937 results about "Identification rate" patented technology

The present application provides a vulnerability remediation code identification method, comprising: acquiring commit information of open-source software, the commit information comprising a committed code snippet and text information associated with the code snippet; inputting the commit information into a first language model and performing identification to obtain function change information and test case information; inputting the function change information into a vulnerability remediation identification model to obtain a vulnerability remediation code snippet for a target vulnerability, and inputting the test case information into a vulnerability test case identification model to obtain a test case code snippet for the target vulnerability; and inputting the vulnerability remediation code snippet for the target vulnerability and the test case code snippet for the target vulnerability into a second language model and performing integration and identification to obtain an identification result. On the basis of notable characteristics in a test case for a vulnerability, the identification rate of vulnerability remediation code can be effectively increased. In this way, a user can discover potential vulnerabilities in advance, thus reducing security risks.

The invention provides an anti-fraud method and system based on a complex relation network, and the method comprises the steps: carrying out the preprocessing and association rule mining of multi-source data, constructing a heterogeneous relation network model, carrying out the abnormal behavior detection of real-time data, updating the heterogeneous relation network model based on an abnormal behavior detection result and historical data, and carrying out the recognition of the abnormal behavior. And performing risk quantification, community discovery, abnormal path analysis and related result visualization on the updated heterogeneous relation network model, calculating a fraud risk score of each node through a pre-trained machine learning model based on a related result, and finally generating a service instruction of each node based on the fraud risk score and a preset score interval. By adopting the method, the problems of poor adaptability, high false alarm rate, low gang crime recognition rate and the like of the existing anti-fraud technology can be relieved.

The invention provides a financial fraud detection method based on a large language model. The method comprises the steps of obtaining a to-be-recognized text; performing word segmentation on the to-be-recognized text through the target word segmentation tool and the financial fraud dictionary, and determining a fraud sensitive word list; calculating the weight of each sensitive word in the sensitive word list according to a target algorithm to obtain a sensitive word weight feature vector; inputting the sensitive word weight feature vector and a to-be-recognized text into a large language model, and determining a context semantic vector of the sensitive word in combination with a word embedding technology; determining the similarity between the context semantic vector of the sensitive word and a preset financial fraud type semantic vector; and determining a financial fraud type according to the similarity. Through the implementation of the method, the generalization ability of the pre-training model is utilized to capture text deep semantics, priori knowledge is injected in combination with a sensitive word weight mechanism, the model is guided to focus high-risk vocabularies, the defect of a traditional method in semantic comprehension is overcome, the financial fraud recognition rate is increased, and the omission ratio is reduced.

The invention relates to the technical field of graph recognition, in particular to a ship type recognition method and system based on transfer learning, and the method comprises the following steps: carrying out the combined preprocessing of a visible light ship image and a synthetic aperture radar image, and separating a ship inherent feature layer and an environmental noise feature layer through an adversarial domain decoupling network; generating an interference-removed cross-domain feature matrix; enhancing space-time consistency features related to the ship motion mode, and outputting an enhanced feature matrix after enhancement; and S2, constructing a differentiable decision tree classifier based on the enhanced feature matrix in S2, selecting a category with the highest probability as a final recognition result to be output, comparing the confidence coefficient of the final recognition result with a preset threshold, and if the confidence coefficient is lower than the preset threshold, triggering incremental learning. The adaptive capacity to complex sea conditions, ship type diversity and shielding conditions is improved, the error recognition rate and the attack resisting success rate are remarkably reduced, and the reliability of the system in the open environment is effectively improved.

The invention discloses a federated learning model training method and system based on differential privacy image noise addition. The method comprises the following steps: S1, initializing an evaluation network, a target network and hyper-parameters; s2, initializing a global model and parameters; s3, acquiring a selected client and a privacy budget; s4, the selected client performs local training and uploads the disturbance gradient to the central server; s5, the central server aggregates the disturbance gradient of the client to obtain an updated global model and a reward corresponding to the current state; s6, the next state of federated learning is obtained, and the evaluation network and the target network are updated; s7, repeatedly executing the steps S3 to S6 until the set number of global iterations is reached; and S8, updating the exploration rate and repeatedly executing the steps S2 to S7 until the set federal learning times are reached. The invention provides a dynamic differential privacy method in combination with a federal learning framework, which can improve the image recognition rate while ensuring the privacy of the user.

The invention discloses an intelligent traceable hazardous waste management system based on RFID, particularly relates to the technical field of hazardous waste intelligent supervision, and is used for solving the problems of unreliable data acquisition, misreading and misreading and state updating delay caused by signal interference of an existing RFID system in a high-density hazardous waste storage scene. The method comprises the following steps: dynamically dividing warehouse sub-regions, distributing a unique grouping identification code, and combining a directional antenna array partition scanning technology to avoid signal crosstalk; by analyzing the multi-path phase fluctuation rate and the polarization direction offset, a main interference frequency band is dynamically identified by using a covariance decomposition algorithm, and the communication frequency is switched, so that the tag identification rate in a complex electromagnetic environment is improved; scheduling a target sub-region scanning process based on a warehouse-in and warehouse-out instruction, synchronously verifying a label state and an actual position, generating an operation link space-time atlas, actively tracing cross-region time logic contradictions and triggering full-disk scanning; and full-process accurate tracing of the hazardous waste container and complete data management and control in a high-density scene are realized.

The invention discloses a dam measurement system and method based on multi-modal data processing, and relates to the technical field of dam measurement system data processing, a physical model is adopted to extract dam physical form change features in visual data, dam physical change associated sound features in audio data and micro-displacement features in radar point cloud data; according to the method, a three-dimensional monitoring network from macroscopic deformation to microcosmic displacement is constructed through cooperation of visual, audio and radar three-mode data, a monitoring blind area of a single sensor is broken through, and correlation analysis of dam body surface deformation and internal stress change is achieved. Feature fusion driven by physical constraints is carried out, finite element model constraints are embedded in a feature level fusion stage, radar point cloud displacement vectors need to conform to geological structure direction constraints, audio and voiceprint features need to be in space-time synchronization with visual crack expansion rates, and environmental noise interference is eliminated; the crack propagation early warning time is greatly shortened; the cavitation damage positioning is more accurate, and the abnormal working condition recognition rate is greatly improved.

The invention discloses a multi-source heterogeneous knowledge fusion question and answer solution system, which relates to the technical field of computers and comprises a knowledge base layer, a recall layer, an analysis and pruning layer and an answer generation layer. The knowledge base layer is used for constructing a dynamic heterogeneous entity fusion engine and a self-adaptive knowledge slice storage mechanism, and the dynamic heterogeneous entity fusion engine comprises a cross-modal entity alignment algorithm, relation topology completion and an incremental entity evolution model; the self-adaptive knowledge slice storage mechanism is used for performing semantic density perception slicing on the RAG document and constructing a three-level index tree; according to the method, through the technologies of cross-modal entity alignment, dynamic recall weight adjustment, context sensing entity priority correction, RAG cross-block reasoning enhancement, double-encoder conflict detection, knowledge graph guide generation and the like, knowledge fusion reasoning is achieved, the complex question answering accuracy is effectively improved, the conflict recognition rate is increased, and the manual maintenance cost is reduced.

The invention relates to the technical field of computer vision and deep learning, in particular to an aerial image target detection method based on frequency domain decoupling multi-scale feature fusion, which comprises the following steps of: acquiring an aerial image of an unmanned aerial vehicle, establishing a data set, and performing preprocessing and data division; an aerial image target detection network is constructed, and the aerial image target detection network receives an input image and outputs a target category and a bounding box position; loss functions are determined, wherein the loss functions comprise classification loss representing matching quality, coordinate loss representing prediction coordinate relevancy and bounding box regression loss representing bounding box positioning accuracy; training the aerial image target detection network based on the data set and the loss function; inputting a to-be-detected aerial image into the trained aerial image target detection network to obtain a to-be-detected target category and a bounding box position; the method can improve the feature fusion degree, retains high-frequency details, and enhances the small target recognition rate.

The invention provides a palm vein feature extraction and depth adaptive matching method, which relates to the technical field of biological recognition, and is characterized in that a three-dimensional space-time convolutional neural network and a graph neural network are combined to extract space depth features, time blood flow dynamic features and blood vessel topological features of palm veins, and information bottleneck is utilized to optimize and screen a feature subset with the most distinguishing property; the method breaks through the limitation of traditional two-dimensional texture feature extraction, obtains abundant palm vein information through multi-source data acquisition (such as near-infrared imaging, a light field camera and a spectrum sensor), overcomes the interference of environmental factors such as illumination, angle and noise, achieves the high-precision modeling and stable extraction effects of palm vein features, and can achieve the real-time extraction of the palm vein features in a complex scene. A high recognition rate is kept under different illumination conditions or user hand posture changes, the robustness of the palm vein recognition system is remarkably improved, and the method is suitable for scenes with high safety requirements.

The invention belongs to the technical field of image processing, and discloses a frame image and event stream feature fusion-based target detection method, which comprises the following steps of: acquiring RGB (Red, Green, Blue) images and event stream data, and constructing a time-space synchronous multi-modal data pair; constructing a double-flow feature extraction backbone network to extract events and RGB features; based on a cross attention mechanism, multi-head attention is utilized to realize semantic alignment; the fusion weight is adaptively adjusted according to the statistical distribution; a multi-level fusion network is constructed, cross-scale fusion is performed by using an FPN pyramid, and target positioning and classification feature expression are cooperatively enhanced through a spatial semantic aggregation module. According to the target detection method based on frame image and event stream feature fusion, through a multi-modal data collaborative perception and self-adaptive feature optimization mechanism, the detection robustness in a complex scene is remarkably improved, the dynamic target omission ratio and the error recognition rate are effectively reduced, and the target detection efficiency is improved. And an all-weather high-precision environment perception capability is provided for an automatic driving system.

The invention discloses a cable tunnel fire risk feature identification system and identification method, and belongs to the technical field of cable tunnel safety monitoring. The whole stage of a fire is covered through multi-class cooperative detection, the target identification precision and scene adaptability are greatly improved, and the safety of the cable tunnel is improved. According to the whole system, a flame and smoke detection unit and a flame / smoke special detection unit are innovatively added to a detection engine module, an original heat source detection unit and an original human body detection unit are combined, a heat source-flame-smoke-human body four-category collaborative detection framework is formed, high-precision recognition is achieved on the basis of a YOLO model framework, and the detection efficiency is improved. Compared with the problems that a traditional system is high in single target detection omission ratio and cannot cover the whole stage of smoldering-initial open fire-violent combustion of a fire, the system has the advantages that the recognition rate of early flame and weak smoke is increased, the false alarm rate is greatly reduced, meanwhile, a heat source of operation and maintenance personnel is prevented from being misjudged as a fire hazard through human body detection, and the safety of the fire hazard is improved. And cable monitoring and personnel safety protection are both considered.

The invention provides a file uploading attack interception method based on semantic entropy enhancement, and aims at overcoming the defects of an existing file uploading security protection technology in the face of complex attacks. The method specifically comprises the steps that S1, file format analysis and content extraction are conducted, hidden scripts are mined through nested content recognition, and intermediate representation is generated through grammar cleaning and coding specifications; s2, constructing an abstract syntax tree and semantic entropy calculation, tracking a pollution chain, analyzing a high-risk function, identifying a high-entropy character string, modeling and controlling flow complexity, and generating a semantic entropy vector; s3, dynamic scoring and decision making are carried out, and accurate judgment is carried out in combination with white list perception, feature comparison, multi-modal model scoring, adaptive threshold and sandbox observation; and S4, carrying out real-time interception and feature synchronization, blocking malicious file landing, generating an attack log and synchronizing an attack fingerprint. The method takes the semantic entropy vector as a core, breaks through the limitation of static features, remarkably improves the recognition rate of complex attacks, reduces missed judgment, and guarantees the safety of Web applications.

The invention discloses an industrial surface defect image classification method based on hybrid query strategy active learning, and belongs to the technical field of computer image processing and machine learning. The invention aims to solve the technical problems of high cost, long period and low rare defect recognition rate caused by category imbalance due to dependence on large-scale manual labeling. The core of the method is to execute a hybrid query strategy in an iterative loop: firstly, screening out a candidate sample set of model cognitive ambiguity through uncertainty measurement of prediction entropy; secondly, in the candidate set, a core set and hierarchical thought diversity sampling method is adopted to select a final to-be-labeled sample with both characteristic representativeness and category balance. And the query batch is manually annotated and then is used for carrying out iterative updating and optimization on the model. According to the method, the recognition precision and generalization ability of the classification model on various defects can be remarkably improved with extremely low manual labeling cost, and the model development period is greatly shortened.

The invention discloses a PDF (Portable Document Format) drawing recognition and information structured extraction method. The method comprises the following steps: generating a high-resolution bitmap through image preprocessing; positioning and classifying a text region, a table region and a symbol region in the drawing based on a target detection model of transfer learning; hough transform is combined with SIFT feature matching to identify engineering symbols, and sub-pixel positioning is realized through an RANSAC algorithm; after the oblique text is corrected through affine transformation, the content is extracted through OCR; reconstructing a table structure based on OPTICS clustering and projection analysis; constructing an RDF knowledge graph according to a coordinate association rule; and using U-Net difference to detect and position an omission area and complementing the omission area. According to the method, deep learning and image processing technologies are fused, the problems of low rotating text recognition rate, table structure loss and semantic association deficiency in a traditional method are solved, through lightweight model compression and TensorRT acceleration, the analysis accuracy is remarkably superior to that of the traditional method, and the method can be widely applied to the fields of constructional engineering, petrochemical engineering and the like and has wide application prospects. And the drawing information processing efficiency and the data integrity are improved.

The invention provides a highway automatic monitoring system, and relates to the technical field of monitoring systems, and the system comprises an intelligent sensing layer which carries out the data collection and preliminary processing, and outputs structured data and an original video stream to an edge calculation layer. The edge computing layer receives the data of the sensing layer and outputs a local event alarm and prediction result to the central decision-making system, the central decision-making system performs global analysis and strategy generation and issues a grading processing instruction to the collaborative early warning platform, and each terminal performs linkage response through information distribution and execution of the collaborative early warning platform; the highway automatic monitoring system realizes road environment total factor perception through the multi-source sensing array, the video analysis unit and the pavement state monitoring module of the intelligent perception layer, effectively integrates multi-source data through a spatio-temporal data fusion technology, significantly improves the recognition rate of the system to abnormal events, and improves the safety of the system. And the system can respond to various traffic events more accurately.

The invention discloses a metal defect identification method trained by a small number of training samples, and the method comprises the steps: completing the labeling of a data set through Label labeling software, and carrying out the data enhancement through random frequency domain filtering; a wavelet convolution M-WTConv module based on multi-level feature interaction extraction is added; adding a windowed cross attention DM-FAM module based on frequency domain analysis, dynamic frequency band selection and dynamic frequency perception; a semantic enhancement cross-modal fusion module SECF is added to improve the feature representation capability of a detection head in the YOLOv10; and gradually training the improved neural network in two stages to complete metal defect identification. According to the method, the target detection capability of metal surface defects under high-reflection and few-sample scenes is effectively enhanced, and higher robustness, recognition rate and accuracy are achieved for metal defect detection and recognition tasks under the condition of few samples.

The invention relates to the technical field, in particular to an offline handwritten mathematical formula recognition method based on deep learning, which comprises the steps of data acquisition and processing, formula detection and symbol segmentation, symbol feature extraction and recognition, structural analysis and reconstruction and evaluation and optimization. According to the offline handwritten mathematical formula recognition method based on deep learning, the robustness of a model to partial shielding is improved through decomposition and recombination, real handwritten deformation is simulated through local distortion, the generalization ability is enhanced through global distortion, a low-contrast image is improved according to CLAHE equalization, the illumination influence is reduced, and formula data enhancement diversification processing is ensured; a formula area is positioned through YOLOv5, adhesion symbols are accurately segmented and processed according to U-Net, global features are extracted by adopting Transform, a Tesseract OCR is called, a symbol library is self-defined, the OCR and a rule engine are combined, the special symbol recognition rate can be increased, a syntax tree is generated through syntax tree and PCFG analysis and recursive descent, an optimal structure is selected, and a two-dimensional layout is accurately analyzed.

The invention provides a risk early warning method and system for medical insurance settlement data based on a knowledge graph, and relates to the technical field of medical insurance settlement, and the method comprises the steps: formulating a risk rule, obtaining the medical insurance settlement data, extracting a first entity and a relation thereof from the data, constructing the knowledge graph, and storing the knowledge graph in a graph database; and searching the second entity according with the risk rule and the relationship thereof in the graph database by utilizing the reasoning ability of the knowledge graph, extracting the risk characteristics of the second entity, determining the risk level of the second entity according to the risk characteristics, and sending a corresponding early warning signal, thereby improving the recognition rate of the complex associated risk in the process of medical insurance settlement.

The invention relates to the technical field of block chain technology and asset transfer, in particular to a multi-risk interception and artificial intelligence error correction method in RWA asset cross-chain transfer, comprising the following steps: step 1, cross-chain transfer initialization and asset mapping verification; step 2, multi-level dynamic risk interception; step 3, an artificial intelligence error correction mechanism; step 4, asset right confirmation and log auditing after cross-chain completion; according to the method, technical risks (such as contract vulnerabilities), data risks (such as information inconsistency) and behavior risks (such as abnormal transactions) in RWA asset cross-chain transfer are covered through a multi-level interception system of "pre-transaction-in-process-contract layer" in combination with AI dynamic learning ability, the risk identification rate is greatly improved compared with a traditional method, and the risk identification efficiency is greatly improved. An artificial intelligence error correction mechanism is used for automatically classifying risk types and generating strategies, so that the average error correction time is shortened, and the error correction success rate is improved.

The invention discloses a radar communication radiation source identification method and system based on multi-modal feature alignment, and belongs to the technical field of electronic reconnaissance and signal processing. The method comprises the following steps: preprocessing a received radar signal to generate a standardized time-frequency graph; extracting a signal feature vector through a specially designed convolutional neural network encoder; extracting a text feature vector by using a Transform encoder; through joint optimization of cosine similarity loss and physical parameter constraint loss, alignment of signal-text features in a unified vector space is realized; and finally, zero sample identification of unknown radar signals is realized through vector similarity calculation. According to the method, the problems of low recognition rate and insufficient cross-modal information fusion in a low signal-to-noise ratio environment of a traditional method are effectively solved, and the recognition precision and the system robustness are remarkably improved.

The invention discloses a multifunctional radar working mode recognition method based on an LSTCN network, and the method comprises the steps: firstly analyzing the working mode characteristics of an airborne phased array radar, generating multifunctional radar pulse group data, inputting a pulse data generation program, carrying out the data preprocessing, dividing a data set into a training set and a test set, and carrying out the recognition of the working mode of the airborne phased array radar; and then constructing a multifunctional radar LSTCN working mode identification network model, inputting the preprocessed training set data into the model for training, and finally sending the test set data into the trained model to complete the multifunctional radar working mode identification test. According to the method, the LSTCN recurrent neural network and the TCN network are combined, an LSTCN network model is provided, the problems that an existing working mode recognition method is incomprehensive in working mode feature analysis and low in recognition rate are solved, simulation results show that in a noise environment, compared with the LSTM network and the like, the LSTCN neural network has better performance, and the recognition efficiency is improved. And compared with an existing network model, the corresponding identification accuracy is higher.

The invention discloses a multi-scale convolution self-adaptive partial discharge fault type identification method based on characteristic parameters and moment characteristics, and relates to the field of discharge fault identification, and the method comprises the steps: obtaining a partial discharge phase three-dimensional statistical graph, and extracting a distribution relation to generate a statistical parameter graph; extracting statistical parameters based on the statistical parameter atlas, introducing preposed partial discharge occurrence judgment parameters to obtain statistical characteristic parameters, training a neural network, and constructing a statistical characteristic parameter model according to a training result; processing the statistical parameter atlas by using a multi-scale convolution technology to construct a moment characteristic model, and respectively deploying the statistical characteristic parameter model and the moment characteristic model to a partial discharge online monitoring and diagnosis platform; and obtaining a partial discharge fault type identification result based on the deployment result, analyzing the partial discharge fault type identification result by using an analytic hierarchy process, and evaluating the health state of the equipment. According to the method, the partial discharge fault type identification rate is improved, the anti-interference capability is improved, and the partial discharge type can be effectively and accurately identified.

The invention discloses a valve side sleeve partial discharge identification method and system based on artificial intelligence, and relates to the technical field of power equipment state monitoring and fault diagnosis, and the method comprises the steps: synchronously collecting multi-source signals; pre-processing and feature fusion are carried out based on the multi-source signals, and a fusion feature matrix of partial discharge identification is obtained; and based on the fusion feature matrix of partial discharge identification, performing partial discharge type identification and classification through an artificial intelligence hybrid model. According to the method, the identification robustness of the model to multi-type discharge behaviors in a complex scene is improved, high-quality input with more complete semantic expression and more reasonable structure is provided for a subsequent artificial intelligence model, accurate identification and stable classification of partial discharge types are realized, the probability of false alarm and missing report is effectively reduced, and the accuracy of partial discharge behavior identification is improved. And meanwhile, the system has online learning and model evolution capabilities, so that the system can still keep a high recognition rate in long-term operation, and the dual requirements of an intelligent power grid on online monitoring accuracy and self-adaptability are met.

The invention provides an active defense-based multi-source log data fusion sub-tree traceability graph attack detection method, which comprises the following steps of: constructing an attack behavior graph according to CTI (Computer Telephony Integration) data, fusing multi-source log data and constructing a traceability graph; constructing a node alignment rule, and performing node alignment on the attack behavior graph and the traceability graph; taking the attack behavior graph after node alignment as a query graph, taking the traceability graph after node alignment as a target graph, and respectively performing subtree construction on the query graph and the target graph; performing subtree isomorphic calculation and recursive updating on the constructed query graph subtree and target graph subtree to obtain a subtree isomorphic matrix, and calculating a final subtree isomorphic index according to the subtree isomorphic matrix; performing similarity calculation on the query graph and the target graph to obtain a generalized index matrix; and attack detection is carried out based on the generalized index matrix and the subtree isomorphic indexes. According to the method, the low time complexity can be guaranteed, meanwhile, the high recognition rate is maintained, and the APT attack detection efficiency and accuracy are remarkably improved.

The invention relates to the technical field of grease detection, in particular to a grease detection data analysis system and method based on multi-source data. The method comprises the following steps: acquiring grease multi-source detection data, and performing near infrared spectrum detection feature extraction according to the grease multi-source detection data so as to obtain infrared spectrum detection data; performing signal enhancement processing on the infrared spectrum detection data to generate enhanced infrared spectrum data; carrying out physicochemical index feature extraction based on the grease multi-source detection data so as to obtain physicochemical index data; performing detection anomaly analysis according to the physicochemical index data and the enhanced infrared spectrum data so as to obtain grease anomaly detection data; and performing type division on the grease anomaly detection data so as to obtain grease chemical component anomaly data and grease microorganism anomaly data. Based on the grease detection technology, the abnormal recognition rate and the quality evaluation accuracy of grease detection are improved.

The invention is suitable for the field of radar signal processing, and provides a multifunctional radar working mode identification method based on a knowledge graph ROTATE model, and the method comprises the steps: randomly generating a radar sample data set according to the preset parameter range and parameter modulation mode of each working mode of a radar; preprocessing the radar sample data set to form a basic unit triple for constructing the knowledge graph; constructing a knowledge graph for describing a relation between a radar working mode and a parameter value and a parameter modulation type based on the basic unit triad; entities and relationships in the knowledge graph are mapped to a low-dimensional continuous vector space for representation, and the confidence coefficient of each basic unit triple is calculated by using a score function defined by an ROTATE model algorithm; and maximizing the confidence coefficient of the basic unit triad to train the model, and performing working mode recognition on the test set by using the trained model. According to the method, the recognition time can be shortened while the high recognition rate is guaranteed, and good generalization is achieved.

The invention discloses a Web interface element recognition method, system and device and a medium, belongs to the technical field of artificial intelligence and Web interface development, and aims to solve the technical problem of how to improve the recognition rate of dynamic elements and CSS hidden elements so as to enable Web interface element positioning to be more accurate. Performing screenshot by using a playwright screenshot function, converting the screenshot into a Base64 coded character string by using a function after the screenshot is completed, inputting the character string and a url into an intelligent agent, calling a DeepSeek-VL2 visual model by the intelligent agent, analyzing the screenshot and outputting identified structured element data; dOM structure analysis: inputting a page url address, and generating DOM structure data by using a DOM analysis function; dOM tree analysis and visual element matching: calling an element matching function, performing visual element and DOM structure matching, and obtaining all successfully matched element data; interface element extraction and positioning generation; and element application.

The invention relates to the technical field of intelligent video monitoring, in particular to a park intelligent video monitoring system and method based on deep learning. The method comprises the following steps: collecting video monitoring deployment data of each camera in a park, and analyzing an overlapping area of a view field of a neighborhood camera pair on the ground and a shortest passable path length to obtain a park monitoring topological graph; performing space-time transition probability processing according to the park monitoring topological graph to construct a space-time trajectory prediction model; and capturing a monitoring target by using a camera, and carrying out intelligent video trajectory prediction by using the spatio-temporal trajectory prediction model to obtain a high-confidence monitoring trajectory. According to the method, high-precision target identity relay and automatic track correction are realized through a feature matching mechanism of a graph convolutional network and occlusion perception, so that the intelligent video monitoring recognition rate and track continuity in a complex environment are improved.

The invention relates to the technical field of environment monitoring and early warning, in particular to a forest fire monitoring and early warning system based on big data, which performs multi-period data superposition calculation in combination with a temperature change difference value, remarkably improves the recognition precision of a suspected point of a fire source, enhances the capture capability of dynamic change of forest environment temperature, and improves the accuracy of forest fire monitoring and early warning. According to the method, the identification rate of potential fire behavior is improved, difference analysis is performed by using a time period highly overlapped with humidity in a temperature trend curve, a high-risk section is accurately marked in combination with the temperature rise rate and the wind direction continuity, the judgment capability of a fire risk range is improved, and the judgment is matched with the change rate of a risk point through path overlapping number judgment. Quantitative control of trigger conditions is realized, so that a risk control decision has multiple response capabilities of spatial dimension, time dimension and dynamic evolution characteristics, and timeliness, spatial accuracy and risk response perspectiveness of forest fire prediction are improved.