49 results about "Boundary crossing" patented technology

The invention relates to the technical field of chip packaging, in particular to a PCIe (peripheral component interface express) equipment function number border crossing prevention method supporting ARI (automatic repeat interface). The hardware layer protection process is as follows: S1-1, hot plugging; s1-2, if it is detected that the equipment declarates to support ARI, the PCIe controller hardware circuit immediately sets an equipment number register to zero; s1-3, only allowing the PCIe controller of the system to read the configuration space unidirectionally during the low level period of the reset signal PERST #; the interception process of the driving layer is as follows: S2-1, in an equipment enumeration stage, an operating system drives to dynamically read an ARI capability register; s2-2, if the equipment declarates to support ARI, forcing an equipment number register to be equal to 0, and releasing an original equipment number bit space for function number expansion; s2-3, border crossing access is blocked; and S2-4, synchronizing the state register. Compared with the prior art, correctness of device function numbers is guaranteed through two dimensions of hardware register state solidification and driver layer access truncation, and'perception-free 'fault tolerance of the PCIe device under ARI / traditional mode switching is achieved for the first time.

The invention provides a metamaterial robust optimization method based on a genetic algorithm, and relates to the technical field of metamaterial optimizing.The method comprises the steps that adjustable parameters of an electromagnetic metamaterial are determined, the adjustable parameters form individual vectors, and an initial population is generated; simulating equivalent magnetic conductivity corresponding to each individual vector through a performance simulation model; determining an equivalent magnetic conductivity distribution condition in a neighborhood space of each individual vector, and constructing a structure family thickness index; selecting individual vectors based on the equivalent magnetic conductivity and the structure family thickness index to carry out random crossing or boundary crossing, executing a random variation strategy and a pullback variation strategy on the crossed individuals, and carrying out iteration until a preset number of iterations is reached; and exporting the adjustable parameter of the individual vector with the maximum equivalent magnetic conductivity in the last generation as the optimized adjustable parameter.

The invention belongs to the technical field of building progress management, and particularly discloses a multi-modal data-based identification and acquisition method and system, and the method comprises the steps: detecting the process progress deviation of a current working face in a building construction process in real time, and recognizing a cross risk through combining with the subsequent process plan time when the progress deviation exists, therefore, a multi-dimensional rule-driven necessity evaluation mechanism is constructed, quantitative, objective and traceable evaluation on the feasibility of the cross operation is realized, and a feasible time window and a space authorization boundary of the cross operation are defined on the current working face while the cross operation is judged to be necessary. And then, monitoring a motion track of a subsequent process in an authorized area, synchronously predicting completion feasibility of a previous process in a feasible window period, and carrying out space-time two-dimensional compliance judgment, so that real-time perception and risk identification of cross operation are realized by depending on multi-modal data fusion, and anomalies such as border crossing operation or progress lag can be actively discovered. And regulation intervention is triggered in time.

The invention discloses a foundation pit monitoring self-adaptive regulation and control method based on reinforcement learning, and aims to solve the problems that safety constraints are difficult to meet continuously under complex working conditions and disturbances, and'controllability and no boundary crossing 'are difficult to guarantee. The foundation pit monitoring self-adaptive regulation and control method comprises the following steps: constructing a monitoring-prediction-check-projection-execution-learning closed loop taking safety embedding as a unified drive; according to the method, multi-source data and partition state estimation are fused, and distributed risk actor-commentator, physical constraint nerve prediction and adaptive confidence calibration, model prediction safety filtering and micro quadratic programming projection are adopted; the technical effects of real-time self-adaptive regulation and control within the standard limit value and the dynamic safety margin, reduction of the border crossing risk and improvement of the stability and the reliability are achieved.

A system, method, and computer-readable medium enforce compliance, policy governance, and jurisdiction-aware control of digital artifacts generated by AI models. A deterministic digest of each artifact is stored in a registry, and a compliance token is generated encoding policy predicates, consent scope, jurisdictional lineage, monetization parameters, and a zero-knowledge proof component including a proof statement ID, cryptographic commitment, and verification key. A gateway verifier receives the artifact and token, verifies the signature, performs zero-knowledge proof validation without revealing regulated data, evaluates encoded policies, and renders a permit-or-deny decision. Verification results and compliance decisions are recorded via a dual-ledger atomic commit engine that updates both compliance and financial ledgers. A settlement engine calculates royalty values based on validated performance metrics and appends the results to the commit entry. Upon jurisdictional transitions, a transition engine detects boundary crossings and issues a forked compliance token with updated predicates. The system ensures verifiable, policy-aware, and jurisdictionally adaptive governance.

The invention provides a method, a device, equipment and a medium for identifying abnormal behaviors of people crossing boundaries, and the method comprises the steps: collecting T frames of video images based on a target monitoring video, dividing each frame of video image into a plurality of independent image blocks, and converting the independent image blocks into image block feature vectors; generating a boundary distance map based on each frame of video image and the spatial boundary model; encoding the generated boundary distance map into a boundary sensing position embedding vector consistent with the feature vector dimension of the image block; adding the boundary sensing position embedding vector and the original space-time position code corresponding to each frame of image block to obtain an enhanced position code fused with boundary information; fusing the feature vectors of the image blocks of the enhanced position coding, and obtaining a feature map of enhanced spatial correlation information by using a spatial attention branch in a Division Attention structure; the feature map of the enhanced space correlation information utilizes a time attention branch in a Division Attention structure to obtain features fusing time sequence and motion information; and the space-time fusion feature identifies a border-crossing abnormal behavior based on an abnormal behavior identification model.

The invention discloses a production area border crossing monitoring method and device based on contour comparison, and the method comprises the steps: obtaining video data of a production area, and judging whether a human body outer contour frame is overlapped with a first area or a second area frame by frame; if the human body outer contour frame is overlapped with the first area and the second area, selecting a target track point in the second area according to the human body outer contour frame, and judging whether the to-be-detected object is in the production area according to the human body outer contour and the target track point; if it is judged that the to-be-detected object is located in the production area or the human body outer contour frame is completely overlapped with the first area, an identification model is called to conduct part identification on the human body outer contour frame, and warning information is output according to an identification result. Therefore, the position of the to-be-detected object is judged in a manner of overlapping and comparing the human body outer contour frame and the region, and meanwhile, part dimension recognition is performed in combination with the recognition model, so that the risk behavior prediction accuracy is effectively improved while the false alarm probability is reduced.

The invention discloses a millimeter wave radar on-bed / off-bed detection method and system based on micro-motion point cloud track maintenance and border crossing trend identification, and belongs to the technical field of millimeter wave radar perception and human body behavior identification. The method comprises the following steps: installing and calibrating a millimeter wave radar; synchronously calculating a dynamic point cloud for capturing significant motion and a micro-motion point cloud for capturing micro-displacement; during target track tracking, when the number of dynamic point clouds is insufficient, switching to micro point clouds to perform data association and track updating so as to maintain track continuity in a resting or deep sleep state; based on the continuous track, extracting distance sequence gradient features representing that the target approaches or is far away from the center of the bed body, and spatial position deviation features representing that the target is located inside or outside the bed; and finally fusing the two features to carry out consistency discrimination, thereby realizing robust identification of the on-bed or off-bed state of the human body. According to the method, the problems of track interruption and false alarm caused by sparse point clouds are effectively solved, and the accuracy, continuity and anti-interference capability of detection are remarkably improved.

The embodiment of the application discloses a kind of staff visit training voice interaction method and system, method includes: the voice data and voiceprint feature information of staff are collected by voice interaction terminal;The real-time position of staff is obtained by positioning base station;According to voice data, voiceprint feature information and real-time position, staff identity association information is generated, and is given training interactive equipment simultaneously, staff identity association information includes the terminal number, staff state information, staff position information and staff question information associated with staff identity mark one-to-one, staff identity mark is associated with voiceprint feature information;According to staff identity association information, execute position boundary-crossing warning operation and multi-concurrent audio processing operation.The application can solve the problem that new staff position boundary-crossing risk is difficult to early warning and multi-person concurrent question when guiding response is chaotic in the scene of staff pre-service visit training, improve the intelligence and safety of staff pre-service visit training.

The invention discloses an employee subject authorization verification and authorization change verifiable control method and system. The method comprises the following steps: segmenting a natural language authorization task and establishing session isolation for each subtask; calling a neural network model in the first session to output an employee subject name set, an action enumeration and a task phrase set in a fixed format; generating a candidate task subset based on the cross index, calling a neural network model in a second session to carry out constrained semantic alignment, and allowing to output candidate inner task codes or newly creating task names; the execution layer rejects border-crossing output through a hard check gate; a compensation transaction log is generated before writing, a pre-change snapshot and a fingerprint value of an influence domain file are stored, the fingerprint value is recorded after writing, a submission mark is written, and rollback recovery can be performed according to a transaction state during abnormal restart; the authorized cache is maintained or reconstructed based on the increments and the differential set, providing a withdrawal or compensation transaction based on fingerprint value collision detection.

The invention relates to the technical field of AI-based central air-conditioning system energy-saving optimization, and discloses an AI-based central air-conditioning system energy-saving optimization simulation platform. Setting a unified sampling period, reading rated temperature difference, temperature and heat exchange amount sensor resolution and maximum heat exchange amount from a specification, initializing thermal resistance and thermal capacity double-exponential smoothing parameters, and setting non-zero protection; the condenser temperature, the evaporator temperature and the heat exchange amount are collected in real time, and abnormity is eliminated through border crossing and lower limit verification; calculating a temperature difference and a first-order change rate thereof, and estimating instantaneous thermal resistance and thermal capacity according to the temperature difference and the first-order change rate; a smoothing coefficient is adaptively generated based on the continuous period parameter difference value, and double-exponential smoothing iteration updating is adopted; performing physical correction on the smoothed parameters and storing history in a rolling manner to ensure that the result is reliable; and finally, recommended heat exchange amount is calculated and subjected to range correction on the basis of the newest smooth thermal resistance and serves as closed-loop control reference of refrigerant flow or compressor load in the next period.

The invention discloses an intelligent identification risk early warning method for power distribution network uninterrupted operation, and particularly relates to the technical field of power distribution network live-line operation and maintenance, and the method comprises the following steps: arranging a high-definition camera, a positioning sensor and a global satellite navigation system for time service in an operation area, and collecting image and position information; a fusion sequence is generated through unified time reference alignment; inputting a deep learning model to identify personnel protection, insulation shielding and boundary crossing, and calculating a dynamic risk index in combination with relative potential of personnel, tools and live wires; continuously monitoring the time service integrity and self-correcting redundancy; if necessary, sound-light early warning is given out, and safety interlocking is switched in; after the alarm is released, storing and adaptively updating by incremental learning; according to the invention, based on unified time reference fusion acquisition, deep learning analysis protection, insulation shielding and border crossing, dynamic risk indexes are generated, early warning interlocking is carried out, and electric shock high falling misoperation is reduced; monitoring time service integrity self-correction redundancy verification is carried out, so that false alarm and missing alarm are reduced; and through incremental learning updating, the recognition precision and the power supply continuity are improved.

The invention provides a bastion host operation instruction border crossing behavior identification and control method, which comprises the following steps of: extracting an input value of a current operation instruction and a security boundary value preset by a system through a bastion host operation log, and synchronously acquiring a CPU occupancy rate, a memory usage amount and a running task number of a target host; analyzing the incidence relation among the CPU occupancy rate, the memory usage amount and the number of running tasks according to the host resource state data and the task load data, and identifying whether the host is in a light-load running state or a heavy-load running state currently to obtain a host load level; and according to the operation instruction input value, the security boundary value and the host load level, calculating the deviation degree of the operation instruction input value and the security boundary by adopting a dynamic risk quantification algorithm, and adjusting a risk weight coefficient in combination with the host load level to obtain a boundary crossing risk level.

This invention discloses a warning system and method for maintenance areas of hydropower stations based on laser sensing and light and shadow, comprising: a back-end management terminal, a local control unit, an area boundary crossing warning device, a projection matrix device, and a panel door opening warning device. The area boundary crossing warning device is used to identify personnel intrusion and determine its direction; the projection matrix device is used to create a virtual visual isolation zone on the ground; the panel door opening warning device is used to provide a live signal after the equipment panel door is opened; and the local control unit is used to send data from the back-end management terminal to the area boundary crossing warning device, the projection matrix device, and the panel door opening warning device, or to send data received from the area boundary crossing warning device, the projection matrix device, and the panel door opening warning device to the back-end management terminal.

A speed control system is configured to control a propulsion system of a mobile agricultural machine to ramp from a first ground speed to a second ground speed based upon a location of a boundary crossing point. The speed control system generates an effective boundary crossing point location based on the location of the boundary crossing point and a system delay between when a section control signal is generated and when a corresponding section control function is performed. A section control system generates section control signals based upon the effective boundary crossing point location.

The application discloses a kind of agent sensitive operation tracking methods and systems for permission boundary, specifically related to computer security and access control audit field, including obtaining multiple operation events generated by agent in a task execution process as input, and extracting event identification, subject identity information, accessed object information, operation type information and occurrence timing information for each operation event, output event sequence, wherein the acquisition and extraction are completed by edge computing node nearby, by boundary context labeling to operation event in task and generating identity delegation chain digest based on token backtracking, the traceable association of cross-system cross-identity is realized;Further construct real boundary crossing set and solve the minimum boundary crossing set and carry out differential evidence, output cause-effect slice evidence package.

The application belongs to the field of industrial safety monitoring, and discloses a foreign matter border crossing detection method, computer equipment and a storage medium. The method comprises the following steps: acquiring a current video frame; inputting the current video frame into a warning line segmentation model and a foreign matter detection model respectively to determine a binary mask of the current video frame and a dynamic detection frame for foreign matter; determining at least one warning boundary point set of the current video frame based on the binary mask; performing linear regression fitting on each warning boundary point set to obtain a corresponding candidate fitting warning line, and generating a corresponding candidate warning area based on each candidate fitting warning line; selecting a target warning area with the highest overlap degree with a current calibration warning area from each candidate warning area; if the overlap degree is less than a preset threshold, regarding the target warning area as a new calibration warning area; and determining whether there is foreign matter border crossing in the current video frame based on the dynamic detection frame and the new calibration warning area. The application can improve the accuracy of foreign matter border crossing detection.

This invention relates to a vision-based method and system for identifying and classifying the boundary crossing behavior of intelligent agents in oilfield electronic fences. Belonging to the fields of oilfield perimeter security and computer vision technology, it addresses the problems of existing electronic fences, such as inability to identify target types, inability to quantify boundary crossing risks, single warning levels, and low system integration. The method involves target detection and continuous tracking of oilfield perimeter image data to obtain target type and movement trajectory. Boundary crossing verification is performed on the target's center position based on a polygonal electronic fence, and the boundary crossing distance from the target to the fence boundary is calculated. This distance is then fused with target type risk weights to obtain an adjusted boundary crossing distance. Classified warnings are implemented based on preset thresholds. Simultaneously, the warning information is visualized and logged, achieving accurate identification and differentiated warnings for boundary crossing behavior. This invention is applicable to perimeter security monitoring and intelligent risk warning in complex environments such as oilfield mining areas, well sites, and oil and gas storage areas.

The invention belongs to the technical field of natural resource inspection, and particularly relates to a natural resource multi-sensing integrated monitoring method and platform. According to the method, through fusion correction of time synchronization, coordinate unification and historical tracks, the influence of time and coordinate abnormity and accidental positioning deviation on border crossing judgment can be effectively eliminated, so that misinformation and misscheduling are reduced, the self-learning responsibility range is automatically expanded and shrunk according to the space density and the inspection frequency, inspection resources can be concentrated in a high-demand or high-risk area, and the inspection efficiency is improved. Unnecessary repeated inspection on a low-risk area is avoided, manpower and operation cost is reduced, time-space matching and continuity inspection of events and tracks can restore inspection behavior paths and trends before and after the events occur, managers are assisted in judging event properties and responsibility ownership, the treatment pertinence and efficiency are improved, and the management cost is reduced. Through fusion evaluation of three types of risk indexes of space, border crossing and events, grading early warning of the natural resource state is realized.

The invention provides a minimally invasive surgery robot master-slave hand posture boundary impedance control method and device, a medium and a program product. The method comprises the steps that posture information of a master hand is obtained, and a target posture of a slave hand is generated; determining a feasible attitude boundary based on the structure constraint of the slave hand, and performing boundary detection on the target attitude to judge whether the master hand operation may cause attitude boundary crossing or autorotation boundary; when it is detected that the posture of the slave hand crosses the boundary, boundary limitation processing is conducted on the target posture, the slave hand is controlled to execute actions under the limited posture, meanwhile, first impedance feedback information is generated based on the posture deviation between the posture of the master hand and the actual posture of the slave hand, and the first impedance feedback information acts on the master hand to conduct boundary-crossing rollback; when it is detected that the rotation movement of the tail end of the slave hand reaches the rotation boundary, second impedance feedback information is generated based on the posture deviation in the rotation direction, acts on the master hand and retreats from the rotation limit position. According to the invention, real-time limitation and feedback control on the attitude boundary and autorotation boundary state of the slave hand are realized.

The invention particularly relates to a frontier defense perception alarm decision-making system based on a multi-mode visual model, which relates to the technical field of frontier defense security, and comprises the following steps: receiving video slices and disposal principle txt data sent by a background processing module, identifying and analyzing target behavior actions in the video slices, judging the alarm type and level in combination with the disposal principle, and sending the alarm type and level to the background processing module; and processing suggestions are generated and fed back to the background processing module. According to the method, the cross-modal understanding capability of the multi-modal visual large model is combined with frontier defense specific scene optimization and thinking chain reasoning mechanisms, so that the accuracy of frontier defense target behavior recognition and alarm judgment is greatly improved; the video is segmented through an adaptive slicing algorithm, the behavior action integrity is ensured, and model input interference is further reduced through format standardization processing; on the basis of fine adjustment of open source models such as qwenVL, a frontier defense scene exclusive data set is fused, and the recognition capability of professional behaviors such as illegal border crossing and suspicious article carrying is enhanced.

Embodiments of the present disclosure relate to dynamic throttling of workload-based video processing functions using machine learning. Systems and methods are disclosed for dynamically throttling video processing and / or streaming based on workload. Live video is captured from one or more sources (e.g., cameras) and stored. The video is then provided to a video processing engine and a video streaming engine. The video processing engine can perform one or more operations such as object detection, object tracking, and object classification to produce characterization data (e.g., bounding boxes, object tracks, alerts, object labels, object counts, boundary crossings, intersection highlights, etc.). System resource usage and performance of the video processing and streaming are monitored to produce workload data (e.g., metrics). Based on a policy and the workload data, the video streaming and / or processing is dynamically reconfigured by adjusting parameters provided to the video streaming and processing engines.

The invention relates to the technical field of process monitoring, in particular to a personnel intrusion detection method based on an artificial intelligence truck loading system, which comprises the following steps of: acquiring a continuous motion image sequence of a target, calculating a node included angle and direction deflection to generate positioning information, comparing an operation boundary to judge intrusion and generating an identification state. And determining an equipment response instruction in combination with the direction vector, loading a track in the three-dimensional scene, marking a risk area, and obtaining a personnel path intrusion calibration record. According to the invention, through continuous extraction of a human body structure bounding box in an image frame and dynamic displacement analysis of a track node, a target path with continuous gait features is identified and static and equipment interference areas are eliminated; whether the target crosses the boundary is judged by combining with the real-time loading operation boundary, and the minimum distance between the turning point and the boundary path is calculated to complete intrusion behavior recognition, so that the efficient linkage of real-time judgment of the behavior track and alarm direction control is realized, and the accurate recognition and real-time response capability of personnel intrusion events is enhanced.

The invention relates to the technical field of mobile communication, security and differential privacy, and particularly provides an electronic fence method for protecting user privacy, which comprises the following steps: acquiring original position data; and performing localized differential privacy perturbation processing on the original position data to generate perturbation position data satisfying a preset differential privacy budget, the perturbation processing including optimizing noise distribution to maintain subsequent electronic fence determination precision, dynamically binding the dynamic pseudo identity identifier with the current timestamp and the area identification information of at least one target electronic fence area; receiving a fence judgment request from the user terminal equipment; and based on the disturbance position data in the fence judgment request and the area identification information of the corresponding target electronic fence area, executing border crossing state judgment by using a pre-deployed lightweight geo-fence boundary matching algorithm. According to the invention, operation cost and resource consumption are reduced, and system performance and expandability are improved.

The embodiment of the application discloses a kind of staff visit training voice interaction method and system, method includes: the voice data and voiceprint feature information of staff are collected by voice interaction terminal;The real-time position of staff is obtained by positioning base station;According to voice data, voiceprint feature information and real-time position, staff identity association information is generated, and is given training interactive equipment simultaneously, staff identity association information includes the terminal number, staff state information, staff position information and staff question information associated with staff identity mark one-to-one, staff identity mark is associated with voiceprint feature information;According to staff identity association information, execute position boundary-crossing warning operation and multi-concurrent audio processing operation.The application can solve the problem that new staff position boundary-crossing risk is difficult to early warning and multi-person concurrent question when guiding response is chaotic in the scene of staff pre-service visit training, improve the intelligence and safety of staff pre-service visit training.

The invention provides an ephemeris pointing interpolation method and device, and the method comprises the steps: converting the ephemeris pointing at a first moment into a first unit vector, and converting the ephemeris pointing at a second moment into a second unit vector; determining a rotation axis angle between the first unit vector and the second unit vector; determining a rotation matrix of a to-be-interpolated moment according to the rotation axis angle, wherein the to-be-interpolated moment is located between the first moment and the second moment; and determining an interpolation vector of the to-be-interpolated moment according to the rotation matrix, and converting the interpolation vector into ephemeris orientation of the to-be-interpolated moment. According to the method, ephemeris directional interpolation is carried out at any angle, and the problems of singularity and border crossing are also avoided.

The invention provides a multi-dimensional operation rule detection method and system, electronic equipment and a medium. The method comprises the steps that metadata configuration information and historical rule combinations of all rule dimensions are acquired; wherein the metadata configuration information comprises a value set of a non-interval dimension, an upper limit boundary, a lower limit boundary and a boundary specification of at least one interval dimension; generating a theoretical combination rule complete set based on the metadata configuration information; performing conflict and deletion detection on the theoretical combination rules in the theoretical combination rule set based on the historical rule combination to obtain a detection result; wherein the conflict and deletion detection at least comprises starting point coverage detection, adjacent connection fault detection, overlapping conflict detection, end point coverage detection and border crossing and zero length interval detection. According to the invention, the coverage degree of multi-dimensional operation rule detection can be improved.

The invention discloses a role permission-based chemical reagent operation management method, which comprises the following steps of: acquiring biological characteristic data of an operator and reagent identification information, and matching to obtain a role permission configuration file; the operation type of the extraction reagent is limited, and an allowable operation range is determined; generating a temporary operation token and associating the temporary operation token with the timestamp to register as an authorized operation record; extracting a key event node to identify a permission border-crossing deviation; if the border crossing deviation exists, the key event nodes and the associated data are stored in an independent storage partition in an isolated mode and marked as abnormal log entries, and a behavior record set of hierarchical storage is formed; log aggregation processing is carried out on the behavior record set stored in a layered mode, a responsibility tracing path is determined according to the time sequence relation and role permission information, and audit report data is generated. Accurate authority control can be achieved through dynamic authorization and abnormal behavior real-time isolation, the responsibility tracing path is automatically constructed based on the abnormal logs stored in a layered mode, and operation safety and auditing efficiency are improved.

This invention relates to the field of risk perception and identification technology, specifically a risk perception and identification system for key indicators of cold chain systems. It includes collecting multi-point environmental data inside refrigerated trucks, extracting disturbance features and path trends, simultaneously analyzing the relationship between temperature, humidity, and wind changes, identifying boundary drift and trajectory overlap, labeling risk path levels, and outputting cold chain transportation risk identification results. This invention improves disturbance identification accuracy and structural analysis capabilities by introducing maximum difference node location, wind speed jump extraction, and humidity gradient segment analysis. It achieves collaborative identification by combining path unidirectionality and synchronous change relationships, constructs a drift identification mechanism through fluctuation ratio and boundary crossing judgment, establishes a concern level system based on trajectory overlap and directional consistency, and forms a classification structure based on continuity and coverage breadth, thereby enhancing the accuracy of path risk identification and trend judgment capabilities.

This invention provides a 3D Gaussian point cloud electronic fence alarm system for substations, comprising: an intelligent calibration module that calibrates the electronic fence in three-dimensional space based on a 3DGS model of the substation, determining the 3D Gaussian point cloud electronic fence of the substation; a spatial positioning module that combines multiple positioning methods to analyze the position of a work object, determining the position information of the work object; a boundary crossing analysis module that combines the position information of the work object with the 3D Gaussian point cloud electronic fence of the substation to perform spatial boundary crossing analysis and obtain the boundary crossing analysis result; and an alarm push module that performs multi-level linkage alarms and pushes based on the boundary crossing analysis result. This invention effectively improves anti-interference capability, enables accurate acquisition of the position information of the work object, thereby allowing for more accurate spatial boundary crossing analysis, avoiding false alarms or missed alarms, improving the accuracy of the 3D Gaussian point cloud electronic fence alarm system for substations, and reducing the false alarm rate.