1213 results about "Network processing" patented technology

A system and method for adaptive image and video processing using a Persistent Cognitive Machine (PCM) architecture with visual strategy caching. The system receives degraded input media and extracts degradation fingerprints to query a PCM-based visual strategy cache containing previously successful processing strategies. When matching cached strategies are found above a relevance threshold, they are retrieved and applied directly. When no match exists, the input is processed through transform-domain networks to generate new strategies. A pattern synthesizer combines multiple strategies for complex degradation types. The system evaluates processing effectiveness using a feedback controller and stores successful strategies in the hierarchical cache. This cognitive approach enables real-time processing with continuously improving performance as the cache learns from successful patterns. The adaptive architecture eliminates redundant processing while maintaining high-quality output, making it suitable for diverse imaging and video applications requiring efficient enhancement capabilities with superior performance over traditional methods.

The invention relates to the technical field of agricultural meteorological prediction, in particular to an agricultural meteorological disaster time sequence prediction system and method based on multi-modal data fusion, and the method comprises the steps: collecting and preprocessing agricultural meteorological disaster related data, constructing a dynamic semantic association graph, carrying out the multi-layer feature abstraction processing, and generating a semantic enhancement feature vector; the dual-branch prediction network processes time sequence dependence and local mode features, multi-granularity attention processing identifies key feature information, multi-scale feature fusion extracts different time scale feature information, and cascade fusion is carried out; the multi-target optimization module carries out model training based on the comprehensive feature representation and optimizes a plurality of targets; the multi-time scale prediction output module generates short-term accurate prediction, medium-term trend prediction and long-term risk assessment results, and provides prediction confidence, error range and risk level information; a dynamic semantic association graph and a multi-layer feature mapping mechanism are constructed, and deep fusion of multi-modal data on the semantic level is achieved.

The invention provides an intelligent customer risk assessment system and method based on a large language model, and relates to the technical field of risk assessment, and the method comprises the steps: obtaining multi-modal data of a customer, carrying out the preprocessing, and extracting structured and unstructured features; constructing a hierarchical risk knowledge system, and realizing adaptive evolution of the knowledge system through a graph neural network and a generative model; constructing an initial negative sample library, and constructing a negative sample database in combination with a non-risk mode labeled by an expert and derivative layer analysis; optimizing the large language model by adopting a strong supervision, weak supervision and reinforcement learning cooperative training mechanism under each classification according to the customer type; mining risk features in a text by using the optimized large language model, processing multi-modal data through a multi-level attention network, and generating a positioning report including contradiction type coding, service influence dimension evaluation and risk level quantification; the accuracy, efficiency and flexibility of customer risk assessment are improved, and the risk management strategy is optimized.

The invention relates to the technical field of artificial intelligence, can be applied to business scenes of financial science and technology, medical health and the like, and discloses a time series data processing method, device and equipment and a medium. Performing a data enhancement operation on the original time series data to generate enhanced time series data; training a feature encoder based on the enhanced time sequence data in a comparative learning mode to obtain a pre-trained feature encoder; connecting the pre-training feature encoder with a sparse attention mechanism module to construct a time sequence modeling network; target time series data is processed using the time series modeling network to generate a processing result. According to the method, the enhanced view is constructed on the unlabeled data and the contrast learning training feature encoder is introduced, so that the time sequence representation with generalization ability is obtained, effective modeling of the long dependency relationship is realized in combination with a sparse attention mechanism, and the accuracy of time sequence modeling is improved under the condition of not depending on a large amount of labeled data.

According to the multi-modal automatic knowledge graph construction method based on the large language model, a multi-modal data stream is preprocessed, features are extracted, and the multi-modal data stream is mapped to a unified semantic space through a cross-modal alignment network after being processed through the large language model, a visual converter and a time sequence neural network. In the space, entities and categories are recognized based on a large language model, a triple is generated by combining a multi-modal feature judgment entity relationship, mapping fusion is performed through an ontology alignment algorithm driven by a graph neural network and a predefined domain ontology, finally knowledge is stored in a graph database, and dynamic updating is performed by means of incremental learning and online reasoning. Standardized APIs and visualization components are provided. According to the method, the construction efficiency and the automation degree of the knowledge graph are remarkably improved, the cross-modal information fusion and knowledge maintenance capability is enhanced, and the application requirements of intelligent retrieval, recommendation, decision support and the like are met.

The invention relates to the technical field of medical related information and communication, and discloses a hospital logistics one-stop service inspection and repair platform. The hospital logistics one-stop inspection and repair method comprises the following steps: processing equipment operation data through a multi-modal data fusion network, and generating a state feature vector set; constructing an equipment association graph model by using the space-time diagram attention network in combination with the topological information; detecting the abnormity of a single device and predicting a propagation path through a hierarchical time sequence abnormity detection network, and generating a minimum intervention strategy; an abnormal root cause analysis report is generated by means of a multi-stage interpretability engine, and decision making is supported; routing inspection routes and repair task allocation are optimized based on reports and strategies; according to the invention, integrated management of the whole process from abnormity prediction, inspection planning, fault diagnosis to maintenance dispatching can be realized, the hospital logistics pain point is solved, and the system has significant technical innovation and practical value.

The invention discloses a temperature monitoring and control system in a goat breeding house. The system comprises an environmental data sensing unit, an edge data preprocessing unit, an optimized long-short term memory neural network processing unit, an air internal and external circulation regulation and control strategy generation unit, an equipment linkage execution unit and a system operation state monitoring unit. Environmental data in a breeding house is collected in a self-adaptive mode through multiple types of high-precision sensors, after edge processing, temperature spatial-temporal characteristics are extracted deeply through an optimized long-short-term memory neural network, an air circulation regulation and control strategy is generated in combination with a multi-target optimization model, equipment is driven to execute precisely, and system dynamic optimization is achieved through closed-loop feedback. The system overcomes the defects of insufficient monitoring coverage and extensive regulation and control in the prior art, considers the coupling relationship between temperature and air components, realizes multi-factor collaborative accurate regulation and control, effectively reduces energy consumption, creates a stable and comfortable growth environment for goats, and improves the breeding benefit and intelligent management level.

The invention discloses an electric power scene defect small target detection method based on Gaussian mask supervision and cross-layer attention guidance, and the method comprises the steps: inputting an electric power scene image into a detection model, extracting an initial feature map through a backbone network, carrying out the multi-stage feature extraction of the initial feature map according to a convolution path, and carrying out the multi-stage feature extraction of the initial feature map; processing the multi-stage features based on a path aggregation network, and outputting a plurality of fusion feature maps with different feature levels from shallow to deep; and based on cross-scale window attention, guiding a shallow fusion feature map to carry out semantic information modeling by using a deep fusion feature map with high semantics in every two adjacent fusion feature maps, and after a plurality of output feature maps are obtained, respectively processing and outputting prediction results by using a multi-branch detection head. According to the method, shallow feature activation prediction and cross-scale window attention guidance are fused, and the detection robustness and positioning precision of a tiny fault target in an unmanned aerial vehicle inspection image can be effectively improved.

The invention discloses a three-dimensional shielding target tracking method based on multi-modal space-time interaction, and relates to the technical field of target tracking. The method comprises the following steps: acquiring a point cloud and an image and preprocessing to obtain global fusion features; obtaining an initial detection frame and region-of-interest features through region proposal network processing; projecting the non-empty voxel point cloud to the image features, and reconstructing shielded target features; convolution and neural network processing are utilized to obtain a refined detection frame; screening legal detection frames through distance calculation and legality judgment; the bipartite graph and the self-adaptive channel graph are adopted for convolution, and appearance correlation scores are calculated; and matching the detection frame and the trajectory based on a Hungary algorithm to realize whole-course tracking. The target identification accuracy and robustness are improved, the shielding problem is solved, the accuracy of the detection frame is ensured, the correlation accuracy is improved by using the bipartite graph and the adaptive convolution, the nodes are matched in combination with the geometric cost matrix, and whole-course tracking and error calibration are realized.

The invention provides a gear geometric parameter rapid detection method and system based on wireless transmission, and relates to the technical field of gear detection, and the method comprises the steps: for a to-be-detected gear, extracting gear features based on a deep learning network, and determining a position combination of optimal measurement points in combination with a gear precision grade requirement; collecting geometric parameter data of the gear based on the position combination; and processing the geometric parameter data by adopting a cascade neural network to obtain actual geometric parameters, sending the actual geometric parameters to a remote monitoring terminal through a wireless transmission module with a self-adaptive channel selection function, and generating a detection result report.

The invention discloses a compressed sensing MRI (Magnetic Resonance Imaging) reconstruction method and system based on a double-domain fusion expansion model. The method comprises the following steps: acquiring an MRI image, preprocessing the MRI image to obtain a preprocessed image and a corresponding compressed sensing measurement value, and constructing a data set; training, testing and verifying the double-domain fusion expansion model by using the data set; and outputting a reconstructed image from the MRI image after mask sampling by using a double-domain fusion expansion model. According to the double-domain fusion expansion model, a complex convolutional neural network and a K-space attention mechanism are introduced for complex value data processing and double-domain information fusion in compressed sensing MRI reconstruction. The complex value data is processed through the complex network, the amplitude and phase information of the complex field is fully utilized, the detail recovery capability and the noise suppression effect of the model are enhanced, and particularly, excellent robustness is shown at a low sampling rate. The method is excellent in performance in complex MRI data reconstruction, and has high reconstruction precision and generalization ability.

The invention discloses a traffic flow prediction method and system based on feature fusion and frequency enhancement. The method comprises the following steps: acquiring traffic flow data of a traffic monitor and external meteorological data and carrying out preprocessing; a dynamic space-time diagram generation module constructs a dynamic space-time diagram fusing a geographic neighbor diagram and a function similar diagram based on the preprocessed traffic flow data; inputting the dynamic space-time diagram into a space-time attention convolution layer to obtain a space-time matrix; fusing the space-time matrix with the external meteorological data subjected to one or more times of long-short-term memory network processing, and executing near-time slicing processing on the fused data; and constructing a hierarchical encoder-decoder model based on Transform, and generating traffic flow prediction data after the data subjected to slicing processing passes through the model. According to the method, the traffic flow prediction precision can be improved, and a more reliable prediction result is provided.

The invention discloses a chlorophyll calculation method based on a double-branch multi-scale fusion network model, which belongs to the technical field of chlorophyll calculation, is used for calculating a chlorophyll a concentration value, and comprises the following steps: taking the chlorophyll a concentration value of a measurement site as a label of a water body characteristic sequence data set, dividing the data set into a training set and a verification set, inputting the training set into the double-branch multi-scale fusion network model to obtain a trained double-branch multi-scale fusion network model; inputting the verification set into the double-branch multi-scale fusion network model to obtain a verified double-branch multi-scale fusion network model; and taking the neighborhood spectral feature tensor and the neighborhood inherent optical feature tensor as the input of the verified double-branch multi-scale fusion network model to obtain the chlorophyll a concentration value of the pixel at the corresponding position. According to the method, the calculation result of the chlorophyll a with relatively high precision is obtained through neural network processing, and the complex relationship among the concentration change of the chlorophyll a in the water body, the spectrum and other optical parameters can be captured.

The invention provides a falling detection method and system based on millimeter wave radar fusion human body postures, and relates to the technical field of falling detection, and the method comprises the steps: collecting human body echo signals, and constructing three-dimensional point cloud data through distance, angle and speed estimation; the point cloud is processed to generate a dynamic sequence, and skeleton features and key point coordinates are extracted in combination with a graph convolutional network. A skeleton connection relation is constructed based on the key points, attitude features are calculated, key change features are extracted through a self-attention mechanism, and the key change features are fused with a point cloud sequence to construct multi-modal features. And the double-branch state recognition network processes the fusion features, and when abnormality is detected, further analysis is carried out through residual attention and space-time diagram convolution, and finally, the falling state is recognized and the risk level is evaluated.

The invention provides a method and a system for detecting surface defects of a flange sealing element. The method comprises the following steps: expanding an annular surface image into a rectangular image; constructing a defect segmentation network to process the rectangular image to obtain a preliminary defect mask; the defect segmentation network adopts an encoder-decoder structure, a parallel multi-scale feature extraction module is arranged at the tail end of an encoder, and the parallel multi-scale feature extraction module comprises a plurality of parallel convolution branches with different receptive fields; training the defect segmentation network by using a composite loss function, and performing post-processing on the initial defect mask by using a full-connection conditional random field to obtain an optimized defect mask; and inversely transforming the optimized defect mask to a Cartesian coordinate system, and identifying and marking the position and contour of the defect on the original annular surface image.

The invention relates to an oncogene prediction method based on graph variation self-coding, and belongs to the field of bioinformatics. The method is based on a dual-path neural network framework: a main path processes an original network and features enhanced by a variational auto-encoder (VAE) by using a graph attention network (GAT) so as to capture a complex relationship between nodes; the auxiliary path generates an auxiliary network and features containing global information through an APPNP algorithm, and the auxiliary network and features are aggregated through GraphSAGE to retain structural information. The model introduces jump connection and residual connection to relieve gradient disappearance and enhance feature complementarity. And finally, integrating dual-path information output prediction through a linear layer. The method is verified on a plurality of biological network data sets, the prediction accuracy, robustness and hidden relation recognition capability are remarkably improved, and a reliable tool is provided for cancer research.

The invention relates to the technical field of medical image intelligent analysis and artificial intelligence auxiliary diagnosis, in particular to a medical image disease course prediction system based on an industrial neural network, and the system comprises a reference generation module which is used for obtaining static image data; processing the static image data by using a physical perception neural network to generate a pure ideal state reference; a perturbation simulation module; the industrial kinetic parameters are used as perturbation terms to be superposed to a pure ideal state reference, and a theoretical damaged state is generated; the projection verification module is used for acquiring real multi-modal observation data; generating a real residual error; generating a theoretical residual error based on the theoretical damaged state and the pure ideal state reference; calculating a manifold coupling confidence coefficient; the closed-loop correction module is used for performing inversion optimization on the industrial kinetic parameters; outputting a disease course prediction result according to the manifold coupling confidence coefficient; according to the method, the problem that a traditional medical model lacks physical consistency explanation is solved, and the credibility of artificial intelligence auxiliary diagnosis is remarkably improved.

A neural network processing device, an electronic apparatus, and an operation method The neural network processing device comprises a storage module, an input data processing module, a plurality of matrix compute-in-memory units, an output data processing module, and a control module, wherein the storage module comprises: a first storage sub-module configured to store an input feature map for neural network processing, and a second storage sub-module configured to store an intermediate feature map for neural network processing; and the plurality of matrix compute-in-memory units are configured to store the weight of at least one network layer of a neural network and use the weight to perform matrix-vector multiplication computation of the neural network. The neural network processing device can be applied to a compute-in-memory system for image processing, thereby lowering the computation delay, and reducing the area and power consumption.

The invention provides a scene semantic occupancy prediction method based on probabilistic three-dimensional Gaussian, and the method comprises the steps: processing a multi-view camera image through employing a two-dimensional backbone network, generating multi-view and multi-scale image features, predicting the three-dimensional occupancy distribution aligned with pixels through the image features, and initializing the position attribute of the three-dimensional Gaussian according to the distribution; the method comprises the following steps: taking three-dimensional gauss as point cloud data, generating sparse voxel data through voxelization, and processing the sparse voxel data by using three-dimensional sparse convolution to complete feature interaction between the three-dimensional gauss; by using a cross attention mechanism, updating three-dimensional Gaussian features by image features; performing layer-by-layer iterative optimization on each attribute in the three-dimensional Gaussian features by using a Gaussian attribute fine tuning model; and aggregating geometric attributes and semantic attributes of the three-dimensional Gaussian by using a probability Gaussian superposition method, and generating a three-dimensional semantic occupation prediction result of the scene. According to the invention, the problem of efficient adaptive three-dimensional semantic occupancy prediction of an automatic driving scene can be solved.

The invention relates to the technical field of computer vision and neural network processing, in particular to a human body posture recognition method based on a multi-modal neural network framework. Comprising the following steps: collecting event data, and representing the event data by using a five-channel event representation method; dividing the preprocessed five-channel event data into three channels, namely a space channel, an event characteristic channel and a point cloud channel; after channel splitting is completed, feature extraction is carried out respectively, and features of different modals are optimized and integrated through dynamic feature enhancement and cross-modal fusion; a skeleton constraint module and a space-time joint processing module are adopted to enhance understanding of the network on a human body structure and modeling of a space-time dependency relationship; a prediction head module is adopted to output final human body key point coordinates; and carrying out training optimization by adopting a multi-task learning framework to obtain a posture recognition result. The method has the advantages that the sparse asynchronous characteristic of event data is effectively processed, the blind spot problem of static limbs is solved, low power consumption is kept, and high-precision human body posture estimation is achieved.

The invention relates to the technical field of deep learning and computer vision, and discloses a histogram interactive attention-based pavement crack segmentation network processing method and system, so as to enhance the edge detail fidelity and improve the crack segmentation precision. The method comprises the steps of image preprocessing, up-sampling, down-sampling, feature fusion and image reconstruction processing. Wherein global feature modeling in the intensity sub-boxes and among the sub-boxes is realized by constructing a histogram interactive attention module (HIA); a double-branch detail enhancement feedforward module (DDEF) is introduced to enhance spatial detail and high-frequency edge information expression; meanwhile, a Fourier jump enhancement module (FFSM) is adopted to jointly refine jump connection features in a spatial domain and a frequency domain. Through the synergistic effect of the modules, the network can realize continuous recovery and structural consistency modeling of a crack boundary in a complex pavement environment, so that the accuracy and the stability of a segmentation result are remarkably improved.

The invention relates to the technical field of switch cabinet temperature measurement, and discloses a method for measuring the temperature of a switch cabinet in real time through an infrared matrix based on multi-modal sensing. The method comprises the following steps: synchronously acquiring an infrared image sequence, an acoustic vibration signal and current waveform data of the switch cabinet through a multi-mode sensor array, and capturing equipment operation and temperature associated information; constructing a three-dimensional temperature field reconstruction mechanism model based on a heat conduction physical equation, outputting a theoretical temperature distribution value fitting a physical rule, processing multi-source data by using a spatial-temporal feature extraction network, and outputting a temperature prediction value reflecting a real-time working condition; inputting the two into a dynamic fusion module based on a Markov decision process to generate a fusion temperature distribution value, and generating a temperature anomaly positioning instruction through a rolling time domain optimization strategy according to the fusion temperature distribution value; when the deviation between the actual temperature sampling value and the fused temperature distribution value exceeds a preset threshold value, calibrating a network output layer parameter; and triggering a linkage control protocol of the switch cabinet protection system according to the abnormal positioning instruction.

Described herein are systems, methods, and instrumentalities associated with using a multi-layer perceptron (MLP) neural network to process medical images of an anatomical structure. The processing may include padding an input image in accordance with the training of the MLP neural network, splitting the input image (e.g., the padded input image) into patches of a same size, and processing the patches through the MLP neural network over one or more iterations. During an iteration of the processing, the patches may be processed separately and re-combined into an intermediate image before the intermediate image is shifted to concatenate portions of the image that are derived from different patches. This way, global features of the anatomical structure may be learned and used to improve the quality of the image generated by the MLP neural network, without incurring significant computation or memory costs.

The invention relates to an Internet of Vehicles channel selection method based on a multi-agent depth deterministic strategy gradient algorithm, and belongs to the technical field of communication. The method comprises the following steps: establishing an Internet of Vehicles system model, and setting local and global environment variables, a vehicle action space and a reward function of the Internet of Vehicles system model; establishing a dynamic channel model of the Internet of Vehicles based on the system model of the Internet of Vehicles, and determining a state space and steady-state probability distribution under multi-channel combination; a graph neural network GNN is introduced into a multi-agent depth deterministic strategy gradient algorithm, and a GNN-Critic network is established to process dynamic vehicle input information; the intelligent body vehicle learns a channel selection strategy by using a global environment state so as to complete centralized training; and after training is completed, the intelligent body vehicle observes local channel state distribution and executes channel selection. According to the method, the channel utilization rate can be remarkably improved, the vehicle can make a better channel selection decision according to the global environment information, the channel idle time is reduced, and the overall network performance is improved.

The invention relates to the technical field of photovoltaic power generation, in particular to a photovoltaic power generation prediction method of a multi-algorithm hybrid model, which comprises the following steps: S1, historical data preprocessing: carrying out meteorological condition clustering on historical photovoltaic data by adopting a BIRCH algorithm, dividing the historical photovoltaic data into a sunny day data set, a cloudy data set and a cloudy and rainy data set, and screening key influence factors through a Pearson's correlation coefficient; s2, EEMD (ensemble empirical mode decomposition) and reconstruction: performing ensemble empirical mode decomposition on the preprocessed photovoltaic data, and reconstructing the data according to the energy distribution and contribution rate of an IMF component; and S3, constructing a hybrid prediction model: processing a non-stationary IMF component by using a BiLSTM neural network, processing a stationary IMF component by using a Holt-Winters model, carrying out weighted fusion on a prediction result through back propagation, and outputting a final photovoltaic power prediction value. According to the method, the advantages of deep learning and a statistical model are combined, refined preprocessing and decomposition reconstruction are carried out on data, and the precision and stability of photovoltaic power prediction are effectively improved.

The invention discloses a body finger representation reaching method based on a visual sparse hybrid expert, and the method comprises the following steps: 1, constructing an end-to-end basic model which adopts a cross-modal coding structure based on a transformer layer, and comprises a language instruction encoder, a panoramic observation encoder and a cross-modal encoder; step 2, replacing FFN layers in the panoramic observation encoder and the cross-modal encoder with MoE layers; and step 3, constructing a routing network, and dynamically allocating expert networks to process different visual features according to the visual lexical elements and the category information. According to the method, collaborative optimization of navigation view angle selection and target object recognition tasks is realized through the multi-task processing capability and the dynamic routing mechanism of the hybrid expert model, and the problem of difficulty in view and object heterogeneous relationship modeling in language-guided navigation is effectively solved.

The invention provides a structure health early warning method and system based on space-time correlation characteristics and digital twinning, and relates to the technical field of data processing. The method comprises the following steps: acquiring multi-source heterogeneous data of a target structure; performing dynamic sampling alignment and wavelet packet decomposition on the multi-source heterogeneous data to extract energy features to obtain synchronous data, and performing abnormal data filtering on the synchronous data to obtain cleaned fusion data; performing wavelet decomposition on a high-frequency vibration signal in the fused data to obtain a damage impact feature, performing time sequence processing on low-frequency temperature data in the fused data to obtain a temperature time feature, and performing dynamic graph convolutional network processing on strain data in the fused data to obtain a spatial correlation feature; constructing an input vector; calculating a damage degree index; and according to the damage degree indexes, early warning grades are divided, and corresponding control instructions are triggered for different early warning grades. By implementing the technical scheme provided by the invention, the accuracy of structural health early warning can be improved.

The invention provides a vector geographic target overall extraction method and device for a large-format remote sensing image. The method comprises the following steps: a context sensing segmentation stage: processing the whole remote sensing image by adopting a context attention network, extracting context information through a local-to-global attention mechanism, and generating a segmentation mask of a target; in the mask edge reconstruction stage, a target contour is extracted from the segmented mask, and a regular polygon sequence with uniformly distributed vertexes is generated through polygon simplification and fixed distance interpolation reconstruction; in the polygon sequence tracking stage, the regularized polygon sequence is input into a polygon sequence tracker, and a complete vectorized geographic target is output through position offset correction and vertex classification; wherein in the three stages, end-to-end whole-scene processing is carried out on a large-format remote sensing image, and blocking operation is not needed. The method aims at overcoming the limitation of a traditional method and providing a complete and coherent vector result for high-precision geographic information extraction.

Embodiments described herein are generally directed to an end-to-end trainable degradation restoration network (DRN) that enhances the ability of a super-resolution (SR) subnetwork to deal with noisy low-resolution images. An embodiment of a method includes estimating, by a noise estimator (NE) subnetwork of the DRN, an estimated noise map for a noisy input image; and predicting, by the SR subnetwork of the DRN, a clean upscaled image based on the input image and the noise map by, for each of multiple conditional residual dense blocks (CRDBs) stacked within one or more cascade blocks representing the SR subnetwork, adjusting, by a noise control layer of the CRDB that follows a stacked set of a multiple residual dense blocks of the CRDB, feature values of an intermediate feature map associated with the input image by applying (i) a scaling factor and (ii) an offset factor derived from the noise map.

The invention discloses a dynamic single-target long-time tracking method and system, and belongs to the technical field of computer vision and artificial intelligence. The method comprises the following steps: initializing a deep twin tracking network for main tracking and a trajectory prediction model as a standby; processing the video stream using the main tracking network to obtain a preliminary tracking result and a response graph; judging the real-time tracking state of the target based on the average peak correlation energy of the response diagram and the color histogram similarity; when normal tracking is judged, the result of the main tracking network is adopted and used for updating the prediction model; and when it is determined that the target is blocked or lost, activating the prediction model, and predicting the target position according to historical motion information. According to the invention, through dynamic strategy switching of the main tracker and the predictor, stable and efficient tracking in a complex scene, especially when the target is shielded, is realized, and the method is suitable for edge computing platforms such as an unmanned aerial vehicle and a robot.