137 results about "Change analysis" patented technology

The invention discloses a hardware fault real-time detection method and system based on cooperation of a CPU (Central Processing Unit) and a BMC (Baseboard Management Controller), and the method comprises the following steps: respectively collecting high-frequency state data and tendency indexes by establishing a communication mechanism between the CPU and the BMC; the processor uses a CUSUM algorithm to carry out abrupt change analysis on the periodically collected operation state to generate an abrupt change event; and the management controller uses an EWMA algorithm to model the trend data, and extracts abnormal changes. And the system performs fusion analysis on the mutation event and the trend anomaly to form a fusion anomaly vector. The risk assessment module calculates a risk score based on a preset rule, determines a fault level, positions a target component, and outputs a processing strategy. And triggering a response action according to the strategy and recording an execution state. And the fusion data and the response record are input into the adaptive module together for dynamically adjusting CUSUM and EWMA parameters, so that adaptive updating of the algorithm is realized, and finally a fault detection signal is generated.

The invention relates to the technical field of equipment fault prediction, in particular to an on-line monitoring analysis and fault pre-judgment method and system for a switch cabinet of a switch cabinet, and the method comprises the following steps: collecting a voltage sequence, recognizing a main peak, recording the main peak, extracting a rear section of the main peak, analyzing amplitude change, judging discontinuous features, constructing a track trend sequence, recognizing continuous offset, and carrying out fault prediction. According to the method, through extraction and serialization processing of voltage main peak characteristics in the contact conduction period and combination of comparison of change amplitudes of sampling sections after main peaks, accurate identification of transient anomalies is achieved, and through dip angle change analysis of main peak tracks in the continuous periods, accurate identification of the transient anomalies is achieved, so that the accuracy of the transient anomalies is improved, and the accuracy of the transient anomalies is improved. According to the method, the deviation behavior of the running trend is recognized, abnormal sections which are consistent in direction and are distributed in a concentrated mode are positioned through cross comparison of abnormal periods and track deviation, the deviation trend is comprehensively judged based on the period number and direction characteristics, and the timeliness and accuracy of fault pre-judgment are effectively improved.

The invention relates to the technical field of sensing data processing, and discloses an old people home safety decision service system fusing multi-mode sensing data, which comprises a data interaction analysis unit, a data change analysis unit, a data calculation unit, a safety analysis unit and a safety decision unit. The hidden association of the multi-modal indexes is accurately captured, the early degradation of the collaboration level can be perceived through the graph condensation degree and the graph equilibrium value, the collaboration degradation track can be used for analyzing the collaboration state, and the finally generated safety decision instruction can perform early warning in advance. According to the scheme, single-mode threshold dependence is broken through, hidden dangers hidden in data correlation evolution are recognized in advance, and the lagging of home safety decision of old people is solved.

The invention discloses an embedded system integration test case multiplexing method and system, and the method is characterized in that the multiplexing of a test case is carried out based on the existing software assets: firstly, based on the existing software requirements and the test case, an existing requirement-test case tracking chain is constructed; performing demand change analysis on the new demand activity graph information; performing change influence range analysis based on a change analysis result in combination with a demand-test case tracking chain, and determining a test case reuse and generation scheme; and finally, executing the test case multiplexing and generating scheme. According to the method, the tracking relation between existing requirements and test case assets is established, meanwhile, the execution cost of downstream test tasks caused by requirement change is reduced through multiplexing, and the test efficiency of integration testing is improved.

The invention provides an automatic management interaction method, system and device of a whole vehicle software version and a medium, and relates to the technical field of vehicle software engineering.According to the method, functions of version monitoring, change analysis, test management and version management are provided through an integrated main interface, and a user can check the software version state and tasks to be done in real time; and performing multi-dimensional version comparison and risk assessment to generate a detailed change and risk report. In addition, the method supports configuration of a test strategy, execution of an automatic test and generation of a test report and a defect positioning list, and the test efficiency and accuracy are effectively improved. And in the aspect of version management, the whole process operation from creation to release is realized, the timeline and historical report of the whole life cycle can be displayed, and the transparency and traceability of version control are ensured. On the whole, the method effectively improves the efficiency and reliability of whole vehicle software version management, simplifies the operation process, and enables the version control to be more efficient, more accurate and easier to manage.

The invention relates to the technical field of automobile state monitoring, in particular to an automobile electric drive state intelligent recognition system which comprises a data acquisition module, a change analysis module, an association recognition module, a state layering module and a state recognition module. According to the method, the temperature and flow velocity dynamic states of the cooling liquid at the inlet, the outlet and the middle part of the channel are captured in real time, the time sequence characteristics reflecting the heat dissipation process of the electric drive equipment are established, the dynamic change rate of the electric power of the electric drive equipment is synchronously associated, and the linear coupling strength between the electric power input and the heat flow response output is deeply analyzed; a multi-dimensional joint feature fusing electric power change, heat flow dynamic response and response delay between the electric power change and the heat flow dynamic response is constructed, the similarity between the current operation state and various preset typical working condition features is calculated through clustering analysis, and finally affiliation judgment of the current operation state is output in a probability form. The problem of recognition fuzziness caused by complex and changeable working conditions is effectively solved.

The invention relates to a spoil field three-dimensional form change analysis method, which comprises the following steps: constructing an air-space-ground collaborative multi-scale three-dimensional monitoring network by adopting an unmanned aerial vehicle carrying a laser radar system; dynamic construction of a multi-period digital elevation model and a digital terrain model is realized through a weighted fusion algorithm based on point cloud density difference and an improved progressive triangulation network encryption algorithm; performing time sequence alignment on the digital terrain models in different periods, and constructing a digital elevation difference model; and through a digital elevation difference model and a self-adaptive threshold change detection technology, the space-time differentiation characteristics and evolution rules of a slope curvature sudden change area, a surface roughness abnormal zone and a deformation rate of the spoil site are disclosed. According to the scheme provided by the invention, high-precision three-dimensional form dynamic modeling and deformation process precise monitoring of the spoil ground can be realized, and reliable technical support is provided for ecological safety situation assessment and geological disaster advanced early warning of the spoil ground.

The invention relates to the technical field of air tightness detection, in particular to an air tightness detection method and system for a plastic pipe connecting piece, and the method comprises the steps: obtaining real-time pressure data and associated temperature data of the interior of a detected plastic connecting piece in an inflation and pressure maintaining process; inputting the real-time pressure data into a temperature and pressure coupling model for pressure value compensation correction to obtain a compensated pressure value only reflecting the leakage effect; performing time sequence change analysis on the compensated pressure value, and identifying and determining a starting time point when the pressure enters a stable state; extracting the dynamic attenuation characteristic of the compensated pressure value; and performing matching degree calculation on the dynamic attenuation characteristics and the reference model of the leakage-free basis behavior to generate an air tightness detection result of the plastic pipe connecting piece. According to the invention, the problem of interference of non-leakage factors such as environment temperature in air tightness detection is effectively solved, the inhibition capability of interference of non-leakage factors such as temperature is enhanced, and the accuracy and reliability of a detection result are improved.

The invention relates to an intelligent fire exit door state monitoring method, system and device, and the method comprises the steps: collecting magnetic field distribution data through a magnetic field induction array disposed on the contact surface of a door body and a door frame of a fire exit door, carrying out the change analysis, and obtaining the opening and closing information of the door body; performing signal filtering and feature extraction on the door body opening and closing information to generate door body state information; the door body state information is uploaded to a cloud analysis platform, safety analysis is carried out in combination with fire exit door data, and a door body evaluation conclusion is generated; and generating equipment regulation and control parameters according to the door body evaluation conclusion, and performing parameter optimization adjustment on the magnetic field induction array to obtain optimized monitoring parameters. According to the method, a complete closed-loop system from data acquisition, local processing, cloud analysis to parameter optimization can be constructed, and the accuracy, reliability and operation and maintenance efficiency of state monitoring of the intelligent fire exit door are comprehensively improved.

The invention discloses an automatic optimization system for an administrative approval process, which relates to the field of administrative approval and comprises a twin modeling module, a simulation prediction module, a scheme generation module, a feature modeling module, a dynamic routing module, a semantic analysis module, a change analysis module, a behavior intervention module, a fusion inference module and a decision support module. According to the method, a real process structure and an operation rule can be accurately copied, bottleneck nodes can be identified in different pressure scenes, intervention measures are ensured to consider efficiency, throughput and resource utilization rate, task allocation is efficient and fair by dynamically generating a routing instruction, the response speed and accuracy of approval to regulation updating are improved, and the method is suitable for popularization and application. Quantitative policy implementation effect prediction is provided for decision makers, the scheme is ensured to have transparency and auditing performance, and manual review and risk control are facilitated.

The invention relates to the technical field of medical image processing, in particular to a pulmonary nodule follow-up visit intelligent management system which comprises a pulmonary nodule intelligent matching and change analysis engine, a pulmonary nodule intelligent follow-up visit scheme generation engine, a cloud storage and collaborative service module, an early warning module and a user management and operation module. The surface of the pulmonary nodule is modeled into a differential manifold, accurate matching of the pulmonary nodule and quantitative analysis of small changes are achieved through a multi-scale differential invariant feature extraction framework and a geodesic distance-based change analysis technology, the system generates a personalized follow-up visit scheme based on a risk assessment model, and through a multi-stage early warning mechanism and multi-channel notification distribution, the risk assessment accuracy of the pulmonary nodule is improved. The system also adopts a GPU acceleration computing architecture, so that the processing efficiency is remarkably improved, multi-terminal cooperative operation is supported, the accuracy, intelligence and individuation of pulmonary nodule follow-up visit management are realized, and the follow-up visit efficiency and the early detection rate of lung cancer are improved.

The invention discloses a coastline characteristic change analysis method based on a remote sensing image, and relates to the technical field of coastline monitoring. Comprising the following steps: collecting Landsat TM / OLI data, preprocessing the data, carrying out sea-land classification and water body index extraction on a remote sensing image of a research area by adopting an object-oriented nearest neighbor algorithm, carrying out post-processing such as vector data conversion and shoreline smoothing processing on final shoreline information to obtain a shoreline data set, and carrying out spatial-temporal change analysis. And by combining a plurality of remote sensing image processing technologies and a statistical analysis method, the coastline feature change can be accurately extracted and analyzed, and the influence of human factors and natural environment in a traditional method is avoided.

The invention relates to the technical field of oral medicine and biological information analysis, in particular to an evaluation system for oral microbiological analysis and decayed tooth risk prediction. The system comprises a sample collection module, a microorganism detection module, a data preprocessing module, a microorganism population dynamic analysis module, a multi-dimensional feature extraction module, a decayed tooth risk assessment module and a personalized suggestion generation module. By constructing microbial population evolution analysis based on a Lotka-Volterra competitive model, diversity dynamic change analysis based on Shannon entropy and evolution trend prediction based on a Lyapunov stability theory, the system deeply reveals a dynamic evolution law of an oral micro-ecosystem, and a multi-dimensional feature adaptive weighting mechanism based on mutual information is adopted, so that the dynamic evolution law of the oral micro-ecosystem is improved. Personalized accurate risk assessment is realized, a causal reasoning theory is introduced to generate personalized intervention suggestions with causal support, and the accuracy and the prevention effect of dental caries risk prediction are remarkably improved.

This invention discloses a method and system for intelligent analysis of road and bridge construction data based on big data, belonging to the field of big data analysis technology. The method includes: acquiring road and bridge construction data; grouping the road and bridge construction data to obtain multiple construction data groups; performing change analysis on the construction data groups to calculate the data change rate corresponding to each type of road and bridge construction data; matching the data change rate corresponding to each type of road and bridge construction data with a micro-trend pattern template to identify change pattern combinations; and performing risk assessment on the change pattern combinations to generate early warning information, wherein the early warning information includes the affected structural parts, the change pattern combinations, and intervention measure suggestions. This invention can combine construction data grouped from different regions and pattern templates to identify change pattern combinations and generate early warning information, thereby realizing road and bridge construction data analysis and improving comprehensiveness, accuracy, and road and bridge safety.

The application discloses a kind of based on flavoromics's sun vinegar aging period flavor dynamic change analysis method, it is related to flavoromics technical field, to solve the current concentration rainstorm causes the temperature and humidity gradient of inner and outer wall of pottery jar to increase dramatically, causes the technical problem that pottery jar astringency is generated, including the following steps: S100, real-time monitoring and risk assessment of heat and humidity stress;S200, rainstorm event fusion early warning;S300, dynamic deduction and strategy optimization based on digital twinning;S400, dynamic microenvironment blocking control execution;S500, closed-loop feedback based on proxy index;S600, post-evaluation and model optimization.The application realizes early quantitative perception to astringency risk by real-time monitoring and calculating heat and humidity stress index representing ion precipitation driving force, combined with the research and judgment model of early warning confidence of fusion meteorological forecast and real-time data, starts the blocking control system, carries out physical intervention to pottery jar, guarantees the quality stability of sun vinegar in rainy season.

The invention relates to the technical field of vibration feature recognition, in particular to a method and device for recognizing the upper step or lower step vibration trend of large rotating equipment. The method comprises the following steps: acquiring monitoring data of a vibration measuring point of to-be-monitored large rotating equipment; starting from an initial data point in the monitoring data, sequentially selecting continuous data points of which the number is gradually increased to form a plurality of data subsets; for each data subset, performing linear regression fitting according to the data subset, and calculating a root-mean-square error corresponding to the data subset based on the obtained regression equation corresponding to the data subset; sequentially comparing the root-mean-square error corresponding to each data subset with a preset root-mean-square error reference value so as to identify a trend abrupt change point; and according to the monitoring data, carrying out local change analysis and front and back data segment stability analysis on the trend abrupt change point, and carrying out upper step and lower step trend identification on the monitoring data based on the obtained analysis result to obtain a trend identification result.

This invention discloses a multi-level network-type energy storage control and protection system and method, relating to the field of energy storage technology. The method includes: S1, performing feature modeling on the multi-source operating signals of the acquired energy storage system to construct a multi-source operating feature set; S2, performing time series rate of change analysis on the multi-source operating feature set to extract dynamic features of current direction change and power reversal; S3, performing energy disturbance amplitude analysis on the dynamic energy feature set to calculate an adaptive protection threshold set in real time; S4, comparing the real-time sampled signal with the adaptive protection threshold set to identify whether an over-limit state has been triggered; S5, when an over-limit state is detected to be triggered, extracting the real-time dynamic features of the over-limit event, matching the real-time dynamic features of the over-limit event with the energy backlash judgment model, determining the anomaly type of the over-limit event based on the matching result, and outputting a protection action command or dynamic balance control command corresponding to the anomaly type.

Since IP network topology often has a clear hierarchical structure, using community detection algorithm in graph theory can divide the topology into several regions with high cohesion and low coupling, and selectively spread the routing update in the region to other regions, which can effectively avoid the calculation amount of full routing simulation, and realize fast and accurate IP network routing simulation reasoning. The application proposes an IP network routing simulation reasoning method based on community detection, which first divides the whole network into multiple cluster structures based on the community detection algorithm, and the shortest path between the nodes in the cluster does not pass through other clusters, and uses this division result as the basis for calculating the routing update. Then, the change operation of the network topology is accurately identified, and according to the network elements involved in the change point and the cluster information to which they belong, the network element node pairs that need to be recalculated are further arranged, and the SPT tree of the node pairs is recalculated and the routing update is performed. At the same time, an IP network routing simulation reasoning device based on community detection is proposed, which includes a topology management module, a network configuration module, a network segmentation module, a change analysis module and a routing calculation module, to realize the topology segmentation and routing simulation reasoning process of the IP network.

The invention relates to the technical field of intelligent teaching evaluation, in particular to an education quality intelligent analysis and evaluation system which comprises an image area parameter extraction module, a drawing feature change analysis module, a main color distribution recognition module, a drawing expression repetition detection module and a theme preference affiliation evaluation module. According to the invention, on the basis of an image grid division and feature extraction mode, fine parameter acquisition of the drawing image in the spatial dimension can be realized, and in combination with feature time sequence construction, the change trend of dimensions such as the graph area, the edge line and the pixel density can be dynamically tracked; the color expression analysis depth is improved through linkage recognition of the color proportion and main color space distribution, quantitative description of painting behavior continuity and stability is achieved through repeated comparison of expression areas and feature synchronization rate evaluation, and finally the affiliation probability of an image in a subject category is output through multi-dimensional feature clustering. And data support is provided for the drawing preference and expression orientation trend of students.

The invention discloses a training method for agent template pair training and automatic evaluation, relates to the technical field of agent training, and is used for solving the problem that the training efficiency of agent pair training is low due to the fact that behavior response characteristics and a cognitive evolution process of an agent in the actual use process cannot be effectively reflected. The method comprises the following steps: receiving user input through an interaction end, recording agent response time, setting a learning time reference, obtaining processing time, calculating a processing learning ratio by utilizing the processing time and the learning time reference, classifying agent states in combination with the response time, and judging whether to enter a pair training; detecting a user instruction, calling an instruction answer benchmark, recording an agent answer, setting recording time and generating a training score within the time after the user enters a pair exercise; the memory occupation is detected in real time, the inscribing variation is calculated, the training effect is analyzed in combination with the training score and the memory change, and whether the training transfer is performed or not is judged according to the analysis result, so that the agent training efficiency is improved.

The invention discloses a lake water ecological restoration monitoring method based on big data, and relates to the technical field of submerged plant monitoring, and the method comprises the following steps: collecting original water quality data of a to-be-restored lake area before the to-be-restored lake area is restored by using submerged plants, and collecting initial water quality data of the to-be-restored lake area after the to-be-restored lake area is restored by using the submerged plants; carrying out correlation analysis treatment to obtain related water quality physicochemical indexes; collecting water quality physicochemical data of related water quality physicochemical indexes of the lake area to be restored, and performing water quality physicochemical change analysis to obtain plant related change data; monitoring and analyzing the submerged plants in the lake area to be restored based on the related change data of the plants to obtain related change information of the submerged plants; the method is used for solving the problem that the development condition of submerged plants cannot be accurately obtained through water body parameters on the premise of not damaging the submerged plants when the submerged plants are monitored at the initial stage of repairing lake water ecology by using the submerged plants in the existing submerged plant monitoring technology.

The invention relates to a data change management method, device and equipment for a financial system and a medium, the method can be applied to change processing of system data in a financial science and technology scene, and according to historical data flow direction information of known data fields, a flow path link corresponding to the known data fields in the financial system is constructed; training a preset data change analysis model by using all the data fields and the corresponding flow path links to obtain a trained data change analysis model; inputting a to-be-changed data field into the trained data change analysis model, and outputting an influence range corresponding to the to-be-changed data field; and determining a data change strategy according to the influence range of the to-be-changed data field, and performing data change synchronization on data of the to-be-changed data field based on the data change strategy. According to the invention, the problem of inconsistent use of data in the financial system due to lack of an efficient data change notification mechanism in the existing financial system is solved.

This invention relates to the field of tunnel construction technology, and more particularly to a system and method for dynamically optimizing tunnel construction parameters based on geological information from point cloud data. At multiple monitoring time points, point cloud data is collected from the tunnel segment to be constructed, obtaining multiple point cloud datasets. The tunnel segment is divided into multiple construction areas, and multiple regional point cloud datasets are obtained from these datasets. Point cloud change analysis is performed on each regional dataset to obtain multiple point cloud change rates. Based on these regional point cloud datasets, reflection intensity analysis is conducted to obtain multiple reflection intensity coefficients. Combined with the multiple point cloud change rates, geological analysis is performed to obtain the distribution of concealed geological information. Based on this distribution, tunnel construction parameters are dynamically optimized to obtain optimized tunnel construction parameters for tunnel construction. This achieves the technical effect of improving the efficiency and safety of tunnel construction while reducing construction risks and costs.

A multi-parameter feedback-based intelligent control method and system for aquaculture wastewater, relating to the field of wastewater treatment and control technology, includes: acquiring key water quality monitoring indicators for each treatment subsequence; real-time monitoring of each treatment subsequence; critical change analysis of historical monitoring periods marked as abnormal for the treatment subsequences, obtaining critical feature sets for historical monitoring periods, comparing significant differences in the critical feature sets, and constructing a critical correlation feature network of abnormal indicators for abnormal historical monitoring periods; pseudo-correlation analysis of the critical correlation feature network, obtaining pseudo-correlation nodes and interference nodes, and constructing a critical directed acyclic graph and critical interference link graph of abnormal indicators; constructing a multi-device collaborative control model based on the critical directed acyclic graph and critical interference link graph of several abnormal indicators, outputting specific control parameters of the associated devices for each treatment subsequence and performing preventive control to achieve efficient, economical, and stable treatment of aquaculture wastewater.

The invention relates to the technical field of data processing, in particular to a data quality analysis method, system and device and a storage medium. The data quality analysis method comprises the steps of obtaining training data; inputting the training data into a neural network model to execute a model training operation so as to obtain data processing information of the neural network model on the training data, the data processing information comprising a blood relationship graph of the training data and quality information of the training data in each processing link of model training; inputting the blood relationship graph of the training data and the quality information of the training data in each processing link of model training into a data quality change analysis model, and outputting a quality change result of the training data in each processing link of model training by the data quality change analysis model; and determining a data quality analysis result according to the data quality change result. According to the method, the specific processing link causing the quality reduction of the training data can be quickly positioned.

The application relates to the technical field of building management, in particular to a building abnormal change identification method and device based on an intelligent algorithm, which comprises a building abnormal change identification center, an abnormal change analysis unit, an abnormal change division unit, a cause analysis unit, an abnormal change prediction unit and a visual response unit; the application preliminarily analyzes point cloud data at different time points to determine whether abnormal changes occur in a building, meanwhile, the abnormal building is distinguished in a color marking mode, so that the building which has abnormal changes in a target area can be directly and intuitively understood, the abnormal building is analyzed in an information progressive mode, so that it can be directly and intuitively understood whether the abnormal change of the abnormal building is caused by a sudden event or a slow long-term change accumulation, and the sudden abnormal change and the long-term accumulated abnormal change are analyzed, so that the management efficiency of the sudden abnormal change of the target building is improved, and the structure state in the target building is understood based on the strain change trend, so that advanced early warning management is facilitated.

The present invention relates to a device for automatically detecting changes in a contract and assessing risks, which may be used in the fields of legal risk management, contract review, compliance monitoring, and the like and, more specifically, to a device for automatically detecting changes in a contract and assessing risks that utilizes artificial intelligence (AI) and legal data analysis technologies to automatically detect changes in a contract and evaluate legal and business risks of the changes. A device for automatically detecting changes in a contract and assessing risks, according to one embodiment of the present invention, comprises: a content classification module; a change detection module; a risk assessment module; and a change analysis and reporting module.

The invention provides an airborne system ICD change method and system, and relates to the field of airborne systems.The airborne system ICD change method comprises the following steps that S1, an airborne system ICD change management environment is initialized, and an interface dependency relationship model is established; s2, receiving a change request, analyzing the change request, and identifying a change type; s3, calling the interface dependency relationship model, executing change influence analysis based on the change type, and generating a change influence report; s4, performing change review and implementation based on the change influence report to obtain a change result; and S5, verifying the change result, and generating a change closed-loop report. The problems that in the ICD changing process of an existing airborne system, the flow is not standard, changing analysis is not comprehensive, different types of ICD changing lacks a targeted method, and changing system collaboration is insufficient are solved, the scientific, controllable, efficient and accurate changing method and system are provided, and the compatibility and reliability of the changed system are guaranteed.

This invention discloses a wetland landscape pattern classification method based on optimized feature parameters. First, multispectral and topographic data of the target wetland are preprocessed to extract feature parameters from the preprocessed data. Wetland landscape categories are defined, identifying 13 land cover types. Training samples are determined by referencing field survey data, original multispectral image data, principal component analysis (PCA1) of original remote sensing data, and the tasseled cap transformation brightness component of original remote sensing data. For the preprocessed multispectral data, wetland information extraction is performed using both maximum likelihood classification and decision tree classification. The classification results from the two extractions are combined with the extracted feature parameters to form fused data. Information extraction based on maximum likelihood classification is then performed on the fused data to obtain dynamic change analysis results of wetland landscape classification. Combined with optimized feature parameters, multiple feature parameters are fused to achieve information extraction from remote sensing images and refined classification of wetland landscapes.

The invention relates to the field of fan control algorithms, in particular to a fan non-stop routing inspection blade tracking algorithm and system, and the algorithm comprises the steps: S1, tracking a fan in a fan image in a subsequent frame through a long video object segmentation model XMem, and carrying out the semantic segmentation to generate a mask graph; s2, the longest intersecting line section of the vertical line of the adjacent points in the boundary point set and the contour and a corresponding vertical line equation are calculated and screened out, two vertical lines meeting preset conditions are determined, the intersection point is determined, and the intersection point is the center point of the draught fan; and S3, carrying out position offset and change analysis on each blade key point so as to realize accurate tracking of the blade key points. According to the method, the fan blade can be accurately segmented and tracked, the fine positioning of the blade is realized by accurately extracting the edge, key point and form information of the blade, the accuracy and stability during tracking are improved, shutdown and manual intervention are avoided, the workload and maintenance cost of manual inspection are reduced, and the working efficiency is improved. And misjudgment and risks caused by human factors are reduced.