223 results about "Stream network" patented technology

The invention belongs to the technical field of logistics supply chain management, and discloses a logistics supply chain dynamic risk identification method based on a large language model. A real-time heterogeneous data stream is subjected to space-time normalization processing through a dynamic sliding window mechanism, a three-dimensional space-time data tensor set with entity types, timestamps and space grid codes as dimensions is constructed, dynamic entities in a logistics supply chain and the incidence relation of the dynamic entities are recognized through an entity-relation-time triple extraction module, and the real-time heterogeneous data stream is obtained. Constructing a dynamic knowledge graph with space-time attributes; generating an incremental graph version according to the change event of the entity state, performing multi-dimensional anomaly detection in combination with a corresponding graph change log, and generating a structured risk tag; when a risk event occurs, the time-space coordinates of the root cause of the risk event are accurately positioned through a version backtracking function. The real-time performance, the accuracy and the interpretability of supply chain risk identification are remarkably improved, and a systematic solution is provided for dynamic risk management of a complex logistics network.

The invention provides a deep learning-based die process die surface automatic supplement generation method and system, and relates to the technical field of deep learning, and the method comprises the steps: extracting the contour line and feature point information of a die surface as a training sample; feature learning is carried out based on a multi-scale image convolutional network and a double-flow network, and optimized local geometric features and global topological features are extracted; generating a supplementary modular surface through the generative adversarial network; and carrying out stress field constraint verification and then fusing with an original model. According to the method, intelligent supplementary generation of the mold surface is realized, and the mold design efficiency and quality are improved.

The invention relates to a medical data dynamic management method based on multi-modal data fusion and deep reasoning, which comprises the following steps: establishing a three-layer dynamic management architecture, collecting various medical data of a patient in real time through a bottom layer, performing primary processing on the various medical data of the patient, receiving the primarily processed medical data through a middle management layer, and performing dynamic management on the medical data through the middle management layer. The double-flow Transform network architecture performs medical event prediction, the event-driven architecture triggers a data processing flow, medical data after primary processing is subjected to dynamic extraction, data conversion, data loading and multi-modal data fusion processing, a data processing result is transmitted to the application layer, the application layer performs dynamic field generation, and the event-driven architecture performs multi-modal data fusion processing. And data automatic dynamic management and real-time feedback calibration are carried out based on a data fusion analysis result. The limitation of a static field structure existing in an existing medical data management SEER database is solved, the data quality is improved, automatic management and real-time feedback calibration of follow-up visit data are achieved, and separation of a follow-up visit system and a data management platform is avoided.

The invention provides a fast-charging charger service life prediction method based on deep learning, and the method comprises the steps: collecting electrical, thermodynamic and environmental parameters, carrying out the denoising through extended Kalman filtering, employing adaptive quantile normalization and time window weighted interpolation to process missing values, extracting the time sequence and thermodynamic characteristics in a sliding window, constructing a double-flow Transform-LSTM network, and carrying out the prediction of the service life of a fast-charging charger. The method comprises the following steps: respectively processing an electric current flow and a thermodynamic flow, introducing a stage perception attention mechanism, dynamically adjusting a feature weight, integrating double-current output through weighting, designing a physical loss function, finally outputting residual life, calculating a health state through weighting, triggering multi-stage early warning based on a threshold value, and classifying fault types in combination with Softmax. Abnormal data are selected through incremental training, the model is regularly and finely adjusted, physical constraint and deep learning are combined, charging stage characteristics are dynamically adapted, high-precision life prediction and fault diagnosis are achieved, and the method is suitable for real-time health management of fast charging equipment.

The invention provides an unsupervised industrial anomaly detection method based on an improved full-convolution cross-scale flow network. The method comprises the steps of performing data preprocessing on an industrial image data set; industrial image data is used as input, a pre-trained visual backbone network is used for extracting multi-scale features, a resolution-perceived channel attention module is used for enhancing expression of the multi-scale features, and an enhanced multi-scale feature map is obtained; performing multilayer reversible transformation by taking the enhanced multi-scale feature map as input and taking an ICSF-Net model based on hierarchical attention and expansion convolution as a cross-scale normalized flow network, and modeling multi-scale feature distribution; in the training stage, optimizing and updating ICSF-Net model parameters by maximizing the log likelihood of a normal sample under potential Gaussian distribution based on multi-scale feature distribution and combining an LSGR mechanism; in the reasoning stage, probability density estimation is carried out based on multi-scale feature distribution, an abnormal score graph is generated, and defect detection and positioning are achieved. According to the invention, the accuracy, robustness and pixel-level positioning precision of industrial product defect detection are improved.

The invention provides a damage assessment method based on heterogeneous double-flow network and multi-strategy adaptive decision fusion. The method aims at solving the technical defects of an existing method in the aspects of dual-time-phase feature alignment, multi-scale perception and fusion strategy self-adaption. The method comprises the following steps: registering and enhancing unmanned aerial vehicle images before and after damage; afterwards, feature extraction is carried out through a heterogeneous double-current feature extraction module, a cross-temporal window alignment mechanism is introduced into a global context awareness stream, strict spatial alignment of double-temporal features is ensured, and capture of multi-scale damage details is enhanced through multi-receptive-field optimization of a local detail enhancement stream; finally, through a multi-strategy self-adaptive decision fusion module, a strategy selector is used for dynamically calculating the weight, weighted decision is carried out on output of the association perception fusion strategy, the explicit change detection strategy and the robust weighted fusion strategy, and finally the damage level is output through a classifier. According to the invention, the accuracy, robustness and adaptive ability of damage assessment in a complex battlefield environment are effectively improved.

The invention discloses a method for optimizing micro-crack segmentation based on deep learning and super-resolution reconstruction, and the method comprises the steps: cutting an image in real time, obtaining an image block which takes a component as a target main body, and synchronously recording a homography matrix for geometric mapping; selecting an amplification strategy to improve the resolution, and recording a scale mapping relation; inputting the enhanced image block into a double-flow network; adaptive fusion and reconstruction are carried out on the two branch features, and a high-resolution texture image is output; generating a geometrically corrected ortho-image, fusing the geometrically corrected ortho-image with original illumination information, and outputting a corrected image with a known pixel size; identifying cracks, spalling and honeycomb diseases in parallel; generating a unified defect confidence map; calculating real geometric parameters of the BIM in a BIM global coordinate system through coordinate back projection; and generating quantitative defect reports and maintenance suggestions. The method has the advantage that seamless connection between the detection result and the BIM global coordinates is realized.

The invention relates to a geological disaster change detection method and system based on an improved twin U-Net and a central surrounding double-flow network. According to the method, an improved twin U-Net network is constructed as a feature extraction trunk, and a deformable convolution module is integrated in an encoder to adaptively adjust a receptive field; a central surrounding double-flow network is embedded in a decoding path, detail features are extracted through a central flow path, and a global context is obtained through a surrounding flow path; a feature fusion module is designed to realize double-path feature deep fusion, and a gating attention unit is adopted to calibrate feature response; introducing a contrast feature learning mechanism to reinforce feature space clustering characteristics; multi-modal degradation enhancement training is implemented to improve the robustness of the model; and an objective function containing difference perception loss and feature comparison loss is minimized through end-to-end joint optimization. The method can accurately identify geological disaster change areas such as landslide, debris flow and land subsidence, and has the advantages of high detection precision and strong anti-interference capability.

The invention discloses a heterogeneous data processing method and system for energy big data, and the method comprises the steps: obtaining different types of energy data, inputting the different types of energy data into a pre-trained directed association graph model, and obtaining a cross-type dynamic contribution flow network node forest; a cross-type linkage report is obtained, and a basic directed positioning dynamic correction chain is constructed through report analysis positioning and causal analysis; and inputting the cross-type dynamic contribution flow network node forest and the basic directed positioning dynamic correction chain into a pre-trained positioning correction path model, and performing real-time positioning error correction adjustment on the cross-type linkage report to obtain an updated and adjusted cross-type linkage report. According to the invention, cross-type linkage analysis and report real-time error correction of the energy data are realized, and the accuracy and efficiency of energy data processing are improved.

A method for training a generative flow network is provided, and is applied to the field of artificial intelligence technologies. In the method, in a process of training the generative flow network, for any state of an agent, a plurality of first actions performed in the state and a plurality of second actions that can be transferred to the state are selected from a continuous action space in a sampling manner, then, predicted values corresponding to the plurality of first actions and the plurality of second actions are output by using the generative flow network, and further, a loss function used to update the generative flow network is obtained through calculation. In this solution, a plurality of actions obtained through sampling are used to approximately represent the continuous action space, and then, the generative flow network is trained.

The invention discloses a visual angle robust traffic accident detection method based on space-time attention domain adaptation. The method comprises the following steps: constructing a double-flow network architecture; the progressive multi-granularity spatial domain adaptation module processes appearance change caused by a visual angle through global and local feature alignment, and uses SSAM to generate a semantic mask and entropy-guided mobility weight to realize accurate spatial domain adaptation; the time collaborative attention module realizes dynamic alignment of accident related time information between different visual angles through cross-domain video clip correlation calculation and a collaborative attention mechanism; a comprehensive contrast learning strategy is introduced, and noise robust motion feature learning is enhanced; a multi-view-angle joint training strategy is adopted, cross-domain knowledge migration is carried out by using monitoring view angle, vehicle-mounted view angle and unmanned aerial vehicle view angle data, and model optimization is realized through a comprehensive loss function. According to the method, the performance is remarkably improved in a cross-view migration task, and an efficient solution is provided for a unified multi-view traffic accident detection system.

The invention relates to an intelligent identification and evaluation technology for tornado ground disasters, and the technology comprises a data collection and preprocessing module which collects image data and text data of tornado ground disasters from a plurality of data sources, carries out the preprocessing and cleaning of the data, and generates a structured data set; the multi-modal disaster assessment model is based on a VLLM and an advanced visual identification technology, adopts a double-flow network structure and comprises an image feature extraction branch and a text processing branch, the image feature extraction branch extracts key disaster features from a high-resolution image, and the text processing branch deeply analyzes text information related to disaster grade assessment; and the interactive learning mechanism allows the multi-modal disaster assessment model to request feedback to a user after preliminary analysis, and performs self-adjustment and optimization according to the feedback. The invention further provides a map display method. According to the method, the problems of data scarcity, image diversity, low damage evaluation efficiency, rapid processing requirements and model generalization ability can be effectively solved.

The invention provides a photovoltaic power prediction method and system based on a double-current network and multiple attention, and the method comprises the steps: obtaining historical photovoltaic generating capacity and meteorological information, and obtaining initial sample data; performing variable factor influence analysis on the initial sample data to obtain a simplified initial sample data set; performing similar daily clustering processing on the simplified initial sample data set to obtain each scene data set after clustering; providing a photovoltaic power prediction initial model based on a double-current network and multiple attention, and training the photovoltaic power prediction initial model by using each clustered scene data set to obtain a photovoltaic power prediction model; and taking the historical generating capacity and meteorological data of the target photovoltaic power station in the set time period as input data of the photovoltaic power prediction model, and outputting a photovoltaic power prediction result. According to the invention, through the clustering integration method, the clustering robustness and accuracy are improved; and through multi-dimensional feature extraction and feature fusion, the prediction accuracy is improved.

The invention relates to a layout wiring planning design method based on a 2.5 D core particle interconnection interposer, which is characterized by comprising the following steps of: defining signal logic interfaces for automatic wiring in a passive interposer according to signal transmission requirements among core particles and a communication relationship between the core particles and a packaging substrate, and creating a mapping relationship list among the logic interfaces; according to the method, layout planning is carried out on physical interface endpoints of a passive intermediate layer, the physical interface endpoints are matched with signal logic interfaces, and an optimal layout allocation scheme is quickly determined through a signal allocation algorithm based on a flow network in combination with a window matching method, so that the wiring resource loss of a 2.5 D integrated circuit is effectively reduced, and the wiring efficiency is improved. The system overall performance is improved. According to the invention, the method can cope with a complex interconnection wiring condition generated by a plurality of independently designed core particles in a 2.5 D integration process, and achieves the efficient interconnection design of the passive interposer.

The invention provides a machine vision microdefect detection method and system based on spectrum-polarization-phase three-domain collaborative perception, and the method comprises the steps: constructing a third-order tensor through synchronously obtaining the spectrum, polarization and phase information of the surface of a to-be-detected object, and extracting features through a lightweight multi-stream network; and dynamic fusion of three-domain information is realized by adopting a cross attention mechanism, and finally defect category, position and size information is output. According to the method, the problem of insufficient single-mode information is solved, the surface defect characteristics of the object can be more comprehensively reflected by combining three-domain information, and the capability of identifying the defects of complex objects such as high-reflective surfaces and transparent materials is improved; a lightweight multi-stream network and a cross attention mechanism are adopted, the multi-domain information fusion efficiency is improved, the network parameter quantity is reduced, deployment at a production line edge calculation unit is facilitated, and the detection speed and cost requirements are met; the method can better adapt to surface form changes of complex objects such as curved surface parts and flexible materials, the defect omission ratio is reduced, and the product yield is improved.

The invention discloses a cross-modal pedestrian re-identification method and system based on a multi-scale joint learning network, and relates to the technical field of pedestrian re-identification, and the method comprises a four-flow network architecture which enables a model to extract diversified semantic features through a mode of separating an application data enhancement branch from an original branch. Random channel selection and self-adaptive graying are respectively applied to the data enhancement branch, so that the robustness of the model to color change and the adaptive capacity of the model to different thermal imaging conditions are improved. Important channels are enhanced through a channel attention mechanism, irrelevant channels are inhibited, and two substreams are guided to internally enhance modal specific features. Richer semantic information is reserved through features extracted by a four-flow network, a joint semantic learning module is designed, a group of learnable vectors are defined and spatial position codes are added, global features of original branches and color invariant features of channel data enhancement are fully learned under the guidance of loss, and the overall feature of the original branches is optimized; and the features among different modes have higher semantic consistency.

The present invention provides a skeleton action recognition method based on multi-dimensional dynamic topology learning graph convolution, which relates to the field of computer vision. The skeleton action recognition method based on multi-dimensional dynamic topology learning graph convolution comprises three components: pure node topology learning graph convolution, dynamic temporal specificity topology learning graph convolution, and channel specificity topology learning graph convolution. Among them, in the dynamic temporal specificity topology learning graph convolution, we propose a dynamic skeleton topology modeling method to efficiently model the dynamic skeleton topology rich in global spatio-temporal topology features. And multi-scale temporal convolution is used to obtain multi-scale temporal features. In addition, in order to supplement the spatial information of the skeleton data, we additionally introduce relative node data and relative bone data for the fusion of the multi-stream network. This model can extract more comprehensive and effective action features and perceive more subtle action differences, which is worthy of vigorous promotion.

The invention provides a double-current comparison and DHI combined equipment fault detection method and system, and belongs to the technical field of power equipment state monitoring and fault diagnosis. The method comprises a time sequence perception adversarial enhancement module, a generative adversarial network (GAN) data expansion module, a double-flow contrast attention network (D-CAN) feature extraction module and a dynamic health index (DHI) calculation module. The core lies in that a fault sample with time sequence correlation is supplemented through a physically constrained GAN, robust features are extracted by using a ResNet1D and Transform fused double-flow network, and a health state is quantified in combination with unsupervised clustering and mahalanobis distance. Through simulation and experimental verification, zero-delay detection of the early fault of the transformer can be realized, the output health index and the 3D visualization result can provide an accurate basis for operation and maintenance of the transformer, and the diagnosis accuracy, the data utilization rate and the dynamic adaptability of state evaluation are remarkably improved.

The invention discloses an IPv6 network topology detection method and system based on address prefix prediction, and the method comprises the steps: obtaining an IPv6 address prefix and a survival IPv6 address from a public data source, and carrying out the matching of the survival IPv6 address with the corresponding address prefix based on the longest prefix matching; dividing the IPv6 address prefixes into two types, namely an IPv6 address prefix containing a survival / 64 subnet prefix and an IPv6 address prefix not containing the survival / 64 subnet prefix; an IPv6 network topology detection technology based on a split hierarchical clustering algorithm is adopted for an IPv6 address prefix containing survival / 64 subnet prefixes; and an IPv6 network topology detection technology based on reinforcement learning is adopted for an IPv6 address prefix which does not comprise a survival / 64 subnet prefix. The IPv6 address hit rate, the IPv6 address discovery rate and the IPv6 address coverage of the IPv6 network topology detection method are superior to those of the current mainstream IPv6 network topology detection technology.

The invention discloses a hyperspectral image classification method, system and device under a cross-scene condition and a medium, and mainly solves the problems that the existing cross-domain feature space is not rich and the classification boundary is not robust, and the scheme comprises the following steps: obtaining source domain data, a source domain label and target domain data; a classification model composed of a Mama flow network and a ViT flow network in parallel is constructed, and the Mama flow network comprises a Mama block, a spectrum SKR block and two spectrum-oriented classifiers and is used for optimizing spectrum classification boundaries; the ViT flow network comprises a self-adaptive dynamic mask, a ViT block, a space SKR block and two space guiding classifiers, and is used for enriching a feature space and optimizing a space classification boundary; inputting the source domain data, the source domain data label and the target domain data into a classification model, and performing iterative training on the classification model; and inputting the target domain data into the trained classification model and outputting a classification result. The method improves the classification capability of the cross-scene hyperspectral image, and can be used for environmental monitoring and landform change prediction.

The invention discloses a traffic anomaly detection method based on a space-time double-flow network and multi-modal feature fusion, relates to the technical field of intelligent video analysis and traffic behavior recognition, and is used for solving the problem of high false detection rate of existing traffic anomaly detection in a complex environment. A video image sequence is extracted through a fixed and dynamic combined frame sampling strategy, and motion features are constructed. And in combination with the direction consistency coefficient and the perturbation evolution trend, repairing direction perception distortion by using a mirror image mapping mechanism. Target appearance features are extracted through spatial branches, behavior evolution trajectories are extracted through time branches, and meanwhile, a modal weight dynamic adjustment mechanism is introduced to enhance the adaptability to complex scenes. The multi-target interaction relation is quantified based on a graph structure modeling mode, an adjacent structure is reconstructed through a graph self-encoder, and local behavior mutation is recognized. And finally realizing high-confidence anomaly marking by combining state transition analysis and spatial aggregation characteristics. The method has the advantages of high real-time performance, high robustness and wide scene adaptability.

The invention discloses a strength evaluation method and device for a network following and constructing interconnection system. The method comprises the following steps: constructing an extended admittance matrix and an extended impedance matrix of the interconnection system based on capacity information of various devices on a device side in the interconnection system and information on an alternating current network side; respectively constructing a network following feature subsystem feature equation and a network construction feature subsystem feature equation through the information of the alternating current network side, the extended admittance matrix of the interconnection system and the extended impedance matrix of the interconnection system; based on the network following feature subsystem feature equation and the network construction feature subsystem feature equation, first network following strength of the interconnection system and first network construction strength of the interconnection system are obtained respectively; and based on the first network following strength of the interconnection system and the first network construction strength of the interconnection system, obtaining second network following strength of the interconnection system and second network construction strength of the interconnection system. By means of the scheme, comprehensive quantitative evaluation of the stability of the interconnection system can be achieved, and scientific guidance is provided for reasonable configuration of network following / constructing equipment.

The invention provides a face counterfeiting recognition detection method and device based on image high-frequency noise and a GhostNet network, and the method comprises the steps: 1, selecting a visual converter as a backbone network of a double-flow network, so as to preliminarily extract image features, respectively introducing a GhostNet module into two branches of the double-flow network to carry out deep feature extraction on the image, and introducing a cross-modal attention fusion device behind a backbone network to integrate double-flow features; and adding a perceptron adapter behind the cross-modal attention fusion device so as to realize classification of true and false images. Step 2, using a mask image modeling self-supervision pre-training method to train a double-flow network; and step 3, high-frequency noise in the face image is extracted through a high-pass filter SRM, the high-frequency noise and the low-frequency texture of the image are respectively used as input of the double-flow network and are sent into the double-flow network, and a face counterfeiting recognition result is obtained.

The invention discloses a psychological health cloud service platform system and method based on multi-modal data analysis. The system comprises a multi-modal data acquisition module; a heterogeneous data fusion analysis module; a dynamic intervention decision module; a block chain evidence tracking module; an edge computing terminal; the output end of the multi-modal data acquisition module is electrically connected with the input end of an edge computing terminal, the output end of the edge computing terminal is electrically connected with the input end of a heterogeneous data fusion analysis module, and the output end of the heterogeneous data fusion analysis module is electrically connected with the input end of a dynamic intervention decision module. According to the method, multi-modal data are integrated to construct a dynamic psychological portrait, a double-flow network is adopted to model spatio-temporal characteristics, an intervention scheme is matched based on risk levels, block chain evidence storage closed-loop management is performed, and an edge computing terminal performs cooperative computing through 5G encryption, so that real-time security is guaranteed, and the method is suitable for basic education scenes.

The invention discloses a human behavior prediction method based on a double-flow space-time diagram convolutional network, and belongs to the technical field of computer vision and deep learning, and the method comprises the following steps: S1, obtaining a human skeleton data set; s2, preprocessing skeleton point data; s3, processing the preprocessed skeleton point data through a double-flow space-time diagram convolutional network; s4, constructing a multi-scale joint point spatial dependency relationship; s5, processing the output features of the first stream through frequency weighting and a channel feature importance attention mechanism, dynamically adjusting the feature weights of high-frequency and low-frequency channels, and generating enhanced spatial-temporal features; and S6, fusing the output features of the double-flow network, generating a final behavior prediction result, and realizing human body behavior prediction. According to the method, through a multi-space category modeling graph convolution module and a frequency weighting and channel feature importance attention mechanism, a multi-space category modeling double-flow space-time graph convolution network is further provided, and the accuracy of behavior prediction is remarkably improved.

The invention discloses a cross-modal fusion target detection method and system for a low-light storage environment, and is applied to the technical field of cross-modal fusion target detection. The method comprises the following steps: obtaining a cross-modal image data set, and carrying out timestamp alignment and spatial registration preprocessing on an image; designing a cross-modal dynamic fusion state space model; training a cross-modal dynamic fusion state space model based on the preprocessed cross-modal image data set to obtain a CMDS cross-modal detection model; and inputting a to-be-identified image into the CMDS cross-modal detection model to obtain each target detection result of the to-be-identified image. According to the invention, through optimization of multi-modal training data and a double-flow network flow, the problem of feature dislocation existing in cross-modal target detection of an existing system is solved, and especially in complex environments of low light, no light and the like, the detection precision and robustness are remarkably improved.

The invention discloses a watershed ditch and pond-road low-cost ecological sand control method based on hydrological connectivity, which belongs to the field of ecological environment protection and mainly comprises the steps of data acquisition, DEM data correction, confluence network construction, hydrological structure connectivity identification, SWAT model construction, key source area identification, scene analysis and cost benefit analysis. Under the influence of a ditch pond-road system, the key source area identification technology based on hydrological structure connectivity and the drainage basin key source area ditch pond road low-cost water regulation and soil conservation method achieve the dual purposes of reducing the water and soil regulation and control cost and continuously improving the ecological benefit by identifying the key source area.

The invention relates to the technical field of image processing, and discloses an idle scooter detection method and system. The method comprises the following steps: acquiring a multi-source video through a city monitoring camera, preprocessing the multi-source video, identifying a scooter by using improved YOLOv8, constructing a self-adaptive region of interest through polar coordinate mapping, extracting pedestrian behavior characteristics by using a double-flow network, and calculating interaction strength in combination with space intersection, time continuity and posture identification. And finally, judging the idle state of the scooter based on the hierarchical decision-making tree and scoring. According to the method, the idle state of the shared scooter in the urban environment can be accurately identified and judged in real time, and high-precision use state evaluation is provided.

The invention discloses a human body action recognition method and system based on a skeleton spatial-temporal feature fusion graph convolutional network, and relates to the field of computer vision, the method comprises the following steps: inputting a to-be-detected video to a posture estimation algorithm to obtain a human body skeleton data set of the to-be-detected video; performing data preprocessing on the human skeleton data set to obtain skeleton joint information flow, skeleton joint movement information flow, skeleton information flow and skeleton movement information flow, and fusing the information flows to form a double-flow network branch; respectively inputting skeleton joint information flow-skeleton joint motion information flow branches and skeleton information flow-skeleton motion information flow branches in the double-flow network branches into a skeleton spatial-temporal feature fusion graph convolutional network model to obtain double-flow network branch output; and on the basis of double-flow network branch output, obtaining a human body action prediction result of the video to be detected by adopting a weighted fusion method. According to the invention, the recognition precision and robustness of complex actions can be effectively improved.

The invention provides a target re-identification method and device based on foreground segmentation, and relates to the field of computer vision. According to the method, through foreground segmentation processing, the target area in the to-be-recognized image can be accurately focused, and background interference is effectively eliminated. A double-flow network is utilized to extract features from an original image and a foreground image, and feature fusion is performed by combining a feature channel weight constructed based on a foreground image feature map, so that the fused feature map contains global information and highlights local key features of a target. Then global and local features are obtained through a global and local feature extraction method and are spliced, and the target to be recognized is comprehensively and meticulously described, so that the accuracy of target re-recognition is remarkably improved, and the target to be recognized and the target in the image to be compared can be more accurately matched.