results about "Three dimensional data" patented technology

Embodiments of the present disclosure provide a three-dimensional scene reconstruction method and apparatus, a device, a medium, and a program product. The three-dimensional scene reconstruction method comprises: acquiring a scene image collected for a three-dimensional scene; on the basis of the scene image, determining sparse point cloud data and camera parameter information corresponding to the scene image; performing model initialization on the basis of the sparse point cloud data to obtain a current three-dimensional data model; performing rendering on the basis of the camera parameter information and the current three-dimensional data model to obtain a current color map, a current depth map and a current normal map from the perspective of the scene image, and determining a pseudo normal map from the perspective of the scene image on the basis of the current depth map; and on the basis of the current color map, the current normal map, the pseudo normal map, and an actual color map, training the current three-dimensional data model to obtain a trained target three-dimensional data model. By means of the technical solution provided by the embodiments of the present disclosure, higher-quality automatic reconstruction of three-dimensional scenes can be achieved, reducing reconstruction costs, and improving the level of detail and rendering quality of scene models.

The invention relates to a three-dimensional model compression transmission method and system based on dynamic feature perception, and relates to the technical field of three-dimensional live-action modeling, and the method comprises the steps: obtaining the original three-dimensional data of a three-dimensional model, carrying out the multi-modal feature extraction based on a model scene, dividing a key region, a transition region and a non-key region, and generating a three-dimensional model partition map, compressing each region according to the partition map label and a preset compression algorithm to generate a multi-resolution LOD sequence, determining a transmission data hierarchy in combination with a network state and equipment parameters, and finally rendering the model according to the transmission data hierarchy, user behavior information and an environment state to obtain a target three-dimensional model. The technical effects of effectively compressing, transmitting and rendering the three-dimensional model according to various factors such as different area characteristics, network states, equipment parameters and user behaviors of the model and improving the transmission efficiency and the rendering quality are achieved.

The invention provides a fault detection method and system based on an intelligent electric meter, and relates to the technical field of fault detection and diagnosis of the intelligent electric meter, and the key points of the technical scheme are that three-dimensional data are obtained, and the three-dimensional data comprise intelligent electric meter data, environment perception data, power grid topology data and line load data; predicting a normal data range of the next one minute in real time according to the three-dimensional data, and judging whether actual data exceed the normal data range or not; when the actual data exceed the normal data range, uploading all the data to a cloud end to analyze and detect faults; and collecting fault data in real time and generating a fault early warning list. The fault detection method and system based on the intelligent electric meter have the advantages that the fault false alarm rate is reduced, the fault detection response efficiency is enhanced, and fault risk early warning is achieved.

The invention discloses a road internal damage assessment method, device and equipment based on three-dimensional GPR data and a storage medium, and the method comprises the steps: determining a comprehensive disease weight corresponding to each disease in each layer of horizontal slice based on disease distribution feature information of a three-dimensional GPR horizontal slice data model, determining the internal damage rate of each layer of horizontal slice based on the comprehensive disease weight, determining the initial internal damage index of each layer of horizontal slice based on the internal damage rate, and performing through punishment correction on the through damage related layer to obtain the internal damage index; performing road internal damage evaluation on the road structure based on the internal damage index of each layer of horizontal slice; according to the method, the three-dimensional GPR horizontal slice data model is modeled, the disease distribution characteristics are analyzed in the horizontal direction, the three-dimensional geometric characteristics of the disease are accurately represented, the weight is objectively given to the disease, the internal damage index of each layer of horizontal slice is accurately calculated, and the internal damage of the road is accurately evaluated.

The invention relates to a forest tree diameter at breast height and tree height measurement method based on double-layer unmanned aerial vehicle laser radar point cloud fusion in a complex under-forest environment, and belongs to the technical field of forestry measurement. According to the method, multi-source point cloud registration fusion is carried out based on feature point matching and a weighted fusion algorithm; generating a ground DEM and carrying out point cloud height normalization; carrying out individual tree segmentation by adopting an improved mark control watershed algorithm; constructing a multi-dimensional feature vector for each tree, and calculating each feature; extracting a tree height based on the fused point cloud; an effective point cloud coverage index is introduced to judge the effectiveness of a 1.3 m height point cloud, weighted least square circle fitting is adopted to directly extract the diameter at breast height, and a pre-trained multi-feature fusion gradient lifting model is adopted to estimate the diameter at breast height for a tree with point cloud missing. According to the method, complete stand three-dimensional data is obtained through multi-source point cloud fusion, a multi-dimensional feature system is constructed, a complete mathematical calculation framework is established, and the technical problem that the diameter at breast height cannot be accurately measured due to complex under-forest shrub and grass shielding is effectively solved.

The invention relates to a three-dimensional data processing method and device based on Beidou differential positioning. The method comprises the steps that multiple pieces of original point cloud data collected by different scanning stations at different time points for a construction area are acquired; performing de-noising processing on each piece of original point cloud data to obtain de-noised point cloud data corresponding to each piece of original point cloud data; according to the multiple pieces of denoised point cloud data, identifying and removing point cloud data corresponding to a moving object existing in the construction area in combination with time sequence analysis or an optical flow method to obtain target point cloud data; performing Poisson reconstruction according to the target point cloud data to generate a triangular mesh surface model; and performing texture enhancement processing on each triangular patch in the triangular mesh surface model to obtain a three-dimensional model corresponding to the construction area. By means of the method, the three-dimensional model with centimeter-level geometric accuracy can be obtained.

The invention provides a three-dimensional data generation method and device, and the method comprises the steps: carrying out the planning of a text prompt before generation, optimizing the three-dimensional data in stages, introducing a reflection and correction mechanism in the generation process, recognizing and correcting defects through a multi-modal large model, and introducing a cross-view alignment constraint. And global planning and local fine adjustment of the three-dimensional generation process are realized. According to the method, a complex generation task is decomposed into a closed-loop process of planning-execution-reflection-correction, and multi-view consistency is forced from a feature level, so that geometric consistency and semantic accuracy of generated three-dimensional data can be effectively improved, the problems of structural artifacts and Janus which are easy to occur in the prior art are avoided, and the generation efficiency of the three-dimensional data is improved. And finally, high-quality three-dimensional data with consistent views are generated.

The invention discloses a multi-dimensional time series data dynamic evolution analysis method and device, a medium and electronic equipment. The method comprises the steps that time series data are obtained and subjected to structured preprocessing; performing multi-dimensional quantization and attribute endowing on each piece of time series data, and constructing a three-dimensional data node containing a space coordinate, an initial weight and a timestamp; based on the set of the three-dimensional data nodes in a specific time window, constructing a three-dimensional data density field capable of representing data hotspot distribution and intensity in the time window; in the three-dimensional data density field, identifying a density peak region as a core data cluster, calculating the cluster mass according to the density and the coverage range of the core data cluster, and calculating the association strength among different core data clusters; and serializing a plurality of continuous time window data field state snapshots to form a dynamic evolution view capable of showing data hot spot generation and extinguishing and data cluster fusion and splitting processes. According to the invention, abstract multi-dimensional data is converted into a visual three-dimensional density field.

The invention discloses a geometric constraint fitting point cloud filtering method for sea surface three-dimensional reconstruction, and belongs to the technical field of computer vision and three-dimensional reconstruction. The objective of the invention is to solve the problem of insufficient subsequent three-dimensional reconstruction precision caused by interference of reflection noise, mismatching points and the like in sea surface point cloud. The method specifically comprises the following seven steps: firstly, acquiring sea surface original point cloud through three-dimensional data acquisition equipment; a filtering technology is adopted to obtain a to-be-fitted point cloud; fitting a quadric surface through an improved RANSAC (Random Sample Consensus) algorithm to solve an initial parameter; constructing a comprehensive error function, and optimizing the model through gradient descent; effective inner points are screened through quadratic term coefficient constraint and a distance threshold value; and finally, iterating until a termination condition is met, and outputting an optimal effective point cloud. The method is high in noise rejection rate, the point cloud fits the sea surface form, and high-quality data support can be provided for sea surface fitting, sea wave simulation and unmanned ship control.

A construction process support system that enables diverse users to quickly and intuitively create construction model data according to their intentions using natural language input and grid-based positioning. The system comprises hardware processors, software modules, and databases configured to display user interface pages to terminals and control natural language input messages. When input messages contain grid labels that specify coordinates on pages with character codes regarding construction structures, the system requests a Large Language Model (LLM) to generate instruction codes describing construction model data including the grid labels as position information. The LLM generates the requested instruction codes. The system stores the construction model data as searchable data in the database, forwards input messages to chat areas, and renders each structure as three-dimensional data at grid label positions on page canvases.

The invention discloses a vehicle-mounted human-computer interaction method and device and electronic equipment, and belongs to the technical field of human-computer interaction. The method specifically comprises the following steps: collecting vehicle-mounted scene information; the vehicle-mounted scene information comprises vehicle environment information and / or user behavior information; determining a target interface element corresponding to the vehicle-mounted scene information, and generating target three-dimensional data corresponding to the target interface element according to the vehicle-mounted scene information; the target interface element is an existing interface element, or the target interface element is a newly added interface element; rendering the target three-dimensional data to obtain a three-dimensional HMI interface corresponding to the target interface element; and displaying the three-dimensional HMI interface. According to the method, the display mode, the layout or the interaction priority of the interface elements in the three-dimensional HMI interface can be automatically adapted based on the vehicle environment or the user behavior reflected by the vehicle-mounted scene information.

The invention discloses a marine geological environment multi-dimensional data reconstruction and dynamic visualization mapping method and system, and belongs to the technical field of marine information technology and data visualization. In order to solve the problems that in the prior art, standard data asset structure is isolated, heterogeneous data deep fusion is difficult, and mass data visualization performance is bottleneck, the method mainly adopts the following technical scheme that standardized multi-source data is received, a data volume mapping rule is established, discrete observation data is reconstructed and fused into a unified and continuous three-dimensional data volume, and the three-dimensional data volume is obtained; and executing a three-dimensional space pattern recognition algorithm on the three-dimensional data volume to recognize a feature entity, and providing on-demand query and multi-resolution access services through a visual mapping interface. According to the method, efficient automatic modeling and high-performance interactive analysis of mass and multi-dimensional marine geological environment data can be realized.

The invention discloses a BIM-based building design process dynamic management method, and relates to the technical field of BIM. Spatial state information of a construction task is extracted from a BIM model and is uniformly expressed as a six-dimensional space body vector TBox, standardized processing of a task geometric boundary is realized, and a task time and space three-dimensional data set Cx is generated; and the spatial evolution trend between tasks can be perceived in advance. On the basis, the system carries out space overlapping detection on any task pair in the three-dimensional data set Cx and calculates a task conflict vector CVec, whether conflicts exist or not is judged, directionality and space centralization of the conflicts are quantized by combining task path direction vectors and overlapping volume information, conflict severity index CIdx values of all the task pairs are summarized, and the conflict severity index CIdx values of all the task pairs are calculated. Dynamic judgment and optimization triggering of system-level construction risks are achieved, and building design and construction scheduling can have the capabilities of feedforward perception, self-adaptive adjustment and feedback optimization.

The invention provides a multi-dimensional and dynamic digital assessment system and assessment method. Comprising a support module (comprising an index knowledge base, an algorithm operation base and an evaluation model base), a multi-source index fusion center (for constructing a three-dimensional data model and integrating a multi-dimensional data resource pool), a dynamic rule configuration engine (for realizing flexible configuration through a visual interface, a model adaptation mechanism and double-track verification) and a core business module (for managing assessment indexes, evaluating indexes and evaluating indexes). And evaluating the model and plan). The support module lays a basic resource layer, a fusion center constructs a data base, a configuration engine realizes flexible verification, the business module drives full-process execution, and full-process digital management from index creation, model configuration to examination execution is realized. The problems of iteration lag, multi-scene adaptation redundancy, weak data verification and the like of a traditional system are solved, index legality is guaranteed through static grammar detection and dynamic simulation operation, data traceability is achieved through a temporary database and full-link recording, and the dynamic response capacity, scene compatibility and data reliability of the assessment system are improved.

The invention relates to a dynamic simulation method and system for a tooth movement process based on a Type model, and the method comprises the steps: carrying out the data collection of an artificial tooth and a fixed appliance through three-dimensional scanning, and obtaining the three-dimensional data of a tooth jaw structure; performing overlap analysis on the crown root model according to the three-dimensional data to obtain a digital dental model containing periodontal tissues; performing material attribute assignment on an anatomical structure in the digital tooth jaw model according to preset material parameters to obtain a biomechanical tooth jaw model; and performing time recursion iteration on the biomechanical tooth jaw model through finite element analysis to obtain four-dimensional dynamic simulation data for simulating a tooth movement process, so as to provide visual orthodontic process teaching demonstration based on the four-dimensional dynamic simulation data. According to the invention, the dynamic visual simulation of the tooth movement process is realized, the feedback of the Type Dont system can be quantified, and the teaching interactivity is improved.

The invention provides a rock mass internal pore intelligent characterization and three-dimensional reconstruction method and system, and relates to the technical field of rock mass internal structure analysis, and the method comprises the steps: obtaining a two-dimensional image sequence of a rock core in a target rock mass through a focused ion beam scanning electron microscope slicing-imaging sequence working mode; performing pore segmentation on each two-dimensional image in the two-dimensional image sequence by using the pore segmentation network to obtain a plurality of pore two-dimensional segmentation maps; on the basis of the pore two-dimensional segmentation map, constructing three-dimensional binary volume data of the pores according to a slice sequence; according to the three-dimensional binary volume data, performing three-dimensional reconstruction and structural characterization on rock core pores; according to the method, an imaging-recognition-reconstruction-evaluation integrated workflow is constructed, and a full-chain closed loop from nanoscale real three-dimensional data acquisition to intelligent recognition, three-dimensional reconstruction and quantitative analysis is realized.

The invention discloses a dynamic adaptive operation, maintenance and metering optimization method for an electric energy metering box, and the method comprises the steps: collecting electrical parameters, environmental parameters and box body state parameters in real time, constructing a three-dimensional data set through preprocessing, and carrying out the local storage and preliminary analysis through edge calculation; the condensation occurrence probability is calculated based on the environmental parameters, and self-adaptive regulation and control are triggered and a ventilation structure is linked; abnormal features are extracted, abnormity is recognized in a rule judgment and trend analysis double-engine mode, and graded early warning is carried out; performing comprehensive state scoring in combination with the three-dimensional data set, and matching a differentiated inspection mechanism; presetting a metering error reference value of a temperature and humidity interval, calculating a compensation value in real time and remotely correcting parameters of the electric energy meter; and establishing a shared data platform of the power supply and the power consumption, and realizing full-process data tracing and automatic report generation. According to the method, the condensation prevention and control accuracy is improved, the abnormal false alarm rate is reduced, the operation and maintenance resource allocation is optimized, the metering accuracy is improved, and the data collaboration and tracing capability is enhanced.

The invention provides a three-dimensional part retrieval method based on graph similarity search, relates to the field of computer graphics, and solves the technical problems that in the prior art, geometric and design semantic information of CAD cannot be fully utilized, and a topological relation is difficult to capture and display, so that the retrieval efficiency is low. The method comprises the following steps: acquiring part three-dimensional data of a computer-aided design (CAD) model, and constructing a training data set; constructing an edge-surface connection diagram based on the three-dimensional data of the part; calculating a graph editing distance (GED) matrix of all edge surface connection graphs in the training data set as a supervision signal; based on the GED matrix, training a sorting model, mapping an edge-surface connection graph to a hidden space, and constructing a part vector database according to an output graph-level embedding vector; the sorting model is constructed based on a graph attention network; and inputting a to-be-queried CAD part into the trained sorting model to obtain a feature vector, carrying out nearest neighbor search in the part vector database, and returning a similar part result.

The invention discloses a rotor initial mechanical angle identification method of a built-in permanent magnet synchronous motor, and relates to the field of motor control. When the motor is offline, a fitting function model corresponding to a three-phase high-frequency response current amplitude and a rotor mechanical angle of the permanent magnet synchronous motor is measured and established in advance; constructing a three-dimensional data table by using the fitting function model; when the motor runs, a high-frequency injection voltage between lines is input, a real-time three-phase high-frequency current response amplitude value is collected, and matching search is carried out on the three-dimensional data table, so that a mechanical angle mapped by an obtained error minimum value is used as an initial mechanical angle of a rotor obtained by matching search; and inputting the initial mechanical angle of the rotor into a control system of the permanent magnet synchronous motor. Thus, when the rotor is static at any mechanical angle, the absolute mechanical angle position of the rotor can be directly and uniquely determined without any pre-positioning operation, and the problems of running blind areas, pre-positioning strokes, reverse rotation risks and vibration noise are effectively solved.

The invention relates to the technical field of crossing of geophysical exploration and artificial intelligence, and discloses a three-dimensional VSP data reconstruction method based on a three-dimensional blind sidewalk time frequency-space multi-domain interpolation network model, and the method comprises the steps: obtaining a to-be-reconstructed three-dimensional VSP data volume containing an irregular missing sidewalk; inputting the three-dimensional VSP data volume into a trained three-dimensional space-time-frequency mixed domain convolutional network model; performing down-sampling on the input data through an encoder to extract initial spatial-temporal characteristics; inputting the initial spatiotemporal features into a residual spatiotemporal-frequency wavenumber domain joint convolution feature fusion module to obtain fusion features; and inputting the fusion features into a decoder, and reconstructing a complete three-dimensional VSP data volume through up-sampling operation. According to the invention, through joint learning of ground seismic data (providing a macrostructure) and illuminance data (providing a physical propagation path), a reconstruction result with a reasonable geologic structure and extremely high fidelity of wave field features can be generated.

The invention discloses a three-dimensional model dynamic rendering method and a storage medium, and the method comprises the steps: obtaining three-dimensional data of a target scene, obtaining a plurality of voxels, and enabling each voxel to correspond to a plurality of voxel attributes and an initial comprehensive displacement vector; in any voxel in any time step, obtaining a comprehensive displacement vector of the voxel in the current time step according to a preset physical deformation rule and / or animation curve deformation rule, and performing coordinate transformation on the voxel; after the coordinate transformation, updating the voxel attribute of the voxel corresponding to the current time step length; and performing real-time three-dimensional rendering on the target scene according to the vertex normal of each voxel at the current time step length and the voxel attribute. The physical consistency is ensured in the dynamic deformation process by synthesizing the displacement vector; by updating the voxel attributes, the voxel attributes are kept in smooth transition after deformation; and high-quality real-time rendering of the target scene is realized by recalculating the deformed vertex normal and combining the illumination model.

A computer implemented method for determining a 3D position of a surgical object relative to a region of interest in a patient's anatomy. The method includes receiving an X-ray image and a 3D data set. The X-ray image is a 2D projection image generated with an imaging direction and defining a 3D image coordinate system. The 2D projection image may depict a part of a patient's anatomy and a part of the surgical object. The received 3D data set describes the part of the patient's anatomy and a region of interest in a 3D data coordinate system. In order to determine a spatial relation between the 3D image coordinate system and the 3D data coordinate system, the part of the patient's anatomy is localized based on the X-ray image. Subsequently, the 3D position of the surgical object relative to the region of interest is determined in the 3D coordinate system.

A three-dimensional data encoding method includes: calculating coefficient values from pieces of attribute information of three-dimensional points included in point cloud data; quantizing the coefficient values individually to generate quantized values; and generating a bitstream including the quantized values. Each of the coefficient values belongs to any one of layers. In the quantizing, each of the coefficient values is quantized using a quantization parameter for a layer to which the coefficient value belongs among the layers, and the bitstream includes first information and pieces of second information, the first information indicating a reference quantization parameter, the pieces of second information being for calculating quantization parameters for the layers from the reference quantization parameter.

The invention discloses an in-service expressway full-life-cycle model parametric modeling method, which provides a unified parametric modeling method for standardized design assets, solves the problems that an in-service expressway asset model library is imperfect, low in modeling efficiency and the like, effectively improves modeling efficiency and modeling quality, and is suitable for large-scale popularization and application. And a refined digital base can be provided for highway digital twin. According to the technical scheme provided by the invention, the road center line information is obtained based on the laser point cloud and the panoramic image data; highway assets in non-standardized design are sorted, and asset modeling is completed based on live-action three-dimensional data; sorting highway assets of standardized design, and establishing a highway standardized asset BIM model library; and obtaining detailed information of the standardized design assets, determining a parameterized modeling and fusion scheme of the total factor assets, and finally completing road section modeling.

The invention discloses polishing equipment for large burrs on the surface of a casting, and the polishing equipment comprises a three-dimensional scanner which is arranged on one side of a polishing mechanism and is arranged in parallel with the polishing mechanism; the three-dimensional scanner is fixedly arranged on the visual angle deviation mechanism, the burr sensing mechanism is fixedly arranged on the side edge of the grinding mechanism, and the technical problem that when a common three-dimensional scanner faces large shielding burrs, due to the fact that the sight line is blocked, scanning data are lost is solved. The burr sensing mechanism is in physical contact with and detects the burrs, the linkage control mechanism directly and mechanically drives the visual angle deviation mechanism, automatic deflection of the visual angle of the three-dimensional scanner can be achieved when the visual angle of the three-dimensional scanner encounters the burrs, burr shielding is avoided, three-dimensional data of a shielded area can be more sufficiently obtained, and the three-dimensional scanning efficiency is improved. And a more reliable path planning basis is provided for subsequent grinding.

The invention discloses a semantic-guided lightweight three-dimensional reconstruction method based on three-dimensional Gaussian, and the method comprises the steps: inputting a multi-view two-dimensional image, and extracting two-dimensional semantic features; projecting the two-dimensional semantic features to the three-dimensional Gaussian primitives; learning three-dimensional semantic distribution in a knowledge distillation mode, and predicting three-dimensional semantic embedding features; dividing a large scene into a plurality of sub-regions, and performing divide-and-conquer three-dimensional Gaussian optimization and semantic learning; performing semantic-guided lightweight modeling on the three-dimensional Gaussian primitives in each sub-region; executing sub-scene fusion, and performing quantitative compression on the three-dimensional data in the fused scene; and a rasterization renderer is used to carry out efficient real-time rendering under any visual angle, and a lightweight large-scene three-dimensional reconstruction result is output. According to the method provided by the invention, the real-time rendering performance of a complex scene can be remarkably improved, the model volume is effectively compressed, meanwhile, the model training process is accelerated, and the method is suitable for efficient three-dimensional visual reconstruction tasks.

The invention discloses a storage battery abnormal state identification method, system and device and a medium, and particularly relates to the technical field of storage battery monitoring, and the technical key points are as follows: obtaining working performance parameters and engineering physical parameters of a storage battery to be identified, constructing a multi-dimensional parameter time sequence, and segmenting the multi-dimensional parameter time sequence to obtain three-dimensional data pieces; sliding on the multi-dimensional parameter time sequence by using a sliding window with a preset size, calculating a linear correlation feature, a nonlinear correlation feature and a time sequence trend feature of the multi-dimensional parameter time sequence in each sliding window, and fusing to obtain a correlation feature vector; splicing the three-dimensional data pieces and the associated feature vectors to obtain identification parameter features; inputting the identification parameter features into a pre-trained abnormal state identification model, and identifying to obtain probability distribution of an abnormal state; the abnormal state recognition model is obtained by training an AA-LSTM network; and when the probability distribution meets a preset judgment condition, outputting a corresponding state judgment result.

The invention discloses a machining method, device and equipment for an ultra-long groove cut pipe and a storage medium. The machining method is matched with a pipe with the diagonal line smaller than or equal to 360 mm and the length smaller than or equal to 18500 mm. The method comprises the steps that firstly, a groove cutting parameter prediction model is pre-constructed, and parameters of the pipe to be machined are input to obtain target groove parameters and a tailing critical value; then the first chuck and the second chuck cooperatively clamp the pipe, multiple sets of follow-up supporting mechanisms are started to compensate the deflection of the middle section, forward path cutting is planned according to target parameters, and the remaining length is detected in real time; when the remaining length reaches a critical value, the second chuck loosens and avoids, the first chuck and the third chuck are switched to cooperatively clamp, the first chuck penetrates into an empty clamping hole in the second chuck, and the follow-up support is synchronously adjusted to adapt to the tail section posture; and finally, real-time three-dimensional data of the tail section are collected, a reverse cutting path is generated, the laser cutting head is controlled to complete zero-tailing groove cutting, tailings are greatly reduced, and the pipe utilization rate and the cutting precision are improved.

The invention discloses a method and system for constructing a universal digital twin asset data model for multi-source heterogeneous data exchange and sharing, and belongs to the technical field of digital twin. The method comprises the following steps: obtaining three-dimensional data structures and three-dimensional data indexes of different geometric forms in a digital twin model, the spatial topological relation between the three-dimensional data structures is identified; converting texture map information of the three-dimensional data structure into a unified format, normalizing coordinates of the texture map information, and unifying vertex coloring information of the three-dimensional data structure to a standard color space; and performing material analysis and description on different material characteristics of the digital twin model, dynamically calling node objects corresponding to the different material characteristics to obtain real material information, and finally dynamically outputting to obtain a general three-dimensional data model. According to the invention, the problem that a unified digital twin floor scene is difficult to generate due to different sources and structures of current model data can be solved.

The invention discloses a pronunciation recognition method based on multi-modal fusion and related equipment, and the method comprises the steps: obtaining a to-be-processed millimeter-wave radar signal and three-dimensional visual data, carrying out the preprocessing of the obtained millimeter-wave radar signal, and determining a radar data set; preprocessing the three-dimensional visual data to obtain a three-dimensional data set; performing feature extraction on the radar data set to obtain radar spatial-temporal features; performing feature extraction on the three-dimensional data set to determine three-dimensional visual features; performing feature fusion on the radar spatial-temporal features and the three-dimensional visual features according to a cross attention mechanism to obtain a feature fusion vector; inputting the obtained feature fusion vector into a trained feature decoupling network for identification, and determining pathological features and pronunciation features; and generating a corresponding pronunciation result according to the pathological features and the pronunciation features. According to the embodiment of the invention, the recognition accuracy can be improved. The method can be widely applied to the technical field of speech recognition.