Funeral service full-chain intelligent control cloud brain platform based on digital twinning scene

Abstract

The invention discloses a funeral service full-chain intelligent control cloud brain platform based on a digital twin scene, and relates to the technical field of intelligent funeral services, and the platform comprises a parameter conversion module which constructs a four-dimensional space-time manifold model, converts a four-dimensional tensor of a corruption process into a space-time manifold parameter set through a multi-scale space-time mapping algorithm, and transmits the four-dimensional tensor of the corruption process to a data processing module; and finally generating a four-dimensional space-time manifold comprehensive configuration parameter set. And the scheme generation module is used for performing multi-path parallel deduction by adopting a quantum inspiration tensor network engine according to the four-dimensional space-time manifold comprehensive configuration parameter set, and generating an optimal funeral service process scheme through Monte Carlo simulation and optimization of a variable component child solicitation solver. According to the method, the decay rate predicted by the LSTM network and the atomic-scale displacement data simulated by the AMBER force field are coupled through a multi-scale coupling algorithm, so that the decay process prediction matrix can accurately reflect the space-time correlation between microreactions such as proteolysis and the like and tissue decay.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent funeral services, and in particular to a cloud-brain platform for intelligent control of the entire funeral service chain based on digital twin scenarios. Background Art

[0002] In the field of funeral services, existing technologies typically combine multimodal data acquisition with empirical models to monitor body condition and plan service processes. Conventional methods use medical imaging equipment to acquire structural data, combine it with temperature and humidity sensors to monitor environmental parameters, and establish linear regression models based on historical empirical data to predict the progression of decay. At the data processing level, traditional methods primarily use three-dimensional reconstruction algorithms to construct a geometric model of the body, simulate tissue changes through finite element analysis, and trigger service process adjustments based on preset thresholds. This type of technical solution can achieve basic body condition monitoring and process management, meeting the basic needs of funeral services.

[0003] However, existing methods have significant limitations in coupling spatiotemporal dimensions and multiscale modeling. Traditional linear regression models struggle to accurately characterize the nonlinear spatiotemporal evolution of the decay process, especially when dealing with the cross-scale correlation between protein molecular motion and macroscopic tissue changes. Empirical threshold triggering mechanisms are unable to dynamically adapt to the biological differences between individual remains and lack quantitative analysis of the interaction between genetic factors and environmental parameters. Furthermore, conventional three-dimensional reconstruction algorithms have limited ability to analyze changes in molecular-level dielectric properties, resulting in insufficient spatial resolution of protein distribution heat maps, which affects the accuracy of decay predictions. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a full-chain intelligent control cloud brain platform for funeral services based on digital twin scenarios to solve the problems of insufficient cross-scale prediction accuracy of corruption processes and insufficient dynamic generation of personalized service solutions in existing technologies.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The present invention provides a full-chain intelligent control cloud brain platform for funeral services based on digital twin scenarios, which includes a data acquisition module, which obtains molecular-level point cloud data and environmental parameters of the remains, uses PointNet++ to build a three-dimensional biological tissue model, and builds a protein distribution heat map based on dielectric constant distribution data, and finally generates an initial data set; a corruption prediction module, based on the initial data set, uses the LSTM-molecular dynamics joint prediction method to establish a corruption diffusion equation, generates a corruption process prediction matrix containing time and space dimensions, generates a dynamic corruption heat map through Hinton, and encodes it into a four-dimensional tensor of the corruption process; parameters The conversion module constructs a four-dimensional space-time manifold model, and converts the four-dimensional tensor of the corruption process into a space-time manifold parameter set through a multi-scale space-time mapping algorithm, and finally generates a four-dimensional space-time manifold comprehensive configuration parameter set; the solution generation module uses a quantum-inspired tensor network engine to perform multi-path parallel deduction based on the four-dimensional space-time manifold comprehensive configuration parameter set, and generates the optimal funeral service process solution through Monte Carlo simulation and variational quantum eigensolver optimization; the execution module decouples the optimal funeral service process solution into real-time control instructions and executes them, while making corrections based on real-time body corruption data, and finally generates a full-chain funeral service quality report.

[0007] As an optimal solution for the full-chain intelligent control cloud brain platform for funeral services based on digital twin scenarios described in the present invention, the use of PointNet++ to construct a three-dimensional biological tissue model refers to performing sampling and grouping operations through PointNet++, gradually capturing point cloud features from local to global, and performing interpolation upsampling and cross-layer feature splicing through FP to generate dense feature point cloud data. At the same time, a dynamic graph convolutional neural network is used for point cloud semantic segmentation, and a Poisson surface reconstruction algorithm is combined to generate a three-dimensional biological tissue model.

[0008] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario of the present invention, the protein distribution heat map is constructed based on the dielectric constant distribution data, and the initial data set is finally generated. The steps are as follows: Obtaining the dielectric constant distribution data of the deceased tissue, and converting the dielectric constant distribution data into protein concentration distribution data through the dielectric relaxation spectrum quantitative analysis method; The Kriging interpolation algorithm was used to spatially complete the protein concentration distribution data. The color mapping function of the OpenGL shader was used to map the spatially completed protein concentration distribution data to an RGB gradient color spectrum to generate a protein distribution heat map. Molecular-level point cloud data, environmental parameters, three-dimensional biological tissue models, and protein distribution heat maps are integrated to generate an initial dataset.

[0009] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario described in the present invention, the steps are as follows: the LSTM-molecular dynamics joint prediction method is used to establish the corruption diffusion equation and generate the corruption process prediction matrix containing the time and space dimensions. Predict the decay rate of corpses through a three-layer LSTM network; Based on the grid topology of the 3D biological tissue model, the AMBER force field is used to simulate the chemical reactions of the remains and obtain atomic-level displacement data of the decay process. The decay rate of the remains and the atomic-level displacement data of the decay process are coupled through a multi-scale coupling algorithm to establish a decay diffusion equation and generate a decay process prediction matrix including time and space dimensions.

[0010] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario described in the present invention, the dynamic corruption heat map is generated by Hinton and encoded into a four-dimensional tensor of the corruption process. The steps are as follows: Quantify the substance concentration values in the corruption process prediction matrix to generate the degree of corruption of the remains; The corruption process prediction matrix is mapped into a dynamic corruption heat map through Hinton's method based on Grad-CAM. Using multi-dimensional group structured coding, the time and space dimensions , the degree of decay of the remains, and the RGB values in the dynamic decay heat map are encoded as a four-dimensional tensor of the decay process that can be traced back in time and space.

[0011] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario described in the present invention, wherein: the construction of the four-dimensional space-time manifold model refers to extracting the persistent homology features of the gene sequence through TDA, the steps are as follows: Extract persistent homology features of gene sequences through TDA; The DNA topological energy tensor is generated by performing tensor field transformation on the persistent homology features through TDA; Based on the four-dimensional tensor of the corruption process and the DNA topological energy tensor, the initial space-time manifold is constructed through Riemannian manifold mapping; On the initial space-time manifold, the curvature of the manifold is corrected by conformal geometric flow to generate a four-dimensional space-time manifold model.

[0012] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario described in the present invention, wherein: the four-dimensional tensor of the corruption process is converted into a set of space-time manifold parameters through a multi-scale space-time mapping algorithm, and finally a four-dimensional space-time manifold comprehensive configuration parameter set is generated. The steps are as follows: The four-dimensional tensor of the corruption process is decoupled using a spatiotemporal decoupling algorithm, and macroscopic distribution parameters are generated through spatial gridding. At the same time, corrections are made based on perturbation theory. The corrected macroscopic distribution parameters are then fused with the time evolution sequence through a multi-scale coupling algorithm to generate a set of spatiotemporal manifold parameters. The space-time manifold parameter set is input into the four-dimensional space-time manifold model for dynamic evolution analysis to generate a four-dimensional space-time manifold comprehensive configuration parameter set.

[0013] As a preferred solution of the funeral service full-chain intelligent control cloud brain platform based on the digital twin scenario described in the present invention, wherein: the quantum-inspired tensor network engine is used to perform multi-path parallel deduction, and the optimal funeral service process plan is generated through Monte Carlo simulation and variational quantum eigensolver optimization. The steps are as follows: Through the tensor encoding mapping algorithm, the comprehensive configuration parameter set of the four-dimensional space-time manifold is encoded into the tensor nodes of the quantum-inspired tensor network engine; Based on tensor nodes, the quantum-inspired tensor network engine simultaneously generates and deduces multiple types of funeral paths, performs sampling analysis on each type of path through Monte Carlo simulation, and generates multi-dimensional decision data packets; The multidimensional decision data package is optimized into the Hamiltonian of the funeral path through the variational quantum eigensolver, and the optimal funeral service process plan is generated through the quantum approximate optimization algorithm.

[0014] As a preferred solution of the digital twin scenario-based funeral service full-chain intelligent control cloud brain platform described in the present invention, the optimal funeral service process solution is decoupled into real-time control instructions and executed, and correction is performed based on real-time body corruption data, and finally a funeral service full-chain quality report is generated. The steps are as follows: Using the Poisson spatiotemporal gridding algorithm, the optimal funeral service process solution is parsed into spatiotemporal control nodes containing coordinates, timestamps, and operation instructions. The IoT instruction compilation engine then converts the spatiotemporal control nodes into real-time control instruction sets that can be directly executed by the Cloud Brain platform. During the execution of the control instruction set, the real-time control instructions are corrected through the PID control algorithm based on the real-time collected body decay data, and a quality report for the entire funeral service chain is generated.

[0015] As a preferred solution of the digital twin scenario-based funeral service full-chain intelligent control cloud brain platform of the present invention, wherein: the molecular-level point cloud data of the remains includes tissue dielectric constant distribution point cloud and 5A resolution tomography point cloud data; The environmental parameters include real-time monitored data of temperature, humidity and atmospheric pressure of the body storage environment.

[0016] The present invention achieves the following beneficial effects: A multiscale coupling algorithm couples the decay rate predictions from an LSTM network with atomic-level displacement data from AMBER force field simulations, enabling the decay process prediction matrix to accurately reflect the spatiotemporal correlations between microscopic reactions such as proteolysis and tissue decay. A quantum-inspired tensor network engine encodes the parameters of a four-dimensional spacetime manifold as tensor nodes, utilizes Tucker decomposition to construct a funeral path deduction framework, and combines Monte Carlo sampling with variational quantum optimization to achieve multi-objective decision-making under the dual conditions of DNA constraints and environmental responses. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a schematic diagram of the full-chain intelligent control cloud brain platform for funeral services based on digital twin scenarios.

[0019] Figure 2 Flowchart for data acquisition and construction of 3D biological tissue models.

[0020] Figure 3 Flowchart for corruption prediction and 4D tensor generation.

[0021] Figure 4 Flowchart of the construction and service scheme for space-time manifolds. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4, is an embodiment of the present invention, which provides a full-chain intelligent control cloud brain platform for funeral services based on digital twin scenarios, including the following steps: The data acquisition module acquires molecular-level point cloud data and environmental parameters of the remains, constructs a 3D biological tissue model using PointNet++, and constructs a protein distribution heat map based on the dielectric constant distribution data, ultimately generating an initial dataset. The molecular-level point cloud data of the remains include tissue dielectric constant distribution point cloud and 5A resolution tomography point cloud data.

[0026] Environmental parameters include real-time monitoring of the temperature, humidity and atmospheric pressure data of the body storage environment.

[0027] It should be noted that the tissue dielectric constant distribution point cloud is obtained by scanning the surface and internal tissues of the remains point by point using a terahertz wave scanning array, recording the amplitude attenuation and phase shift resulting from the interaction between the terahertz wave and the biological tissue, and converting this data into three-dimensional spatial distribution data of the dielectric constant using a quantitative dielectric relaxation spectrum analysis method. 5A resolution tomographic point cloud data uses a high-precision X-ray computed tomography scanner to perform sub-nanometer layered scanning of the remains. After reconstructing the tomographic image using a back-projection algorithm, an edge detection algorithm is used to extract tissue boundaries and generate three-dimensional point cloud coordinates. Real-time monitoring of the remains' storage environment temperature data is collected using a platinum resistance temperature sensor, humidity data is collected using a capacitive polymer film humidity sensor, and atmospheric pressure data is collected using a piezoresistive air pressure sensor.

[0028] Based on molecular-level point cloud data from the remains, PointNet++ performs sampling and grouping operations to gradually capture point cloud features from local to global. FP then performs interpolation upsampling and cross-layer feature concatenation to generate dense feature point cloud data with complete spatial details. Furthermore, PointNet++ first performs multi-level sampling on the tissue dielectric constant distribution point cloud and 5A resolution tomographic point cloud data from the remains' molecular-level point cloud data, selecting feature points as local region centers using the farthest point sampling algorithm. A sphere query algorithm is used to construct a spatial neighborhood point set at each sampling level, which is then input into a weighted multi-layer perceptron network to extract local geometric features. During the feature propagation phase, the FP algorithm performs linear interpolation upsampling on the feature points of the low-resolution level to match the feature point density of the previous level. The upsampled point cloud features are then cross-layered with the skip connection features of the corresponding level, integrating local details with global contextual information through a feature fusion unit. The sampling-grouping-feature extraction-interpolation upsampling-feature splicing process is repeated to ultimately generate dense feature point cloud data that retains complete spatial details.

[0029] Based on dense feature point cloud data, a dynamic graph convolutional neural network (DGCNN) is used for point cloud semantic segmentation, and a Poisson surface reconstruction algorithm is combined to generate a three-dimensional biological tissue model; Furthermore, a dynamic graph convolutional neural network constructs a dynamic graph structure based on the spatial neighborhood relationships of dense feature point cloud data. Edge convolution operations are used to extract local geometric features, and a multi-layer perceptron is used to predict tissue type probabilities on a point-by-point basis. The dense feature point cloud with tissue labels is then fed into a Poisson surface reconstruction algorithm, which calculates implicit surface functions using an octree spatial partitioning algorithm. Isosurfaces are extracted to generate an initial triangular mesh. A Laplace smoothing algorithm iteratively adjusts the vertex positions of the triangular mesh to achieve a continuous curvature distribution, eliminating surface noise while preserving anatomical structural features. Ultimately, a 3D biological tissue model is generated.

[0030] It should be noted that the dense feature point cloud with tissue labels is obtained through dynamic graph convolutional neural network processing. The dynamic graph convolutional neural network constructs a dynamic graph structure based on the spatial neighborhood of the point cloud, uses edge convolution to extract local geometric features, and realizes point-by-point classification through a multi-layer perceptron. It assigns probability values of tissue types such as muscle, fat, and bone to each point cloud coordinate, and outputs dense feature point cloud data carrying tissue category labels.

[0031] By scanning the array with terahertz waves, the dielectric constant distribution data of the body tissue is obtained and normalized; Furthermore, a terahertz wave scanning array scans the body's surface and internal tissues point by point, emitting terahertz pulses in the 0.1-10 THz frequency range and receiving reflected signals, recording the amplitude attenuation and phase delay data for each spatial coordinate point. By calculating the complex amplitude ratio of the incident and reflected waves, the real and imaginary parts of the complex dielectric constant at each location are obtained. Then, through minimum-maximum normalization, the real part of the dielectric constant is mapped to the range of 0-1, generating normalized dielectric constant distribution data.

[0032] The normalized dielectric constant distribution data were converted into protein concentration distribution data by dielectric relaxation spectrum quantitative analysis method; Furthermore, the quantitative analysis method of dielectric relaxation spectrum first performs frequency domain decomposition on the normalized dielectric constant distribution data, and based on the quantitative relationship between protein molecules and dielectric relaxation parameters in the three-dimensional tissue model, calculates the contribution weight of each relaxation component to protein polarization through multivariate linear regression, and finally converts the dielectric constant distribution into protein concentration distribution data.

[0033] It should be noted that the relaxation component refers to the three features in the dielectric relaxation spectrum that characterize different molecular motion modes, corresponding to the polarization response characteristics of protein main chain skeleton motion, side chain rotation and free water molecule reorientation, so as to accurately distinguish the different motion states and conformational changes of protein molecules in the deceased tissues, and provide a basis for the subsequent quantitative calculation of protein concentration distribution.

[0034] The Kriging interpolation algorithm was used to spatially complete the protein concentration distribution data. The color mapping function of the OpenGL shader was used to map the spatially completed protein concentration distribution data to an RGB gradient color spectrum to generate a protein distribution heat map. Furthermore, the Kriging interpolation algorithm predicts the protein concentration in unsampled areas by linearly weighting the concentration values of adjacent known points based on the spatial positions and concentration values of known points in the protein concentration distribution data, completing the three-dimensional space completion. The completed protein concentration distribution data is input into the OpenGL rendering process, and color mapping is implemented in the fragment shader. The normalized concentration values are linearly interpolated to a predefined RGB gradient color spectrum using GLSL built-in functions, with the lowest concentration mapped to blue, the highest concentration mapped to red, and intermediate concentrations corresponding to transition colors such as cyan, green, and yellow. The final protein distribution heat map is output as a three-dimensional texture. Each texture pixel contains coordinates and RGB color values, accurately reflecting the spatial distribution characteristics of proteins in the deceased tissue.

[0035] Molecular-level point cloud data, environmental parameters, three-dimensional biological tissue models, and protein distribution heat maps are integrated to generate an initial dataset.

[0036] The corruption prediction module uses the LSTM-molecular dynamics joint prediction method based on the initial dataset to establish the corruption diffusion equation, generate a corruption process prediction matrix including time and space dimensions, generate a dynamic corruption heat map through Hinton, and encode it into a four-dimensional tensor of the corruption process; Based on the protein concentration distribution data, a three-layer LSTM network (including three hidden layers) was used to predict the decay rate of the remains. Furthermore, protein concentration distribution data is fed into a three-layer LSTM network in time series. The first LSTM layer extracts short-term temporal features of protein concentration changes, the second LSTM layer captures medium-term decay patterns, and the third LSTM layer establishes long-term decay trends. Each LSTM unit selectively memorizes or forgets features through a gating mechanism. A fully connected layer at the network output maps the time series features into decay rate predictions. During training, a mean squared error loss function is used to optimize network parameters to ensure that the predicted decay rates are consistent with the measured values.

[0037] Based on the grid topology of the 3D biological tissue model, the AMBER force field is used to simulate the chemical reactions of the remains and obtain atomic-level displacement data of the decay process. Furthermore, the AMBER force field first assigns a corresponding atom type and charge parameter to each vertex based on the mesh topology of the 3D tissue model, establishing a multicomponent mechanical model consisting of protein, fat, and water molecules. Under periodic boundary conditions, the Verlet algorithm integrates the equations of motion, calculating the van der Waals, electrostatic, and bond forces acting on the atoms at each time step. Decay reactions are triggered by defining activation energy thresholds based on historical decay data. When the local environment meets the required temperature, humidity, and pH, reaction pathways such as protein hydrolysis and fat oxidation are automatically executed. During the simulation, the positional changes of all atoms are recorded, generating displacement data containing the 3D motion trajectories of carbon, hydrogen, and oxygen atoms with a time resolution of 1 femtosecond. Atomic-level displacement data of the decay process is dynamically bound to the mesh vertices of the 3D tissue model to ensure consistency between macroscopic tissue deformation and microscopic atomic motion. The final output displacement data contains the spatial coordinate changes of each atom over a time series and can be directly input into a multiscale coupling algorithm to solve the decay diffusion equation.

[0038] It should be noted that the atomic-level displacement data of the corruption process refers to the atomic position movement trajectory data of protein molecules and corruption metabolites that change with time during the corruption process.

[0039] The decay rate of the remains and the atomic displacement data of the decay process are coupled through a multi-scale coupling algorithm to establish a decay diffusion equation and generate a decay process prediction matrix including time and space dimensions. The expression is: ; in, It includes space-time dimensions The corruption process prediction matrix, is the horizontal coordinate of the remains, is the vertical coordinate in the three-dimensional coordinate system of the remains, is the depth coordinate in the three-dimensional space coordinate system of the remains, is the time variable of the corruption process, is the initial moment The corruption state benchmark matrix (including the initial tissue corruption degree distribution data of the remains), It is the initial moment of the corruption process. is the multiscale diffusion coupling coefficient (range is (0, 1]), is the three-dimensional Laplacian operator, is the environmental control coefficient (value range is [0.1, 10]), is the corruption reaction rate term; Furthermore, the multi-scale coupling algorithm first discretizes the three-dimensional biological tissue model into spatial grid units, each unit The initial state is determined by the corruption state benchmark matrix Assignment. In the time step iteration, the atomic displacement data The material transport flux converted to grid scale through spatial averaging processing and the corruption rate predicted by the LSTM network together constitute the source term . Three-dimensional Laplacian operator The central difference format is used to calculate the diffusion effect of corrupt substances among organizations. According to the real-time monitoring temperature ,humidity and pressure Dynamic adjustment to control the weight of environmental factors on the reaction rate. The increment of each time step is integrated by explicit Euler method to update the corruption process prediction matrix The final generated corruption process prediction matrix fully records the arrive The evolution of the degree of decay of the body at each point on the 120×80×60 spatial grid at the moment. The value range of the matrix elements is [0, 1], corresponding to the continuous state of decay from initial to complete decay.

[0040] It should be noted that the corruption reaction rate term The physical meaning is: at a given temperature (Unit: ℃), humidity (Unit: %RH) and pressure (unit: kPa) environmental parameter constraints, combined with the atomic displacement data output by molecular dynamics simulation (Unit: A / ps), quantifies the generation rate of corrupted substances per unit time through a multi-scale coupling algorithm.

[0041] The substance concentration values in the corruption process prediction matrix are quantified by maximum-minimum normalization to generate the degree of corruption of the remains; Furthermore, the maximum-minimum normalization process first scans the concentration values of substances at all time and space points in the decay process prediction matrix to identify the global maximum and minimum values. The minimum value is subtracted from the value at each location in the decay process prediction matrix and then divided by the difference between the maximum and minimum values. All values are linearly converted to a standard range of 0 to 1. The converted value directly represents the degree of decay of the remains, with 0 representing the initial state of no decay and 1 representing complete decay.

[0042] The corruption process prediction matrix is mapped into a dynamic corruption heat map through Hinton's method based on Grad-CAM. Furthermore, the Hinton method based on Grad-CAM first calculates the corruption process prediction matrix The gradient of the final corruption degree prediction result is obtained at each spatial position In time Gradient importance weight on . Gradient importance weight and corruption process prediction matrix The feature map of is weighted summed to generate a spatial attention map. The attention map is combined with the corruption process prediction matrix Element-by-element multiplication highlights key areas of decay. Finally, the result is upsampled to the same resolution as the original 3D tissue model using bilinear interpolation. An OpenGL shader maps the values onto a red-yellow-blue gradient, generating a dynamic decay heatmap. Red areas in the heatmap represent the highest levels of decay, while blue areas indicate the lowest.

[0043] Using multi-dimensional group structured coding, the time and space dimensions , the degree of decay of the remains, and the RGB values in the dynamic decay heat map are encoded as a four-dimensional tensor of the decay process that can be traced back in time and space.

[0044] Furthermore, the multi-dimensional group structured coding first establishes a four-dimensional data cube and transforms the corruption process prediction matrix The spatial coordinates of and time variables As the basic index dimension. At the data structure level, the degree of decay of the remains is taken as the fifth dimension attribute value, and the RGB color value corresponding to each time and space point in the dynamic decay heat map is taken as the sixth to eighth dimension attribute value. The tensor splicing operation is used to combine the time and space dimensions. Orthogonally combine with the (corruption degree, R, G, B) attribute dimensions to construct a structured tensor with 8 dimensions. The original physical meaning of each dimension is strictly maintained during the encoding process. Represents the horizontal coordinate of the remains, Represents the vertical coordinate of the remains, Represents the depth coordinate of the remains, The corruption process time variable is represented by the corruption degree, the corruption degree is derived from the normalization result, and the RGB color value is directly taken from the dynamic corruption heat map. The resulting four-dimensional tensor of the corruption process is combined with the attribute dimensions through dimensionality compression technology to form a traceable data structure containing time and space coordinates and multi-attributes. It should be noted that the four dimensions of the corruption process four-dimensional tensor are: spatial grid coordinates , the time variable of the corruption process , degree of decay of the remains, RGB value.

[0045] The parameter conversion module constructs a four-dimensional space-time manifold model and converts the four-dimensional tensor of the corruption process into a set of space-time manifold parameters through a multi-scale space-time mapping algorithm, ultimately generating a comprehensive configuration parameter set of the four-dimensional space-time manifold; High-throughput gene sequencing was used to collect DNA sequence data of relatives, and TDA was used to extract persistent homology features of gene sequences; Furthermore, after whole-genome sequencing of the relatives' DNA samples using high-throughput gene sequencing, TDA first converted the gene sequence data into a point cloud representation, with each point representing a specific base position and its adjacent sequence features. Using the Vietoris-Rips complex construction method, the connection radius ε was gradually increased to record the simplicial complex structures formed at different scales. By calculating persistent homology groups, the birth-death intervals of 0-dimensional and 1-dimensional topological features were obtained, where 0-dimensional features reflect genomic connectivity components and 1-dimensional features represent loop structures. The extracted persistent homology features include the persistence length, occurrence scale, and spatial distribution density of each topological feature.

[0046] It should be noted that the relatives' DNA sequence data were obtained by whole genome sequencing of the relatives' DNA samples using high-throughput gene sequencing methods.

[0047] The DNA topological energy tensor is generated by performing tensor field transformation on the persistent homology features through TDA; Furthermore, the TDA process feeds the persistent homology features into a tensor transformation phase. First, a higher-order tensor space is constructed based on the topological structure of the Vietoris-Rips complex. By calculating the structural change rates of the persistent homology groups in each dimension as the complex evolves, a differential form describing the dynamic evolution of the topological features is obtained. The Hodge-Star operator is used to transform the differential form into a tensor representation in the dual space while maintaining the homological invariance of the Vietoris-Rips complex. During the tensor field transformation, the birth-death intervals of the persistent homology features are mapped to eigenvalues of the curvature tensor, generating a DNA topological energy tensor with clear geometric meaning.

[0048] Based on the four-dimensional tensor of the corruption process and the DNA topological energy tensor, the initial space-time manifold is constructed through Riemannian manifold mapping; Furthermore, the four-dimensional tensor of the corruption process and the DNA topological energy tensor are mapped onto a manifold using the Nash embedding theorem. First, an isometric embedding is established in Hilbert space. Using the curvature constraint provided by the DNA topological energy tensor, a hyperbolic metric structure is defined in the tangent bundle space. The Levi-Civita connection coefficient is calculated using a parallel transport algorithm, establishing a covariant differential relationship between the four-dimensional tensor of the corruption process and the DNA constraint. During the Riemannian manifold construction, the spacetime gradient field of the four-dimensional tensor of the corruption process provides the principal curvature distribution, while the DNA topological energy tensor provides the normal curvature constraint, jointly generating an initial spacetime manifold that satisfies the Gauss-Codazzi equations.

[0049] On the initial space-time manifold, the curvature of the manifold is corrected by conformal geometric flow to generate a four-dimensional space-time manifold model; Furthermore, the conformal geometric flow implements curvature correction on the initial spacetime manifold via the Yamabe flow. First, the Ricci curvature tensor of the initial spacetime manifold is calculated to identify local areas of curvature anomalies. Based on the principle of conformal transformation, the scaling coefficients at each point on the manifold are gradually adjusted by solving the partial differential equation for the conformal factor φ. During the iteration process, the material distribution provided by the four-dimensional tensor of the corruption process constrains the evolution direction of the conformal factor, while the curvature constraint provided by the DNA topological energy tensor controls the transformation step size. The scalar curvature is recalculated after each iteration until the curvature distribution at each point on the manifold reaches a predetermined uniformity threshold, ultimately obtaining a four-dimensional spacetime manifold model.

[0050] It should be noted that the uniformity threshold is set based on the standard deviation of the Ricci curvature of the initial space-time manifold. For example, when the curvature fluctuation range is less than 0.05, it is determined that the uniform state is reached.

[0051] The four-dimensional tensor of the corruption process is decoupled by using the time-space decoupling algorithm, and the macroscopic distribution parameters are generated through spatial gridding. At the same time, corrections are made based on perturbation theory to generate the corrected macroscopic distribution parameters. The spatiotemporal decoupling algorithm first calculates the four-dimensional tensor of the corruption process Perform modal decomposition to separate the space-time dimensions and time dimensions. Spatial gridding projects the three-dimensional spatial components onto a grid that matches the three-dimensional biological tissue model, generating initial values for the macroscopic distribution parameters. Perturbation theory analyzes the sensitivity of these parameters to initial conditions and makes local adjustments to key grid points. The DNA topological energy tensor acts as a canonical constraint in the correction process, ultimately generating the corrected macroscopic distribution parameters.

[0052] Through the multi-scale coupling algorithm, the corrected macroscopic distribution parameters are fused with the time evolution sequence to generate a set of space-time manifold parameters; Furthermore, the time-space decoupling algorithm first performs Tucker decomposition on the four-dimensional tensor of the corruption process, and converts the time-space dimension into The decomposed three-dimensional spatial factor matrix is separated into a spatial factor matrix and a temporal factor vector. Spatial gridding is then performed to resample the decomposed three-dimensional spatial factor matrix to a grid resolution consistent with the three-dimensional biological tissue model, generating an initial estimate of the macroscopic distribution parameters. Perturbation theory constructs a Lyapunov exponent matrix to assess parameter sensitivity and identify key grid points for local correction. During the correction process, the DNA constraints provided by the DNA topological energy tensor serve as a regularization term. The final output of the corrected macroscopic distribution parameters retains the spatial distribution of the original four-dimensional tensor of the corruption process, but eliminates the coupling effect of the temporal dimension, making them directly applicable to the generation of the space-time manifold parameter set.

[0053] The space-time manifold parameter set is input into the four-dimensional space-time manifold model for dynamic evolution analysis to generate a four-dimensional space-time manifold comprehensive configuration parameter set, which is expressed as: ; in, is the gradient of the four-dimensional tensor space of the corruption process mean (characterizing the curvature of space), is the spatial gradient of the four-dimensional tensor of the corruption process, is the total number of grids in the body point cloud data generated by terahertz scanning (determined by 5A resolution tomography), is the time derivative of the degree of decay of the remains The dot product with the manifold normal n (characterizing the time evolution), is the time derivative of the corruption diffusion equation, is the normal vector of the isosurface in the protein distribution thermodynamic map, It is the persistent homology characteristic of DNA extracted by TDA. is the DNA topological energy tensor, is the DNA constraint parameter (characterizing DNA constraint), is the environmental response parameter (characterizing the environmental response), Atomic displacement data The molecular dynamics simulation calibration coefficients were determined, It is a comprehensive configuration parameter set of four-dimensional space-time manifold.

[0054] Furthermore, the curvature of space The calculation first calculates the central difference of the four-dimensional tensor of the corruption process in the spatial dimension to obtain the spatial gradient of each grid point. Then calculate the gradient norm on all grid points The arithmetic mean of the total number of grids The calculation process fully preserves the spatial distribution characteristics of the four-dimensional tensor of the corruption process, ensuring The value accurately reflects the curvature change of the three-dimensional biological tissue model of the remains in the spatial dimension. The final result is directly used as the comprehensive configuration parameter set of the four-dimensional space-time manifold A quantitative indicator of spatial curvature in .

[0055] Time evolution parameters The calculation strictly follows the expression First, the backward difference method is used to calculate the four-dimensional tensor of the corruption process in the time dimension. Time derivative of the degree of decay of the remains , obtain the corruption degree change rate at each time and space point. At the same time, extract the isosurface normal vector from the protein distribution heat map , through the exact dot product operation · Project the time rate of change to the normal direction of the manifold. This calculation process fully preserves the time derivative of the degree of decay of the remains. Dynamic characteristics and normal vectors of The spatial structural characteristics ensure The value accurately characterizes the evolution law of the corruption process on the space-time manifold.

[0056] DNA constraint parameters The calculation process strictly maintains the same expression First, calculate the DNA persistent homology features extracted by TDA and DNA topological energy tensor The inner product of , and then normalize the result by dividing by Get the normalized value. The ReLU function is used to ensure that only the positive constraint effect is retained, that is, This computational process fully preserves the persistent homology characteristics of DNA Topological properties of DNA and the DNA topological energy tensor The geometric characteristics of the J value ensure that the J value accurately quantifies the binding strength of genetic material on the decay process.

[0057] Environmental response parameters The calculation is strictly based on the expression First, calculate the DNA topological energy tensor Norm of , which is then divided by the atomic displacement data Determined molecular dynamics simulation calibration coefficients The ratio result is processed by the hyperbolic tangent function tanh to ensure that the output value is within the range (-1, 1). This calculation process fully preserves the DNA topological energy tensor. Geometric characteristics and environmental calibration coefficients The physical meaning of The value accurately represents the dynamic response characteristics of environmental factors to the corruption process.

[0058] The solution generation module uses a quantum-inspired tensor network engine to perform multi-path parallel deduction based on a comprehensive configuration parameter set of a four-dimensional space-time manifold, and generates the optimal funeral service process solution through Monte Carlo simulation and variational quantum eigensolver optimization; Through the tensor coding mapping algorithm, the spatial curvature, time evolution, DNA constraint and environmental response of the four-dimensional space-time manifold comprehensive configuration parameter set are encoded as tensor nodes of the quantum-inspired tensor network engine, where Converted to a three-dimensional space grid tensor, Mapped into a time series matrix, and as edge weight and capacity constraint parameters respectively; Furthermore, the tensor coding mapping algorithm first transforms the spatial curvature parameters in the four-dimensional space-time manifold comprehensive configuration parameter set Θ into Converted into a three-dimensional space grid tensor, the space grid of the three-dimensional biological tissue model of the remains is The values are directly mapped to the eigenvalues of the tensor nodes, maintaining the original spatial resolution and topological connection relationship. It is encoded as a time series matrix, where the rows of the matrix correspond to different time steps, the columns correspond to the spatial grid point index, and the element values are the The DNA constraint parameter J is used as the edge weight parameter by calculating the relationship between adjacent spatial grid tensor nodes. The difference in values generates the weight coefficient of the connecting edge, which reflects the gradient change of DNA constraint in space. As a capacity constraint parameter, it is assigned to each spatial grid tensor node to limit the maximum information flow of the node and characterize the regulatory intensity of environmental factors on the local corruption process.

[0059] Based on tensor nodes, the quantum-inspired tensor network engine simultaneously generates and deduces multiple types of funeral paths, performs sampling analysis on each type of path through Monte Carlo simulation, and generates multi-dimensional decision data packets; Furthermore, the quantum-inspired tensor network engine uses Tucker to construct a deduction framework for multiple funeral paths based on the encoded tensor nodes (including three-dimensional spatial grid tensors, time series matrices, edge weight parameters and capacity constraint parameters), and calculates the spatial-temporal correlation of each path through tensor contraction operations; the Monte Carlo simulation uses the Metropolis-Hastings algorithm to randomly sample the path space, and each sampling evaluates the DNA constraint satisfaction and environmental response adaptability of the path, generating a multidimensional decision data package including path feasibility score, spatiotemporal coordination coefficient and resource consumption.

[0060] The multidimensional decision data package is optimized into the Hamiltonian of the funeral path through the variational quantum eigensolver, and the optimal funeral service process plan is generated through the quantum approximate optimization algorithm.

[0061] Furthermore, the component quantum eigensolver maps the parameters in the multidimensional decision data package into a combination of Pauli operators of the Hamiltonian, and constructs an optimization objective function that includes the spatial curvature cost term of the funeral path, the time evolution efficiency term, and the DNA constraint satisfaction term; the quantum approximate optimization algorithm iteratively adjusts the ground state approximate solution of the Hamiltonian through parameterized quantum circuits, and outputs the optimal funeral service process plan when it converges to the quantum convergence threshold. For example, in a high temperature environment (environmental response parameter S>0.8), rapid cremation is preferred over traditional burial plans.

[0062] It should be noted that the quantum convergence threshold is defined based on the quantum state fidelity and the energy gradient change rate, and its value range is the normalized interval of [0.85, 0.95].

[0063] The execution module decouples the optimal funeral service process plan into real-time control instructions and executes them. At the same time, it makes corrections based on real-time body corruption data and finally generates a quality report for the entire funeral service chain.

[0064] Using the Poisson spatiotemporal gridding algorithm, the optimal funeral service process solution is parsed into spatiotemporal control nodes containing coordinates, timestamps, and operation instructions. The IoT instruction compilation engine then converts the spatiotemporal control nodes into real-time control instruction sets that can be directly executed by the Cloud Brain platform. Furthermore, the Poisson space-time gridding algorithm first decomposes the space-time elements in the optimal funeral service process plan into discrete event points, each of which contains three-dimensional coordinates, time dimensions, and time domains. Delaunay triangulation is used to build non-overlapping simplex grids on the four-dimensional space-time manifold to ensure that the time derivative of the degree of decay of the remains in each grid cell is and the spatial gradient of the four-dimensional tensor of the corruption process The IoT instruction compilation engine parses the mesh vertex data, converts the coordinates into GPS positioning codes, converts the timestamps into UNIX time codes, and converts the operation instructions into control statements in JSON format. It is encapsulated into a lightweight data packet through the MQTT protocol and generates a real-time control instruction set containing the device address, execution time, and control parameters. During the execution of the control instruction set, the real-time control instructions are corrected through the PID control algorithm based on the real-time collected body decay data, and a quality report for the entire funeral service chain is generated.

[0065] The body decay data includes the distribution of tissue decay degree, protein degradation status, and environmental parameter changes. Furthermore, during the execution of the real-time control instruction set, the body decay process data is continuously collected through the terahertz wave scanning array to obtain the measured value of the decay process four-dimensional tensor at the current moment. The PID control algorithm calculates the deviation between the measured decay process four-dimensional tensor and the expected value of the optimal funeral service process plan, and respectively obtains the spatial curvature parameters The error proportional term and time evolution parameter The integral term of the DNA constraint parameter J and the differential term of the DNA constraint parameter J. The three terms are weighted and summed to generate the correction coefficient, which dynamically adjusts the operating parameters of each spatiotemporal control node in the real-time control instruction set output by the IoT instruction compilation engine. The corrected control instructions maintain the original coordinates and timestamps. Under the premise of no change, the execution intensity and duration of the operation instructions are updated. At the same time, based on the PID correction records of the complete execution cycle, a summary is generated to generate a full-chain quality report on funeral services, including indicators such as corruption control accuracy, space-time coordinate deviation, and resource consumption efficiency.

[0066] The quality report of the entire funeral service chain includes corruption control deviation ΔM, equipment execution compliance rate, environmental parameter stability and DNA constraint verification.

[0067] The corruption control deviation ΔM in the funeral service full-chain quality report is obtained by calculating the root mean square error between the four-dimensional tensor of the corruption process collected in real time and the expected value of the optimal funeral service process plan, which characterizes the degree of deviation between the actual corruption process and the theoretical prediction. The equipment execution compliance rate statistics the punctuality and parameter matching of the execution instructions of each spatiotemporal control node, reflecting the execution reliability of the IoT device. The stability of environmental parameters is based on the temperature, humidity and pressure data collected by the platinum resistance temperature sensor, the capacitive polymer film humidity sensor and the piezoresistive air pressure sensor, and calculates its coefficient of variation throughout the service cycle. DNA constraint verification is achieved by comparing the persistent homology features of DNA extracted by TDA in the actual corruption process. and DNA topological energy tensor The inner product change rate is used to evaluate the degree of satisfaction of the genetic constraints. All indicator data are derived from dielectric constant distribution data collected by the terahertz wave scanning array, 5A resolution tomographic point cloud data, and environmental sensing parameters.

[0068] In summary, the present invention utilizes a multiscale coupling algorithm to couple the decay rate predictions from an LSTM network with atomic-level displacement data from AMBER force field simulations, enabling the decay process prediction matrix to accurately reflect the spatiotemporal correlations between microscopic reactions such as proteolysis and tissue decay. A quantum-inspired tensor network engine encodes the parameters of a four-dimensional spacetime manifold as tensor nodes, utilizes Tucker decomposition to construct a funeral path deduction framework, and combines Monte Carlo sampling with variational quantum optimization to achieve multi-objective decision-making under the dual conditions of DNA constraints and environmental responses.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A cloud-brain platform for intelligent control of the entire funeral service chain based on digital twin scenarios, characterized by: include, The data acquisition module acquires molecular-level point cloud data and environmental parameters of the remains, constructs a 3D biological tissue model using PointNet++, and constructs a protein distribution heat map based on the dielectric constant distribution data, ultimately generating an initial dataset. The corruption prediction module uses the LSTM-molecular dynamics joint prediction method based on the initial dataset to establish the corruption diffusion equation, generate a corruption process prediction matrix including time and space dimensions, generate a dynamic corruption heat map through Hinton, and encode it into a four-dimensional tensor of the corruption process; The parameter conversion module constructs a four-dimensional space-time manifold model and converts the four-dimensional tensor of the corruption process into a set of space-time manifold parameters through a multi-scale space-time mapping algorithm, ultimately generating a comprehensive configuration parameter set of the four-dimensional space-time manifold; The solution generation module uses a quantum-inspired tensor network engine to perform multi-path parallel deduction based on a comprehensive configuration parameter set of a four-dimensional space-time manifold, and generates the optimal funeral service process solution through Monte Carlo simulation and variational quantum eigensolver optimization; The execution module decouples the optimal funeral service process plan into real-time control instructions and executes them. At the same time, it makes corrections based on real-time body corruption data and finally generates a quality report for the entire funeral service chain.

2. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The use of PointNet++ to construct a three-dimensional biological tissue model refers to performing sampling and grouping operations through PointNet++ to gradually capture point cloud features from local to global, and performing interpolation upsampling and cross-layer feature splicing through FP to generate dense feature point cloud data. At the same time, a dynamic graph convolutional neural network is used to perform point cloud semantic segmentation, and a Poisson surface reconstruction algorithm is combined to generate a three-dimensional biological tissue model.

3. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform as claimed in claim 2 is characterized by: The protein distribution heat map is constructed based on the dielectric constant distribution data, and the initial data set is finally generated. The steps are as follows: Obtaining the dielectric constant distribution data of the deceased tissue, and converting the dielectric constant distribution data into protein concentration distribution data through the dielectric relaxation spectrum quantitative analysis method; The Kriging interpolation algorithm was used to spatially complete the protein concentration distribution data. The color mapping function of the OpenGL shader was used to map the spatially completed protein concentration distribution data to an RGB gradient color spectrum to generate a protein distribution heat map. Molecular-level point cloud data, environmental parameters, three-dimensional biological tissue models, and protein distribution heat maps are integrated to generate an initial dataset.

4. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The LSTM-molecular dynamics joint prediction method is used to establish the corruption diffusion equation and generate a corruption process prediction matrix containing time and space dimensions. The steps are as follows: Predict the decay rate of corpses through a three-layer LSTM network; Based on the grid topology of the 3D biological tissue model, the AMBER force field is used to simulate the chemical reactions of the remains and obtain atomic-level displacement data of the decay process. The decay rate of the remains and the atomic-level displacement data of the decay process are coupled through a multi-scale coupling algorithm to establish a decay diffusion equation and generate a decay process prediction matrix including time and space dimensions.

5. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform as claimed in claim 4 is characterized by: The dynamic corruption heat map is generated by Hinton and encoded into a four-dimensional tensor of the corruption process. The steps are as follows: Quantify the substance concentration values in the corruption process prediction matrix to generate the degree of corruption of the remains; The corruption process prediction matrix is mapped into a dynamic corruption heat map through Hinton's method based on Grad-CAM. Using multi-dimensional group structured coding, the time and space dimensions , the degree of decay of the remains, and the RGB values in the dynamic decay heat map are encoded as a four-dimensional tensor of the decay process that can be traced back in time and space.

6. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The construction of the four-dimensional space-time manifold model refers to extracting the persistent homology features of the gene sequence through TDA, and the steps are as follows: Extract persistent homology features of gene sequences through TDA; The DNA topological energy tensor is generated by performing tensor field transformation on the persistent homology features through TDA; Based on the four-dimensional tensor of the corruption process and the DNA topological energy tensor, the initial space-time manifold is constructed through Riemannian manifold mapping; On the initial space-time manifold, the curvature of the manifold is corrected by conformal geometric flow to generate a four-dimensional space-time manifold model.

7. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 6 is characterized by: The four-dimensional tensor of the corruption process is converted into a set of space-time manifold parameters through a multi-scale space-time mapping algorithm, and finally a four-dimensional space-time manifold comprehensive configuration parameter set is generated. The steps are as follows: The four-dimensional tensor of the corruption process is decoupled using a spatiotemporal decoupling algorithm, and macroscopic distribution parameters are generated through spatial gridding. At the same time, corrections are made based on perturbation theory. The corrected macroscopic distribution parameters are then fused with the time evolution sequence through a multi-scale coupling algorithm to generate a set of spatiotemporal manifold parameters. The space-time manifold parameter set is input into the four-dimensional space-time manifold model for dynamic evolution analysis to generate a four-dimensional space-time manifold comprehensive configuration parameter set.

8. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The quantum-inspired tensor network engine is used to perform multi-path parallel deduction, and the optimal funeral service process plan is generated through Monte Carlo simulation and variational quantum eigensolver optimization. The steps are as follows: Through the tensor encoding mapping algorithm, the comprehensive configuration parameter set of the four-dimensional space-time manifold is encoded into the tensor nodes of the quantum-inspired tensor network engine; Based on tensor nodes, the quantum-inspired tensor network engine simultaneously generates and deduces multiple types of funeral paths, performs sampling analysis on each type of path through Monte Carlo simulation, and generates multi-dimensional decision data packets; The multidimensional decision data package is optimized into the Hamiltonian of the funeral path through the variational quantum eigensolver, and the optimal funeral service process plan is generated through the quantum approximate optimization algorithm.

9. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The optimal funeral service process plan is decoupled into real-time control instructions and executed, and corrections are made based on real-time body corruption data to ultimately generate a full-chain quality report on funeral services. The steps are as follows: Using the Poisson spatiotemporal gridding algorithm, the optimal funeral service process solution is parsed into spatiotemporal control nodes containing coordinates, timestamps, and operation instructions. The IoT instruction compilation engine then converts the spatiotemporal control nodes into real-time control instruction sets that can be directly executed by the Cloud Brain platform. During the execution of the control instruction set, the real-time control instructions are corrected through the PID control algorithm based on the real-time collected body decay data, and a quality report for the entire funeral service chain is generated.

10. The digital twin scenario-based funeral service full-chain intelligent control cloud brain platform according to claim 1 is characterized by: The molecular-level point cloud data of the remains include tissue dielectric constant distribution point cloud and 5A resolution tomography point cloud data; The environmental parameters include real-time monitored data of temperature, humidity and atmospheric pressure of the body storage environment.

Images