140 results about "Track analysis" patented technology

The invention relates to the technical field of directional drilling track control, and discloses a coal mining underground directional drilling track control method and system, and the method comprises the steps: carrying out the fusion track analysis of a drilling machine based on sensor data, drilling parameters and pre-obtained geological hardness parameters, and obtaining a reference track of the drilling machine, a sensor, an azimuth angle sensor and a bit pressure sensor are used for synchronously collecting drill bit attitude data of a drilling machine to generate a real-time track of the drilling machine, then track offset calculation is carried out to obtain inclination angle deviation and azimuth angle deviation, and dynamic gain adjustment analysis is carried out on the drilling machine to obtain adjustment parameters of the drilling machine. And finally, the drill bit posture of the drilling machine is adjusted based on the adjusting parameters. According to the method, the directional control accuracy of the underground drilling track can be improved.

The embodiment of the invention discloses a terminal application behavior data analysis method and computer equipment, and the method comprises the steps: obtaining an interaction behavior data set of an RPA terminal application in a preset time window, the interaction behavior data set comprising a plurality of interaction behavior sequences, each interactive behavior sequence comprises at least one user interface operation event and a corresponding system response event; performing behavior track analysis processing on the interactive behavior data set, and extracting an interface circulation feature and an operation intention feature of each interactive behavior sequence; generating an interface optimization strategy according to the relevance between the interface circulation feature and the operation intention feature, wherein the interface optimization strategy comprises an interface layout adjustment rule and an interface response logic optimization rule; and feeding back the interface optimization strategy to a background management system of the RPA terminal application. Therefore, the problem that an existing RPA application interaction interface is insufficient in design and optimization can be effectively solved, and the usability of the interface and the response capability of the system are improved.

The invention provides an unmanned aerial vehicle flight attitude self-adaptive imbalance control method, device and equipment, and aims to accurately obtain an ideal control track and recognize a beneficial imbalance mode to generate a controllable imbalance strategy library by constructing a reverse track analysis and controllable imbalance strategy generation mechanism. A dynamic balance reference establishment and track reconstruction technology is adopted to construct an unbalance-recovery control track, and dynamic response characteristics are accurately obtained; an unbalance recovery space-time manifold construction and topology analysis technology is introduced, a core unbalance control feature set is extracted through space-time folding transformation, and unbalance control key parameters are determined; establishing an unbalance-recovery mapping rule and self-adaptive parameter fusion mechanism, balancing the control effect and the safety requirement, and generating a disturbance control parameter and a self-adaptive mapping parameter; a real-time control output regulation mechanism is constructed to generate a flight control signal in combination with reverse unbalance combined control and a multi-axis cooperative control mechanism; finally, a self-adaptive unbalance control system is formed, and precise unbalance control over the attitude of the unmanned aerial vehicle is achieved.

The invention discloses a cross-modal large model construction method and system based on track spatio-temporal characteristics, and belongs to the crossing field of artificial intelligence and dynamic spatio-temporal data processing, and the method comprises the steps: carrying out the sliding sampling and spatial distribution difference judgment through the semantic dynamic segmentation of multi-scale track spatio-temporal data, and generating spatio-temporal data blocks with consistent semantics; designing a space-time encoder of a hybrid architecture, extracting track time sequence association and spatial features, and unifying dimensions; constructing a text space-time fusion mechanism, and dynamically adapting cross-modal features by means of an anchor interface and gating fusion; a staged instruction fine tuning strategy is adopted, semantic alignment of space-time and text features is optimized firstly, then model top-layer parameters are trained cooperatively, complex scene adaptation is enhanced in combination with instruction difficulty progression and hard sample mining, and space-time constraint regular terms are introduced to guarantee output rationality. According to the method, high-precision cross-modal reasoning capability is provided for scenes such as track analysis and track prediction.

The invention discloses a one-key performance test method, medium and equipment based on JMeter and template configuration, and the method comprises the steps: obtaining a parameterized performance test template from a template configuration center in response to a test instruction, injecting runtime parameters through a template engine, and generating an executable JMeter test plan; parsing the plan to identify a load pressure demand, thereby dynamically scheduling resources from a distributed load injector cluster and distributing a test plan; uniformly triggering test execution on the cluster, and synchronously acquiring a performance data stream from the JMeter and a resource data stream of the tested system; and finally, timestamp alignment and fusion are carried out on the two types of data streams, intelligent diagnosis is carried out on a formed monitoring data set based on a diagnosis rule base or an intelligent analysis model, and a visual test report containing bottleneck positioning, resource association and link tracking analysis results is generated. According to the invention, full-process automation and intelligentization of performance testing are realized, and the testing efficiency is remarkably improved.

The invention relates to the technical field of road surface monitoring, in particular to a sensor-based road surface quality detection system, which comprises a disturbance extraction module, a trajectory analysis module, a consistency identification module, a sudden change sorting module and a crack confirmation module. According to the invention, through the time sequence correlation analysis of the vertical structure vibration acceleration, the GPS direction vector and the transverse angular velocity signal in the vehicle driving process, the disturbance behavior and the trajectory deviation trend change in the vehicle and road surface contact process can be identified, and after the structure disturbance and the trajectory variation are correlated and compared, the stability of the vehicle is improved. Further screening real disturbance response from change consistency, constructing a space transfer model of crack mutation in combination with a joint state structure of disturbance intensity and offset amplitude, identifying a joint abnormal interval of structural damage and trajectory deviation in time and space, and determining a joint abnormal interval of structural damage and trajectory deviation in time and space through a multi-dimensional index distribution density and time continuity discrimination standard. And accurate judgment and spatial calibration of the crack area are realized.

The invention relates to a method, device and equipment for detecting shopping settlement fraud risk behaviors and a medium, and the method comprises the steps: obtaining a plurality of continuous video frames, and detecting a human body bounding box and a posture key point of a character object in the video frames, thereby obtaining a posture detection result; determining an upper body bounding box according to the human body bounding box, and tracking the upper body bounding box in a plurality of continuous video frames to obtain a motion track; according to the posture detection result and the motion track, determining whether the character object triggers a shopping behavior through track analysis and posture analysis, and if the character object triggers the shopping behavior, outputting and storing a detection record for a shopping behavior triggering event of the character object to obtain a shopping behavior detection result; and under the condition that the shopping behavior detection result represents that a shopping behavior trigger event exists, determining a wrist area image according to the wrist key point, performing commodity identification on the wrist area image through open set target detection, outputting and storing a commodity identification event record, and obtaining a shopping settlement detection result.

The invention discloses a DC power transmission fault diagnosis method and system based on multi-modal data analysis, and relates to the field of DC power transmission line fault diagnosis, and the method comprises the steps: starting temporary line protection measures when a DC power transmission line has abnormal fluctuation, and importing parameters into a data processing center for feature extraction; obtaining fault types corresponding to the effective features; and performing statistics on each associated feature of each fault type to obtain a fault diagnosis result. According to the invention, the multi-modal parameters of the DC transmission line are monitored in real time, and electromagnetic interference is eliminated, so that the accuracy of fault diagnosis is improved, the reliability of data quality is ensured, and the precision of fault diagnosis is improved. And monitoring and protection strategies are adjusted, so that the optimal operation guarantee is achieved. And through dynamic monitoring and tracking analysis of potential faults, the stability and reliability of the system are improved.

The invention discloses an e-commerce platform advertisement effect tracking analysis method based on a big data technology, and the method comprises the following steps: S1, collecting user data, and constructing a user behavior graph; s2, introducing an advertisement intervention path, comparing propagation states with or without advertisement intervention, calculating a disturbance consistency value and labeling an edge weight, and forming a structure chart; s3, inputting the structure diagram into a diagram causal reasoning network, and outputting advertisement individual effect estimation; s4, constructing a behavior disturbance sampler, generating a plurality of anti-fact behavior path training samples, and improving the causal fitting capability of the model; s5, executing path-level attribution on the actual user behavior graph by using the trained model, and outputting a causal weight matrix; and S6, generating an optimized advertisement putting strategy according to an attribution result. The advertisement causal attribution precision is improved, the putting strategy is optimized, and the conversion efficiency and the budget utilization rate are remarkably improved.

The invention relates to the technical field of battery pack capacity attenuation analysis, and discloses a battery pack capacity attenuation characteristic decoupling analysis method, which comprises the following steps: planning an analysis process based on battery pack design parameters and working conditions, determining electrochemical impedance spectroscopy content and cyclic aging test content, and determining different discharge rate types; generating a capacity attenuation analysis process set, determining an execution sequence according to an optimal matching principle, and generating a spectrogram acquisition setting list and a test scheme list of each magnification; and finally, associating to generate an overall analysis scheme. And the analysis content is updated through real-time operation data, the process execution sequence is adjusted, and a database is established for data management and long-term tracking analysis. According to the method, decoupling analysis of the capacity attenuation characteristics of the battery pack is realized, the analysis accuracy and reliability are improved, and technical support is provided for optimization of the battery pack.

The invention discloses a cold-chain pork quality dynamic evaluation method and system based on multi-source data fusion, and relates to the technical field of data processing and intelligent evaluation, and the method comprises the following steps: collecting multi-source signal information of a whole cold-chain transportation process, sorting according to a unified time sequence, continuously arranging all signals on a time axis, and carrying out data processing on all the signals; a complete dynamic record is formed; and extracting each time interval and signal amplitude change track from the formed dynamic record, analyzing a signal section with repeated wave crests in a short time, and generating a preliminary data manuscript for reflection interference judgment. According to the invention, through time sequence arrangement and dynamic association of multi-source signals, a wave crest track analysis mechanism is constructed, real-time identification and correction of reflection interference are realized, and continuity and accuracy of a gas concentration curve are ensured; and by generating a reflection source position list and adjusting numerical value distribution of repeated fragments, false fluctuation is eliminated, and the stability and reliability of cold-chain pork quality evaluation are improved.

The embodiment of the invention discloses a digital management method and system for a main and distribution network project. The method comprises the following steps: performing project information structured modeling on a full life cycle of a target main and distribution network project to obtain a project stage node index model; chain type tracking analysis is carried out on the engineering stage nodes, an information interaction rule and a data transmission path are established, and a data flow topological graph is obtained; extracting front-end node state layer information and rear-end engineering monitoring layer information based on a front-end and rear-end separation strategy; and constructing a multi-dimensional influence factor association network by using the information to perform risk evolution path deduction and risk event simulation to obtain a simulation result and generate a cross-node cooperative response strategy. According to the embodiment of the invention, through the digital support of the node chain type tracking and the front and back end separation architecture, the data island of each stage of the main and distribution network project is eliminated, the cooperation efficiency of the project management is improved, and the controllability of the whole life cycle process is enhanced.

Systems and techniques for generating track geometry data (TGD) for railroad track analysis. In embodiments, rail profile data for each of the rails of a track may be obtained at regular intervals as a vehicle carrying the system traverses the track. The rail profile data may be validated, and reference points on the profiles of each of the rails corresponding to specific geometric features of the rails may be identified. The rail profile data may be filtered and processed to generate track geometry data. The track geometry data may include data for a plurality of data channels. The track geometry data may be transmitted to a backend system receives for further processing to determine rail defects, and to generate notifications and alerts based on the determined defects. The system utilizes machine learning algorithms and models for defect detection and validation.

The invention relates to the technical field of ship safety, in particular to a ship navigation monitoring management system and method based on edge computing, and the system comprises seven cooperation modules: a data collection module synchronously obtains ship multi-source data, and a processing storage module achieves the data compression, marking and storage. The encryption security module completes full-process AES-256 encryption through a hardware encryption chip, the dual-power supply module guarantees stable power supply, the satellite positioning module obtains accurate positioning through Beidou / GPS dual modes, the self-inspection maintenance module monitors faults in real time, and the AI edge calculation unit achieves channel object recognition, track analysis and violation warning through a deep neural network model. According to the method, closed-loop processes of data acquisition, encryption, processing, intelligent analysis and fault handling are completed through module cooperation; the invention aims to solve the problems of high monitoring delay, insufficient data security, unreliable power supply and the like of the existing system.

A distributed link tracking, analysis method and server, global scheduler, in the process of processing user request, when producing detail event, the meta event for detail event search is produced, the meta event contains tracking identification and event production time information, and the data quantity is less than the data quantity of corresponding detail event; The server saves the detail event and the meta event produced separately in local. The global scheduler in the distributed cluster sends a link query request to the server in the distributed cluster, receives the event information related to the tracking identification returned by the server after local search; and based on the received event information, the user request is analyzed by link. The application adopts local storage, on-demand analysis, removes the central warehouse and large database, solves the problems of high cost of data storage, analysis and network bandwidth, and scalability.

The invention provides a sea surface target stable handover and track tracking method and system based on a navigation radar. The method comprises the following steps: the navigation radar performs multi-target stable tracking by using sea surface target echoes; analyzing and screening, and appointing important targets which need to be focused; photoelectric tracking target parameters are calculated by utilizing a sea surface dynamic constraint equation or different-station deployment longitude and latitude conversion according to radar and photoelectric positions; and a photoelectric tracking target is guided, and the track of the target is dynamically and finely adjusted to realize stable tracking of a stable target image. Due to the fact that coordinates of targets measured by the radar are different, each datum cannot be converted to three-dimensional target tracking, and position conversion of the targets cannot be achieved. According to the method, under the condition that a sea surface target lacks measurement dimensions, mutual conversion between measurement coordinates can still be achieved by means of a constraint equation and longitude and latitude conversion, and rapid guide tracking of the target is stably achieved.

The invention discloses a mechanical arm anti-collision analysis system based on an explosive ordnance disposal robot, and relates to the technical field of public safety, the mechanical arm anti-collision analysis system comprises a multi-source heterogeneous data acquisition module, a space-time alignment characteristic distillation module, an attribute-track coupling prediction module and the like, and precise anti-collision is achieved by fusing obstacle attributes and tracks and gradient protection boundaries. The system has the advantages that attribute data such as obstacle material components and mechanical arm state data are synchronously obtained through the multi-source heterogeneous data acquisition module, features are extracted through fusion of the space-time alignment and feature distillation module, and a standardized data set containing rigid, explosive and quality tags is generated; and the attribute tag is incorporated into a state space by the obstacle attribute-track coupling prediction model, and the obstacle attribute and the motion track are associated to carry out collaborative analysis, so that the defect of separation of dynamic obstacle attribute and track analysis in the prior art is effectively overcome, and an accurate attribute-track association basis is provided for anti-collision analysis.

The invention discloses a personalized virtual-real interaction cognitive game system based on AI adaptive brain cognitive enhancement, comprising: a cognitive game design module for designing a plurality of cognitive tasks corresponding to attention, memory, execution function and perception function based on a dynamic difficulty adjustment algorithm according to user characteristics; the dynamic difficulty adjustment algorithm is used for dynamically adjusting the difficulty of a specific level task according to the behavior state and the psychological state of the user; the cognitive task guiding module interacts with the user through a multi-mode guiding mechanism to guide the user to complete a cognitive task; the data tracking analysis module is used for recording the performance and growth trajectory of the user in the completion process of each cognitive task; the instant feedback module provides feedback based on an instant feedback generation algorithm; the cognitive competence evaluation report module is used for generating a corresponding cognitive competence evaluation report according to the performance and data tracking analysis algorithm; and the data storage module is used for storing behavior data, emotion records and cognitive competence evaluation reports of the user.

The invention relates to the field of electric vehicle mileage prediction, in particular to a method and system for predicting the vehicle driving mileage based on real-time data and electronic equipment. The system comprises a battery charging and discharging period identification module, a data extraction and cleaning module, a track segment identification and distance calculation module, an energy consumption behavior modeling and mileage coefficient calculation module, a sample screening and aggregation module and a remaining mileage prediction output module. According to the method, the battery working state, the GPS track and the historical charging and discharging behavior are used as basic data input, multi-layer data processing, track analysis, energy consumption modeling and prediction output are used as intermediate processing paths, and high-precision prediction of the mileage of the electric vehicle under different batteries, different environments and different working conditions can be achieved.

The invention provides a server cluster load balancing state tracking analysis method, and relates to the technical field of servers, and the method comprises the steps: collecting a load migration triggering moment, the number of control messages generated by single migration, and the number of active connections to be migrated of each server from each server in a cluster, meanwhile, the bandwidth bearing upper limit of a network control channel is obtained, migration triggering events are arranged according to the time sequence, and a list of the number of triggered migration servers is obtained; the migration scheduling strategy of each server in the load balancing system is updated by adopting the target migration coordination delay interval, a migration rhythm coordination confirmation signal is obtained by monitoring the network congestion change, and whether the cluster load balancing state is recovered to a safe range is verified by combining a high-risk time period list, so that the stability and efficiency of load migration are improved, and the load balancing efficiency is improved. And a reliable flow coordination guarantee is provided for a large-scale cluster system.

The invention discloses a network service recommendation method and system based on user behavior track analysis, and relates to the technical field of data analysis, and the method comprises the steps: collecting a power interaction behavior data sequence, and carrying out the analysis of a user behavior track; user power demand interactive capture prediction is carried out on the temporal behavior track characteristics and the spatial behavior track characteristics obtained through analysis, and a target user power demand prediction result is obtained; performing recommendation scheme analysis on the power network service platform to obtain an initial network service recommendation scheme; and obtaining a real-time position of a target user, and performing scheme deepening analysis in combination with the initial network service recommendation scheme to determine a target network service recommendation scheme. According to the method, the technical problems that user behaviors and demands cannot be accurately captured and recommendation services are lack of individuation and real-time performance in the prior art are solved, and the technical effects of accurately predicting user power demands and providing real-time and personalized power network service recommendation are achieved.

The application discloses a food full-link quality deterioration tracing method and system, a medium and a product, and relates to the technical field of food quality tracing.The method is implemented, and a sensory description label in consumer feedback information is extracted by a tracing system; with the aid of a preset food quality knowledge graph, accurate mapping of subjective sensory information and objective physical and chemical indexes is realized, and the problem that subjective feedback and objective data are disconnected in a traditional tracing system is solved.The tracing system carries out dynamic deduction on food full-link monitoring data through a preset quality deterioration dynamics model, and upgrades food quality deterioration from static threshold judgment to dynamic numerical evolution track analysis, so that the cumulative deterioration effect of each circulation link can be quantified.The tracing system screens food quality deterioration responsibility links based on average deterioration rate and link deterioration increment, traces deterioration inducement of the responsibility links in combination with the preset food quality knowledge graph, accurately defines the deterioration responsibility and traces the reasons, and improves the tracing efficiency and accuracy.

The invention discloses an electric bed intelligent test method and system based on visual analysis, and the method comprises the steps: carrying out the multi-part spatial positioning and displacement track analysis of an electric bed through a machine vision module in a first test period, generating displacement monitoring data, and evaluating a deviation node; motor operation parameters are synchronously collected, and a multi-dimensional data item set is constructed in combination with displacement data; an FP-Growth association algorithm is adopted to mine an implicit association rule between displacement abnormity and motor parameters, when it is judged that a specific displacement deviation is associated with motor state parameters, an abnormal association node is generated, an abnormal displacement-motor parameter group is defined, and finally a subsequent test scheme is dynamically optimized based on the abnormal characteristic parameter group. And a closed-loop test process of'detection-diagnosis-adaptation 'is formed. According to the invention, adaptive adjustment of the test scheme is realized, and the intelligent degree of quality detection of the electric bed and the tracing efficiency of abnormal factors are remarkably improved.

The invention provides a dynamic passenger flow track analysis method and device, electronic equipment and a medium, and relates to the technical field of data analysis. Comprising the following steps: mapping passenger flow data of every preset time slice into each grid, and generating a plurality of OD matrixes every day; analyzing a flow direction line of the OD matrix of each time slice; connecting a plurality of flow direction lines end to end according to the slicing time to obtain a line dynamic track; searching a plurality of hot tracks based on the line dynamic track to form a hot track set; and optimizing the hot track set by using a full-connection neural network FNN. The real-time hot passenger flow route is generated by updating the passenger flow data in real time. Any threshold does not need to be set in advance, construction can be carried out adaptively according to data characteristics, and the hot route generation precision is improved. The data is processed by setting rules, the hot route generation precision is improved, the influence of noise data is effectively reduced, and the reflecting capacity of the model to the actual situation is improved.

The invention discloses an intelligent video analysis system based on multi-mode AI, and belongs to the technical field of video analysis, and the system comprises a video preprocessing module which is used for extracting an analysis frame of an original video segment; the IFS and Tracking analysis module is used for carrying out cross-frame tracking on the analysis frames and screening important frames; the task arrangement and collaboration module is used for realizing state management and control and progress quantification by relying on Redis, and triggering a corresponding module to carry out parcel logo or face recognition if parcels or figures appear in an analysis frame; the logo identification module is used for screening the optimal frame from the analysis frame with the parcel, and determining the logo type of the parcel in combination with target detection and GPT dual identification; the face recognition module is used for carrying out face detection on the analysis frame with the appearance of the person and recognizing the identity of the person; and the structured video abstract generation module is used for generating a structured abstract of the original video segment. According to the method, the video can be efficiently processed, the recognition precision of the target in the video is improved, and subsequent retrieval and search are facilitated.

This invention relates to a method and system for analyzing test data of tool-type steel pipe truss composite slab structures, belonging to the field of testing and inspection technology for prefabricated building structures. The method includes: data parsing and layered storage, differential internal force calculation, recursive solution of isolated body forces, determination of effective flange width, and construction and comparative verification of beam element finite element models. The present invention uses a combined section co-force method to solve for the effective flange width, overcoming the shortcomings of traditional methods that rely on empirical formulas from specifications. This invention establishes a dual-track analysis method for isolated body analysis and verification of the overall beam element finite element model of composite trusses under asymmetric constraint boundaries. This method can be extended to the stress analysis of other steel truss composite slabs in similar engineering scenarios during the construction phase, providing a quantitative analysis tool from test data to design parameters for the engineering design of unsupported composite slabs.

The invention provides a power conductor galloping track extraction and vibration parameter automatic fitting method, and relates to the technical field of power transmission line state monitoring, and the method comprises the steps: collecting and preprocessing a visible light video of a power transmission line, and obtaining an enhanced image; performing Canny edge detection, contour extraction and screening on the enhanced image to obtain a lead area; determining the observation position of each node of the power transmission line in the lead area; establishing an ID-frame-coordinate mapping relation based on multi-point trajectory tracking of nearest neighbor distance matching in combination with observation positions of nodes; and fitting the galloping trajectory of the power transmission line through a trajectory analysis method based on the combination of frequency spectrum estimation and nonlinear fitting in combination with an ID-frame-coordinate mapping relationship. According to the technical scheme, the position of the wire node can be stably identified under different illumination and background conditions, and the motion trail of the wire node can be automatically tracked and extracted.

The application discloses a kind of based on multi-class deviation's magnetic levitation track smoothness evaluation method, medium and equipment, belong to magnetic levitation track analysis field.This method uses hierarchical amplitude evaluation criterion and hierarchical mean evaluation criterion, assesses the smoothness state of target track, wherein determining hierarchical amplitude evaluation criterion and hierarchical mean evaluation criterion includes: based on the layered deformation characteristics of track beam and stator core system, establish multi-class irregularity superposition model;Based on the superposition model, give the random amplitude of multi-class track geometric deviation in line with the set statistical distribution, and generate theoretical track random irregularity sample;Calculate the power spectral density of theoretical track random irregularity sample, and determine the multilevel confidence interval of power spectral density and corresponding track spectral area value;Based on the multilevel confidence interval of power spectral density, define track smoothness hierarchical amplitude evaluation criterion;Based on track spectral area value, define hierarchical mean evaluation criterion.The evaluation result of the application is more scientific, objective and comprehensive.

Systems and methods include an asset tracking platform for tracking, analyzing, and managing facility assets at an operational facility site. The asset tracking platform includes a site operator device in local communication with a field device. The facility assets are assigned to physical asset tags using an assignment procedure that includes scanning the physical asset tag with the field device (e.g., using a radio-frequency identifier (RFID) sensor) to associate an identifier of the physical asset tag with asset data (e.g., location data, an asset identifier, an asset type, etc.). The field device downloads an asset task list using the local communication to represent a list of daily tasks to be completed with the field device. Task execution procedures are performed for the list of daily tasks by scanning the physical asset tag with the field device which starts a timer. Task completion data is synced with the site operator device.

The present application discloses a work clothes detection method, system, equipment and medium combining yolov8 and target tracking, which relates to the field of safety detection technology. The method comprises acquiring surveillance video data of a target monitoring area and performing data preprocessing; performing work clothes target detection on the preprocessed surveillance video data in a preset work clothes detection model to obtain a work clothes detection result; performing target tracking analysis based on the work clothes detection result to obtain a work clothes tracking result of a continuous video stream; constructing a preset work clothes detection model in combination with the yolov8 algorithm and the target tracking algorithm; generating a defect detection mark based on the work clothes detection result and triggering an alarm instruction; acquiring geographic location information based on the work clothes tracking result, performing spatial matching analysis based on the geographic location information and the work clothes tracking result to obtain a safety target detection result of the corresponding target monitoring area; the present application provides a deep analysis method combining computer vision and deep learning technology to improve the efficiency of safety monitoring.