116results about How to "Implement features" patented technology

The invention discloses an unmanned aerial vehicle routing inspection line adaptive obstacle detection method and system based on a monocular camera. The method comprises the steps of obtaining initial flight parameters, evaluating and optimizing the flight parameters based on scene complexity, collecting an original image sequence and carrying out quality evaluation and exposure compensation. Constructing a multi-resolution feature pyramid, extracting feature descriptors, and performing feature matching and verification; performing multi-constraint pose optimization and depth estimation based on a feature matching result, and reconstructing a three-dimensional point cloud; calculating direction enhancement response to extract a power line candidate area, performing multi-view line segment matching and three-dimensional reconstruction, and analyzing dynamic characteristics of the power line; and based on the power line three-dimensional model and the depth map, carrying out obstacle segmentation, calculating an obstacle distance, evaluating a risk level, and generating early warning information and a monitoring scheme. According to the invention, different scenes can be adaptively processed, the obstacle risk can be accurately identified and evaluated, and the inspection efficiency and reliability are improved.

The invention discloses a large container crane intelligent early warning system based on personnel behavior analysis, and the system comprises the following steps: collecting operator behaviors, equipment operation states and environment parameters in real time through a multi-mode sensor network, constructing a dynamic data set, extracting behavior characteristics through a time difference embedding algorithm, and carrying out the early warning of the operator behaviors, the equipment operation states and the environment parameters; a multi-level spatial-temporal characteristic matrix is generated in combination with dynamic scene influence factors, regional risk factors are dynamically calculated by using a Bayesian inference model, comprehensive risk values are generated through an adaptive multi-objective optimization algorithm, risk levels are divided into low, medium and high classes, and a system triggers graded early warning responses according to the risk levels. And an optimized behavior adjustment suggestion is generated in combination with a quantum probability behavior prediction algorithm, and a feature extraction rule and risk assessment logic are dynamically optimized through a progressive learning algorithm to form a closed-loop optimization process. According to the invention, intelligent analysis and real-time early warning of personnel behaviors are realized, and the safety and adaptability of the crane in an operation environment are improved.

The invention discloses an intelligent remote follow-up visit method and system for tumor patients, and relates to the technical field of medical intellectualization, and the method comprises the steps: collecting follow-up visit data through a remote medical terminal, carrying out the processing of the follow-up visit data, obtaining a physical sign index sequence and a symptom feature vector, and generating a dynamic follow-up visit evaluation score through a time sequence prediction model; performing semantic analysis on the electronic medical record data of the patient based on the medical knowledge map, extracting tumor stage classification data, chemoradiotherapy measurement data, weight nutrition indexes and laboratory inspection data, determining a basic risk coefficient, a cumulative dose ratio and a body state score, and generating a baseline follow-up assessment score through a prediction model; and according to the dynamic follow-up evaluation score and the baseline follow-up evaluation score, calculating a comprehensive follow-up grade and generating a next follow-up plan. According to the invention, the accuracy of patient state evaluation is improved, accurate configuration of medical resources is realized, and the quality and efficiency of remote follow-up visit are effectively improved.

The invention relates to the technical field of reservoir storage capacity measurement, in particular to a reservoir storage capacity measurement method and system based on remote sensing and point cloud data, and the method comprises the following steps: extracting a terrain surface point set based on input remote sensing image data and point cloud data, and calculating a three-dimensional coordinate value, a curvature value and a point density value of each point; and comparing and classifying curvature and density data in combination with the change of the reservoir boundary. According to the method, through comprehensive application of remote sensing image data and point cloud data, three-dimensional coordinates, curvature values and point density values are adopted to accurately extract topographic features, scientific division and feature description of sub-regions are achieved, refined three-dimensional grid parameters are generated through fitting and interpolation calculation, and the accuracy of the three-dimensional grid parameters is improved. The terrain modeling precision and the space geometry description capability of a complex area are improved, boundary node coordinates and normal vectors are dynamically adjusted, the grid continuity is optimized, the integrity of a boundary transition area is ensured, and the boundary complex volume is accurately described by adopting unit volume subdivision and boundary correction.

The invention relates to a fault prediction method for a mining high-voltage power distribution device. Temperature, current, voltage and gas leakage sensors are installed in the mining high-voltage power distribution device; integrating the multi-dimensional data by using a multi-sensor data fusion algorithm to form a real-time operation portrait of the power distribution device; extracting time sequence characteristics of temperature and voltage key variables by combining an operation rule of the high-voltage power distribution device; capturing abnormal fluctuation of the key characteristic parameters by using a sequence analysis model, and identifying a specific fault type in the high-voltage environment; constructing a feature mode database based on a time sequence, and matching new data with a historical fault mode; applying a feature selection algorithm to extract representative fault prediction features from the sensor data; establishing a fault prediction risk level model according to the hazard degrees of different fault modes; based on the fault risk level predicted in real time, an optimized self-adaptive maintenance scheme is generated, and the maintenance period and strategy are dynamically adjusted in combination with the actual operation state of the equipment.

The invention provides a pathological section image analysis method and device based on a large language model, and belongs to the field of computer vision, deep learning and medical image analysis. The method comprises the following steps: constructing an image coding large model which is composed of a pre-training model and an additionally-added task head and aims at a pathological section; freezing the pre-training model, training a task head by using a preset training set, and checking by using a preset verification set to obtain a task head after preliminary training; unfreezing the pre-training model, performing fine tuning training on the large model by using the training set, and performing inspection by using the verification set so as to obtain a finally trained large model; and inputting a to-be-analyzed pathological section image into the large model, and calculating corresponding features of the image based on a result output by the large model to obtain a final analysis result. According to the method, the features in the pathological section image can be automatically extracted, and the diagnosis report is generated based on the features, so that the accuracy and efficiency of pathological section image diagnosis are improved.

The invention provides an interpretable fine-grained image recognition method, device and equipment and a medium, and relates to the technical field of fine-grained image recognition, and the method adopts a double-branch structure frame, and combines feature fusion and prototype learning to carry out fine-grained image classification. Wherein the feature extraction layer captures a complex mode and long-distance dependence in an image through a self-attention mechanism, meanwhile, a feature selection fusion module is introduced, information marks of prototype learning are effectively selected, and it is ensured that the learned prototype pays more attention to the prospect. The prototype learning layer designs a double-branch structure, processes and learns a global prototype and a local prototype of an image respectively, improves the accuracy of the network, and explains a model classification result from a global perspective. And the full connection layer realizes a final classification decision, and obtains a final classification probability through weighted sum. The objective of the invention is to solve the problem of black box characteristics of an existing deep learning model in fine-grained image classification and improve the application credibility of the model in the high-risk field.

The invention provides a pressure test control method and system for a universal pressure testing machine, and relates to the technical field of material testing machines, and the method comprises the steps: obtaining hydraulic cylinder parameters and servo valve parameters, building a pressure control mathematical model, calculating a transfer function, initializing a feedback controller, determining the number of hidden layer nodes, and setting an activation function and a network weight. A self-adaptive feedback controller is obtained; generating a closed-loop control system, constructing an objective function based on an integral absolute error and adjustment time, determining a Pareto optimal solution set in combination with a multi-objective genetic algorithm, selecting a neural network training sample, generating a training sample set, performing offline pre-training, and updating a network weight to obtain a pre-trained neural network; and the feedback controller is embedded into a pressure control system, outputs feedback controller parameters, sets an anti-saturation strategy of the feedback controller, judges whether the feedback controller is saturated or not, limits the amplitude of an integral item according to the relation between an error signal and a saturation direction, determines whether to stop integration or not, and generates a press control strategy under the current working condition.

The invention discloses a projection lens and a projection illumination system. The projection lens comprises a first lens, a second lens, a third lens and a fourth lens which are coaxially arranged in sequence from an image side to an object side, the diopter of the first lens is negative, the diopter of the second lens is negative, the diopter of the third lens is positive, and the diopter of the fourth lens is positive. By limiting the ratio of the focal length of an optical system to TTL * F.NO, the projection lens can be ensured to have the characteristic of large aperture, the size of the lens is effectively reduced, the ultrathin characteristic and miniaturization of projection are realized, the total length of the projection lens along the optical axis direction is reduced to be less than 23mm, the size of the projection lens is reduced, and the size of the projection lens is reduced. Therefore, the lamp is more suitable for the welcome lamp of the vehicle.

The invention discloses an instrument electric signal detection system, which comprises a signal acquisition module, a signal processing module, a data display module and a data storage library, and relates to the technical field of electric signal detection. According to the instrument electric signal detection system, the voltage waveform curve graph is established according to the converted digital signals, the change trend of the corresponding digital signals is determined according to the curve processing algorithm, a plurality of zero crossing points in the change process are searched, then curve segmentation operation is carried out with each zero crossing point as a node, and the segmented curves are covered. According to the method, the signal change period is determined, and the corresponding calculation period is compared with the set period of the output source, so that the electric signal detection efficiency is improved, meanwhile, the accuracy of a data result is verified, the detection operation can be quickly distinguished aiming at waveform change data, and actual application is facilitated.

The invention discloses a semantic segmentation method, a robot grabbing pose estimation method based on semantic segmentation, a computer program product and a terminal, and belongs to the field of semantic segmentation, and the method comprises the steps: constructing a semantic segmentation network, and enabling a decoder of the semantic segmentation network to comprise a lightweight feature fusion module which is used for carrying out multi-scale conversion processing on low-level features and high-level features, performing cross aggregation processing on different scale features, performing heterogeneous convolution processing, and finally performing feature fusion processing; training the semantic segmentation network; and inputting the to-be-segmented image into the trained semantic segmentation network to obtain a semantic segmentation result. Through the lightweight feature fusion module, the problem of feature fusion with inconsistent semantics and scales can be effectively solved, feature fusion and information interaction are better realized, pixel loss is reduced, and the quality of fused features is remarkably improved, so that the precision of semantic segmentation is improved.

The invention discloses an enterprise knowledge graph inference completion method and inference completion system, and the method comprises the steps: obtaining a to-be-completed knowledge graph of a target enterprise, inputting the to-be-completed knowledge graph into a trained knowledge graph embedding model, and predicting an incomplete triple to achieve the inference completion of the knowledge graph; wherein the knowledge graph embedding model comprises an entity node coding module, a multilayer graph convolutional neural network module, a Softmax pooling layer and a gating neural network based on a self-attention mechanism. According to the method, the knowledge graph embedding model can provide more accurate semantic information for feature interaction, multi-level feature interaction and aggregation of the triple on sub-tensor kernel granularity and global granularity can be realized, and hidden semantics behind entities and relationships of the knowledge graph can be more efficiently mined. Therefore, the knowledge graph completion accuracy, the semantic modeling capability, the interpretability and the like are obviously improved.

The invention relates to the technical field of tumor immunotherapy, in particular to a dual-targeting chimeric antigen receptor, a recombinant expression vector, an anti-tumor natural killer cell and application, and provides a PD1-DAP10-41BB-P2A-NKG2D-Ig4-CD28-41BB-CD3 zeta fusion receptor.The dual-targeting chimeric antigen receptor can target PD-L1 on the surface of a tumor cell, can also recognize NKG2DLs and can be used for preparing anti-tumor natural killer cells. The broad spectrum of the NK cells is greatly improved, and meanwhile, the immune escape of tumor cells is also reduced.

The invention discloses a rotating target detection method and system based on phase shift coding and directional feature enhancement, and belongs to the technical field of image recognition. The method comprises the steps of obtaining an image data set and performing enhancement processing; inputting into an ARC network to extract features of a target in any direction; performing adaptive feature optimization on the features through an EMA module, encoding optimized feature direction information by using ARF, generating direction sensitive features for bounding box regression, and extracting directional features for classification through a rotation invariant pooling layer; linearly mapping the angle of the rotating frame into a phase of a corresponding frequency by using a phase shift encoder, predicting a phase shift cosine value, and decoding to obtain a final phase; and finally, removing overlapped detection frames by applying a non-maximum suppression method, and reserving an optimal detection result. Wherein a Pareto optimization theory is used for adaptively weighting a loss item, model parameters are updated through a multi-gradient descent algorithm, and the obtained final model can improve the accuracy and efficiency of detection of a target containing a multi-directional small object.

The invention relates to a tongue feature analysis method and device, electronic equipment and a storage medium. A trained tongue detection model and a trained tongue segmentation model are fused to obtain a target segmentation model, an image data set is segmented through the target segmentation model to obtain a tongue segmentation image data set, and the tongue segmentation image data set is labeled to obtain a tongue feature classification data set. Training a tongue feature classification model based on the tongue feature classification data set to obtain a trained tongue feature classification model, obtaining to-be-detected target image data, inputting the target image data into a target segmentation model to obtain tongue segmentation image data of the to-be-detected target image data, and obtaining the tongue segmentation image data of the to-be-detected target image data. According to the method, the trained tongue feature classification model is used for performing feature analysis on the tongue segmentation image data to obtain a tongue feature analysis result, so that the training efficiency of the model is improved, and the classification stability and reliability are also remarkably improved.

The invention relates to an intelligent product recommendation method and system based on a large model. The method comprises the steps of obtaining user information of a user to be recommended and product information of a product to be recommended; extracting user semantic features, user behavior features and natural language understanding features of the user information and product description features and product category features of the product information by using the large model, and performing feature fusion to obtain fused features; and obtaining a target recommendation score of each product relative to the to-be-recommended user, and recommending the product corresponding to the target recommendation score higher than a preset score to the to-be-recommended user as a target product. According to the method, the Transform model, the GNN model and the LLM model are combined, analysis and feature fusion of the user information and the product information are achieved, the extraction precision of the user semantic features and the product description features is improved, the user behavior features and the product category features can be effectively captured, and therefore the recommendation accuracy and the personalized level are improved.

This application discloses a method and apparatus for constructing a multi-scale adaptive gating MambaPlus network, belonging to the field of artificial intelligence and machine learning technology. The method includes: initializing the network configuration and constructing the basic structure; preprocessing the input data to generate a standard dataset; mapping the input data to the hidden space via an input mapping layer, and extracting backbone features from the Mamba backbone; constructing at least two parallel scale branches in the hidden space to obtain a multi-scale feature set; inputting the backbone features and multi-scale features into an adaptive gating module, dynamically allocating weights and adaptively fusing them through a hierarchical gating mechanism to generate fused features; further enhancing the features through cross-scale attention and feedforward enhancement, and then superimposing the residuals to generate the final discriminative features; finally, completing category prediction and model training evaluation. This application, while retaining the advantages of Mamba's long-range dependency modeling, addresses the problems of insufficient utilization of multi-scale information, poor adaptive feature fusion, and low robustness in complex scenarios.

The invention discloses an electric vehicle charging load prediction method, system and device and a storage medium, and relates to the technical field of electric power, and the method comprises the steps: generating graph structure charging load data based on multi-station charging load data; based on the graph structure charging load data, executing a graph convolutional network-gating time convolutional network adopting an adaptive adjacency matrix, and generating hidden layer representation containing spatial adjacency information; the hidden layer representation and the external features are spliced and input to a time domain fusion Transform, and an electric vehicle charging load prediction result is output; according to the method, the spatial correlation extraction and time-dependent modeling of the charging load are realized by constructing the graph convolutional network-gating time convolutional network based on the adaptive adjacency matrix and the time domain fusion Transform model, and the precision and interpretability of multi-station load prediction are improved in combination with variable selection and an interpretable multi-head attention mechanism.

The invention provides a building electrical safety detection method and system, and relates to the technical field of electrical detection, and the method comprises the steps of first collection, second collection, first processing, second processing, third processing, model construction, model training, fault recognition and the like. The compressed sensing algorithm and the morphological filter are used, noise and interference signals in the data are effectively removed, the accuracy of fault detection of the electrical equipment in a complex electromagnetic environment in a high-rise building is improved, faults of the electrical equipment can be found in time, corresponding maintenance measures can be taken, and the fault detection efficiency is improved. And safe and stable operation of an electrical system is ensured.

The invention relates to the technical field of natural language processing, in particular to a social network user feature mining and matching method and system based on multi-source semantic fusion, and the method comprises the steps: constructing an undirected and unweighted graph according to the original data of a user; extracting target text data, preprocessing the target text data, and extracting a fragment text vector set; constructing prompt word vectors, splicing the prompt word vectors with the fragment text vectors, and extracting a user feature matrix; and based on the user feature matrixes of different social networks, extracting a user pair matching result according to the Euclidean distance. The purpose of the invention is to accurately and efficiently mine user features in different social network scenes and improve the accuracy of a social network user alignment task.

The invention discloses a broadband real-time digital radio frequency simulation method and device, and relates to the technical field of radio frequency simulation, and the method specifically comprises the following processing steps: enabling a radio frequency signal to be accessed to a power matching unit, converting the radio frequency signal into a range suitable for subsequent processing, enabling the radio frequency signal to be accessed to a frequency conversion unit A and a digital optical transmitter and receiver A in sequence, and achieving the remote access of the radio frequency signal; a real-time digital radio frequency simulation environment of N paths of input signals and M paths of output signals is constructed, N paths of intermediate frequency signals are respectively connected into M'path micro-strip power dividers through radio frequency cables, M radio frequency cables are connected behind each micro-strip power divider, and the signals sequentially pass through a circulator A, an amplifier A and a digital signal processing unit and then are connected into a circulator B; the intermediate frequency signal output by the combiner is converted into a radio frequency signal through a frequency conversion unit B and is output; according to the invention, access, exchange and characteristic simulation of large-scale and ultra-wideband radio frequency signals can be realized, modular combination can be carried out according to the number of access signals, and a real-time digital radio frequency simulation environment is constructed.

The present invention relates to the technical field of packaging product detection, and specifically to a quality detection method for packaging products based on the fusion of X-rays and visible light. In the present invention, X-ray images and visible light images are collected; feature engineering processing is performed on the X-ray images and visible light images to generate X-ray feature data and visible light feature data, wherein the data processed by feature engineering includes spatial and structural information; the X-ray feature data is used to detect the internal product quality using an internal quality detection model, and different output results are determined according to different internal product quality detection results; and the visible light feature data is enhanced according to the abnormal position to generate a visible light abnormal image, and the visible light abnormal image is used to detect the external product quality using an external quality detection model.

The present invention relates to the field of antenna technology, and provides a combined antenna applied to a roadside unit, including a transceiver device, a positioning antenna, a first antenna, and a second antenna. Compared with the prior art, the combined antenna applied to the roadside unit provided by the present invention can not only reduce costs and expand the bandwidth of the antenna frequency band by synthesizing the 2.4-2.483GHz frequency band and the 4.910-5.875GHz frequency band in the first antenna, thereby realizing the multi-frequency and broadband characteristics of the frequency band, but also avoid the problem in the prior art that the 2.4GHz frequency band antenna and the 5GHz frequency band antenna are arranged at the upper and lower ends of the transceiver device to affect the performance indicators of the antenna; by setting the frequency range of the second antenna to 698-3800MHz, the frequency range of the second antenna can cover the 5G frequency band, so that the application range of the combined antenna is wider.

The invention belongs to the technical field of time-frequency analysis, and particularly relates to a rolling bearing fault method and system based on Riesz atlas sparse decomposition, which utilizes a new time-frequency analysis method-Riesz four-parameter spectrum time-frequency analysis method to obtain four-parameter information of amplitude, frequency, phase and direction of a vibration signal in a time-frequency domain. The time-frequency aggregation and the anti-noise performance of an analysis result are improved; in order to screen out an optimal parameter spectrum and expand the symbol dynamics entropy to a two-dimensional scale, a new entropy calculation method, namely cosine symbol dynamics entropy, is provided; and reconstructing components of the parameter spectrum obtained by entropy screening, and constructing a sparse filter parameter optimization component amplitude, so that a decomposition result has physical significance. The bearing vibration signal under the variable speed working condition and the bearing vibration signal under the constant speed are analyzed, the RSSD method is provided on the basis, and feature extraction and diagnosis of rolling bearing faults under the constant rotating speed working condition and the variable rotating speed working condition are achieved.

This invention belongs to the field of nonlinear data classification technology and discloses a nonlinear data classification method based on deep transfer learning and visual recognition capabilities. This invention transforms high-dimensional numerical features into structured images, enabling convolutional neural networks to leverage their powerful visual recognition capabilities to handle tabular data classification tasks. This image-based representation naturally reveals the implicit nonlinear relationships within the data. By further integrating deep transfer learning, this method effectively alleviates the data sparsity problem in small-sample scenarios and improves generalization performance through a series of training optimization strategies. The framework proposed in this invention achieves an effective connection between the numerical feature space and the image-based deep learning architecture, providing theoretical innovation and practical value for complex nonlinear data classification tasks.

The present application relates to a kind of based on multimodal feature fusion distribution network disturbance source identification method, belong to the fine disturbance monitoring and diagnosis technical field of smart grid.The method includes: the disturbance class current traveling wave data extracted is carried out time-frequency conversion, generates traveling wave panorama waveform chart and two-dimensional time-frequency chart and is fused into three-channel time-frequency image, then image feature vector is extracted;The disturbance class current traveling wave data is carried out Prony modal parameter fitting, extracts modal parameter feature vector and is projected to high-dimensional feature space;The disturbance class current traveling wave data is carried out Clark transformation, calculates zero mode energy proportion feature and is projected to high-dimensional feature space;Image feature vector, projected physical feature vector and projected modulus energy feature vector are fused;Fusion feature vector is input into classifier, and the class identification result of distribution network disturbance source is output.The present application aims to solve the technical problems that the prior art has single feature representation, weak physical interpretability and difficulty in forming stable power fingerprint.

The present application relates to the technical field of computer vision, deep learning and graphics rendering, in particular to a 3D portrait reconstruction method and system based on cascade diffusion prior and UV space physical constraint, which adopts a parameter efficient adaptation paradigm of freezing a unified DiT backbone and a cascade LoRA adapter, and sequentially completes UV space texture completion, illumination homogenization and intrinsic material decomposition; introduces a cross-intrinsic attention mechanism to realize multi-material pixel-level spatial alignment, and executes differential physical rendering supervision to constrain material physical consistency in the UV domain. The present application realizes 100,000-level field image generalization under less than 100 3D scanning samples, and the generated 4K PBR material assets have no baking light and shadow residues, and the physical behavior is consistent under re-illumination, which is suitable for high-fidelity 3D portrait asset production in the fields of digital human creation, film special effects and virtual reality.

The invention relates to the technical field of medical rehabilitation instruments, and discloses an exoskeleton robot for fracture auxiliary rehabilitation. The robot comprises a shear deformation sensing assembly, a shape memory alloy locking assembly and a joint driving module. The shear deformation sensing assembly comprises a fixed shell, a flexible liner arranged on the inner side of the shell and a sensing beam connected between the shell and the liner. The shape memory alloy locking assembly comprises a locking pin arranged in the connecting rod structure, a shape memory alloy driving piece in linkage with the locking pin and a locking part located on the connecting rod structure and corresponding to the locking pin in position. And the strain gauge is electrically connected with the shape memory alloy driving piece. Transverse micro-slippage representing shearing risks is directly sensed through a mechanical structure, rigid mechanical locking is triggered in a linkage mode, and therefore rapid and reliable physical protection is provided for a fracture part, and secondary injury in rehabilitation training is effectively prevented.

The application provides a medicine cold chain logistics transportation whole-process tracking method and system, and relates to the technical field of logistics transportation monitoring. First, the original sensor tracking data stream output by the internal sensor cluster of a transportation carrier in a continuous collection time window is acquired. Then, the transportation carrier operation trajectory line is generated according to the geographic positioning parameters, and the multi-dimensional environment state marker set is generated by combining the physical parameters at the same time of each geographic position coordinate point. Meanwhile, the logistics transfer node topology connection network containing the logistics transfer node unit and the directed transfer path edge is constructed. The operation trajectory line, the multi-dimensional environment state marker set and the topology connection network are input into the pre-constructed space-time state feature fusion tracking model to generate the whole-process state tracking feature tensor. The abnormal state type identification and the space-time distribution feature are obtained by performing the abnormal state mode matching on the whole-process state tracking feature tensor, and the abnormal early warning instruction set is generated. The application can comprehensively and accurately track the whole process of medicine cold chain transportation, intelligently identify the abnormality and timely generate the early warning.

The application discloses a smart community police management system and method based on Internet of Things security, and belongs to the technical field of smart communities, and solves the problem that existing system data processing cannot utilize edge computing nodes for low-delay real-time analysis, and cannot meet timely early warning requirements, the method comprising preprocessing of multi-source heterogeneous community monitoring information, analysis and processing of the community monitoring information based on edge computing, and generation of a risk propagation path in combination with front-end analysis results and heterogeneity characteristics; in the application, analysis and processing of the community monitoring information based on edge computing avoids the delay of centralized processing, can quickly trigger early warning in the early stage of a risk event, thereby gaining golden disposal time for risk event disposal, and a risk analysis model realizes a leap from single-node anomaly detection to composite risk quantification through multi-dimensional feature fusion and dynamic space-time reasoning, and realizes multi-dimensional feature fusion and dynamic reasoning capability of the risk event.