Security patrol method, patrol device, and patrol apparatus

Abstract

The application provides a safety inspection method, an inspection device and an inspection equipment, and relates to the technical field of building safety. The safety inspection method comprises the following steps: in response to an operation instruction of an editable interface, a multistage process template of building safety inspection is configured, and a corresponding safety inspection task is generated; a shooting device collects real-time inspection data of a construction site; illegal operations, equipment abnormalities and environmental risks in the real-time inspection data are identified, and corresponding risk level results are generated; the risk level results are converted into visual alarm signals, the visual alarm signals are sent to a mobile terminal, and when the visual alarm signals are emergency signals, the visual alarm signals are sent to a monitoring center; response information of the mobile terminal is monitored, the state of the safety inspection task is updated, and operation logs of an inspector handling the risk level results are recorded; if the mobile terminal does not respond within a preset time, alarm information is sent to the mobile terminal again, and alarm information is sent to the monitoring center, so that the inspection efficiency and risk disposal capacity can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building safety, in particular to a safety inspection method, an inspection device and an inspection equipment. BACKGROUND

[0002] In the field of safety supervision in the construction industry, the traditional safety inspection method mainly relies on manual on-site inspection, paper records and scattered electronic management systems. For example, the inspection personnel need to manually fill in the inspection form, record the hidden dangers by taking pictures or videos, and then report the data to the management department level by level. In terms of risk prediction and real-time response, it still relies on manual analysis, which leads to limited efficiency improvement. SUMMARY

[0003] The main purpose of the present application is to provide a safety inspection method, which aims to improve the real-time performance and accuracy of construction site safety supervision.

[0004] To achieve the above purpose, the present application provides a safety inspection method, which comprises:

[0005] In response to the operation instruction of the editable interface, a multi-level process template of building safety inspection is configured, and the template parameters of the multi-level process template are adjusted according to the operation instruction to generate a corresponding safety inspection task. The multi-level process template includes customer management, risk assessment, approval input and supply chain management modules;

[0006] According to the safety inspection task, real-time inspection data of the construction site is collected through the shooting device of the mobile terminal of the inspection personnel, and the real-time inspection data includes at least one of images, videos or sensor data;

[0007] Identify the illegal operation, equipment anomaly and environmental risk in the real-time inspection data, and generate a corresponding risk level result;

[0008] The risk level result generates a corresponding visual alarm signal, and the visual alarm signal is sent to the monitoring center through the 5G / 4G network when the visual alarm signal is an emergency signal;

[0009] Monitor the response information of the corresponding mobile terminal, update the state of the safety inspection task and record the operation log of the inspection personnel processing the risk level result;

[0010] If the mobile terminal does not respond to the visual alarm signal within a preset time, the alarm information is sent to the mobile terminal again through a backup communication mode, and the alarm information is sent to the monitoring center.

[0011] Optionally, the multi-level process template of the building safety inspection is configured in response to the operation instruction of the editable interface, and specifically:

[0012] A standardized process template library corresponding to a building type or a project stage is acquired, and the standardized process template library includes risk management processes of a residence, a commercial building, an industrial plant, and different construction stages;

[0013] In response to the operation instruction of the editable interface, a process node is added or deleted in the standardized process template library, and a corresponding trigger condition is configured for each modified process node, and the trigger condition includes an environmental threshold, a device state threshold, and a manual judgment rule;

[0014] An execution personnel role is bound to each process node, and operation permissions, task execution deadlines, and timeout alarm rules corresponding to the execution personnel role are determined;

[0015] The trigger condition, the operation permission, the task execution deadline, and the timeout alarm rule are imported into the standardized process template library to generate the multi-level process template adapted to the current project.

[0016] Optionally, the corresponding trigger condition is configured for each modified process node, and specifically:

[0017] The historical inspection data of the current project and a similar project risk case library are acquired;

[0018] The historical inspection data and the similar project risk case library are measured for risk types and trigger probabilities by a time series prediction model to generate a corresponding prediction result;

[0019] The prediction result is associated with the trigger condition;

[0020] The trigger condition, the operation permission, the task execution deadline, and the timeout alarm rule are imported into the standardized process template library to generate the multi-level process template adapted to the current project, and specifically:

[0021] The trigger condition is mapped to a corresponding process node to automatically trigger the process node;

[0022] In the case of mapping the trigger condition to the corresponding process node, task execution permissions are assigned to different roles according to the operation permissions to generate and determine task execution deadlines and timeout alarm rules corresponding to the task execution permissions to generate the multi-level process template.

[0023] Optionally, the template parameters of the multi-level process template are adjusted according to the operation instruction to generate a corresponding safety inspection task, and specifically:

[0024] Obtain the project progress data, resource allocation table, historical risk distribution map and material scheduling plan of the current project, and extract the first parameter in the multi-level process template, the first parameter including contract performance period, hidden danger weight coefficient and budget allocation ratio;

[0025] Determine the deviation value of the project progress data of the current project and the contract performance period, the matching degree of the resource allocation table and the budget allocation ratio, and the relevance of the historical risk distribution map and the hidden danger weight coefficient;

[0026] According to the deviation value, the matching degree and the relevance, determine the deviation threshold range of the first parameter;

[0027] Push the deviation threshold range of the first parameter to the editable interface, monitor the parameter adjustment operation of the editable interface, update the template parameter of the multi-level process template, and associate and bind the updated template parameter with the material scheduling plan;

[0028] And the template parameter logic verification of the binding material scheduling plan, and after verification, generate a safety inspection task containing task priority, inspection route planning and resource allocation instruction.

[0029] Optionally, the deviation threshold range of the first parameter is determined according to the deviation value, the matching degree and the relevance, comprising:

[0030] Generate an initial deviation threshold corresponding to the deviation value;

[0031] Calculate the priority weight of resource allocation corresponding to the matching degree by weighting calculation method, and map to the initial deviation threshold to generate a second deviation threshold;

[0032] Predict the hidden danger probability distribution corresponding to the relevance by Bayesian probability model, and fuse into the second deviation threshold to determine the deviation threshold range of the first parameter.

[0033] Optionally, the updated template parameter is associated and bound with the material scheduling plan, specifically:

[0034] By graph modeling technology, the updated template parameter and the key nodes in the material scheduling plan establish a topological relationship;

[0035] Embed real-time monitoring trigger in the topological relationship, when the state of the key nodes in the material scheduling plan changes, automatically trigger dynamic verification of the template parameter;

[0036] And according to the result of dynamic verification, real-time adjust the template parameter.

[0037] Optionally, the step of identifying violations, equipment malfunctions, and environmental risks in the real-time inspection data and generating corresponding risk level results includes:

[0038] The pre-trained violation recognition model is invoked to perform frame-level analysis on the photos or videos in the real-time patrol data to generate corresponding analysis results.

[0039] When the analysis result indicates a violation, the facial features and violation timestamp of the violator are extracted to generate structured data containing the employee's employee number and violation type.

[0040] Obtain equipment operation data and environmental monitoring data at the construction site;

[0041] The structured data, equipment operation data, and environmental monitoring data are input into the risk level assessment model to generate the corresponding risk level results.

[0042] Optionally, generating a corresponding visual alarm signal from the risk level result specifically involves:

[0043] The alarm template library is matched with the risk level results. The alarm template library includes equipment failure alarm templates, environmental exceedance alarm templates, and personnel violation alarm templates.

[0044] Spatial enhancement processing is performed on alarm signals, overlaying the coordinates of the hidden danger location onto the BIM model of the construction site, and marking the association with the affected equipment in the editable interface;

[0045] In the event of an environmental risk alarm, links to historical similar accident handling cases and the status of emergency supplies inventory will be displayed at the corresponding display position on the editable interface.

[0046] Based on the risk level results, alarm icons with different colors and flashing frequencies are generated on the editable interface.

[0047] In addition, to achieve the above objectives, the present invention also provides an inspection device, the inspection device comprising: a memory, a processor, and a security inspection program stored in the memory and executable on the processor, the security inspection program being configured to implement the security inspection method as described above.

[0048] In addition, to achieve the above objectives, the present invention also provides an inspection device, including the inspection apparatus described above.

[0049] This invention, in response to operation commands on an editable interface, configures a multi-level process template for building safety inspections and adjusts the template parameters according to the commands to generate corresponding safety inspection tasks. The multi-level process template includes modules for customer management, risk assessment, approval input, and supply chain management. Based on the safety inspection tasks, real-time inspection data from the construction site is collected using the mobile terminal's camera device. This real-time inspection data includes at least one of images, videos, or sensor data. Violations, equipment malfunctions, and environmental risks in the real-time inspection data are then identified, generating corresponding risk level results. These risk level results are then used to generate corresponding visual alarm signals, which are then distributed to [the relevant network] based on LoRa spread spectrum technology. The system collects real-time patrol data from mobile terminals. When a visual alarm signal is an emergency signal, it sends the signal to the monitoring center via 5G / 4G network, monitors the response of the corresponding mobile terminals, updates the status of the safety patrol task, and records the operation logs of patrol personnel handling risk levels. Finally, if the mobile terminal does not respond to the visual alarm signal within a preset time, it resends the alarm information to the mobile terminal via a backup communication method and sends an alert to the monitoring center. By configuring multi-level process templates, real-time data collection, intelligent risk analysis, and closed-loop alarm management, the system achieves dynamic safety supervision throughout the building's entire lifecycle, significantly improving patrol efficiency and risk handling capabilities, ensuring the accuracy and timeliness of building safety management, and reducing the accident rate. Attached Figure Description

[0050] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of a security inspection method according to an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0055] Figure 4This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0056] Figure 5 This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0057] Figure 6 This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0058] Figure 7 This is a schematic diagram of a security inspection method according to another embodiment of the present invention;

[0059] Figure 8 This is a schematic diagram of a security inspection method according to another embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Well-known modules, units, and their connections, links, communications, or operations are not shown or described in detail. Furthermore, the described features, architectures, or functions can be combined in any way in one or more embodiments. Those skilled in the art should understand that the various embodiments described below are only for illustrative purposes and not for limiting the scope of protection of the present invention. It is also readily understood that the modules, units, or processing methods in the various embodiments described herein and shown in the accompanying drawings can be combined and designed in various different configurations. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] The definitions of various terms or methods used in the following embodiments are, except where logically impossible, generally defined as broad concepts that can be implemented under the premise of the content disclosed in the embodiments. Under this understanding, all specific subordinate limitations of the terms or methods should be considered as part of the invention and should not be narrowly interpreted or biased simply because the specification does not disclose such a specific limitation. Similarly, provided that it is logically feasible, the order of the steps in the method is flexible and varied, and all specific subordinate limitations in the broad concepts of various terms or methods fall within the scope of protection of this invention.

[0063] In the field of safety supervision in the construction industry, traditional safety inspection methods mainly rely on manual on-site inspections, paper records, and decentralized electronic management systems. For example, inspectors need to manually fill out inspection forms, record potential hazards by taking photos or videos, and then report the data to the management department level by level. This method suffers from cumbersome processes and slow response times, especially in risk prediction and real-time response, which still rely on manual analysis, resulting in low efficiency and the potential to miss key risk points. In addition, traditional methods lack a unified multi-level process management mechanism, making it impossible to dynamically adjust inspection strategies according to building type or project stage, and also making it difficult to achieve automated identification and graded alarms for risk data. In terms of communication, traditional systems typically use a single network to transmit alarm signals, and lack an effective backup communication mechanism when mobile terminals do not respond in time, which may delay the handling of emergencies.

[0064] The main solution of this application embodiment is as follows: A multi-level process template for building safety inspections is configured in response to operation instructions from an editable interface. The template parameters of the multi-level process template are adjusted according to the operation instructions to generate corresponding safety inspection tasks. The multi-level process template includes modules for customer management, risk assessment, approval input, and supply chain management. Based on the safety inspection tasks, real-time inspection data of the construction site is collected using the camera device on the mobile terminal of the inspection personnel. The real-time inspection data includes at least one of images, videos, or sensor data. Violations, equipment malfunctions, and environmental risks in the real-time inspection data are then identified, and corresponding risk level results are generated. The system generates corresponding visual alarm signals and distributes them to the mobile terminals that collect real-time patrol data based on LoRa spread spectrum technology. When the visual alarm signal is an emergency signal, it is sent to the monitoring center via 5G / 4G network. The system also monitors the response information of the corresponding mobile terminals, updates the status of the security patrol task, and records the operation logs of the patrol personnel in handling the risk level results. Finally, if the mobile terminal does not respond to the visual alarm signal within a preset time, the alarm information is sent to the mobile terminal again through the backup communication method, and an alarm message is sent to the monitoring center. This is achieved through the configuration of multi-level process templates, real-time data collection, intelligent risk analysis, and closed-loop alarm management.

[0065] In this embodiment, for ease of description, the following description uses the inspection device as the main execution subject.

[0066] This application provides a solution that generates inspection tasks by configuring multi-level process templates, collects and analyzes construction site data in real time, issues alarm signals based on multi-network communication mechanisms, and combines backup communication to ensure response reliability. This solves the problems of cumbersome processes, delayed responses, and limited communication in traditional methods, offering advantages such as improved inspection process efficiency, real-time risk identification and tiered alarms, and enhanced communication redundancy. It enables dynamic safety supervision throughout the entire building lifecycle, significantly improving inspection efficiency and risk handling capabilities, ensuring the accuracy and timeliness of building safety management, and reducing the accident rate.

[0067] Therefore, the present invention proposes a security inspection method; it is understood that the inspection equipment is equipped with an inspection device for storing and executing the following method. The inspection device can be implemented using a main controller, such as an MCU (Microcontroller Unit), DSP (Digital Signal Processor), FPGA (Field Programmable Gate Array), or SOC (System On Chip).

[0068] In existing technologies, safety supervision in the construction industry mainly relies on manual on-site inspections and paper records. Inspectors must manually fill out forms and record potential hazards by taking photos, which are then reported level by level to the management department. Traditional electronic management systems suffer from fragmented processes, and risk prediction and real-time response still depend on manual analysis, resulting in low efficiency. For example, in large construction sites, the complex working environment makes paper records easy to lose, and decentralized systems make it difficult to achieve multi-departmental collaboration, making it difficult to handle sudden risks in a timely manner.

[0069] To address the aforementioned issues, traditional methods suffer from three core shortcomings: a lack of unified standards for process management, insufficient efficiency in data collection and analysis, and unreliable alarm information transmission. To resolve these problems, this paper first constructs a configurable multi-level process template, integrating modules such as customer management and risk assessment to achieve dynamic adaptation of standard processes to project characteristics. Secondly, it utilizes mobile terminals to collect multimodal data in real time, combined with intelligent identification models to improve risk detection efficiency. Finally, it establishes a dual-channel communication mechanism, using LoRa to ensure basic alarm transmission, prioritizing 5G / 4G for emergency signals, and setting backup communication trigger conditions to ensure alarm reachability. This series of designs breaks through the traditional single-point solution, forming a complete closed-loop management system.

[0070] Based on the above, referring to Figure 1 In one embodiment of the present invention, the security patrol method includes steps S100-S600, wherein:

[0071] S100. Responding to the operation instructions of the editable interface, configure the multi-level process template for building safety inspection, and adjust the template parameters of the multi-level process template according to the operation instructions to generate the corresponding safety inspection task.

[0072] S200. According to the safety patrol task, real-time patrol data of the construction site is collected through the shooting device of the patrol personnel's mobile terminal.

[0073] S300: Identify illegal operations, equipment malfunctions, and environmental risks in the real-time inspection data, and generate corresponding risk level results;

[0074] S400. Generate a corresponding visual alarm signal from the risk level result, and send the visual alarm signal to the mobile terminal corresponding to the real-time patrol data based on LoRa spread spectrum technology. When the visual alarm signal is an emergency signal, send the visual alarm signal to the monitoring center through the 5G / 4G network.

[0075] S500: Monitor the response information of the corresponding mobile terminal, update the status of the security patrol task, and record the operation log of the patrol personnel in handling the risk level result;

[0076] S600. If the mobile terminal does not respond to the visual alarm signal within a preset time, the alarm information is sent to the mobile terminal again through the backup communication method, and an alarm message is sent to the monitoring center.

[0077] The multi-level process template includes modules for customer management, risk assessment, approval input, and supply chain management. This multi-level process template refers to a dynamically configurable building safety management framework. It allows for the addition, deletion, and modification of process nodes via a visual programming interface, and supports the integration of customer needs, risk assessment standards, and other elements into a unified management platform. Real-time inspection data includes at least one of image, video, or sensor data. Real-time inspection data collection is achieved through mobile terminals integrating cameras, environmental sensors, and other devices, such as using an industrial-grade tablet computer with waterproof and dustproof characteristics for on-site photography. Risk level identification employs a pre-trained image recognition model, which can use a violation detection algorithm based on YOLOv5 to automatically identify missing safety equipment and abnormal equipment states. Visual alarm signals can be generated through BIM model coordinate mapping, such as highlighting the location of potential hazards in a 3D building model. Backup communication methods can include satellite communication modules or Mesh network devices, automatically switching transmission paths when the primary communication channel fails.

[0078] When project managers configure process templates through an editable interface, the inspection device automatically links the contract terms from the customer management module with the inspection standards from the risk assessment module. After an inspection task is generated, the mobile terminal automatically pushes the inspection route based on GPS positioning, and the video data collected on-site is uploaded to the cloud after preliminary analysis by edge computing devices. When missing scaffolding fixing bolts are detected, the inspection device automatically marks the location of the hazard and assesses it as a level two risk. It then broadcasts an alarm to all terminals within a 200-meter range via a LoRa base station and simultaneously transmits the alarm information to the headquarters monitoring screen via the 5G network. Upon receiving an unconfirmed alarm, the monitoring center automatically activates a backup communication channel to send an SMS reminder to the project manager's mobile phone, ensuring a closed-loop risk management process.

[0079] Compared to existing technologies, traditional methods employ independent risk assessment systems and manual inspection processes, leading to data silos and response delays. This embodiment achieves cross-departmental business integration through multi-level process templates, eliminating the time lost in paper-based work order processing; it uses a dual-channel communication mechanism compared to a single wireless transmission method, improving alarm information delivery rates in complex building environments; an intelligent identification model replaces manual screening of image data, reducing hazard detection time from hours to minutes; and setting backup communication trigger conditions effectively solves the problem of alarm loss caused by signal obstruction in existing technologies.

[0080] This embodiment achieves standardized management of building safety inspection processes, improves cross-departmental collaboration efficiency, ensures the real-time and accurate identification of risks at construction sites, establishes a hierarchical alarm transmission mechanism to ensure the accessibility of key information in scenarios with limited communication conditions, and forms a complete risk handling and supervision chain, enabling accountability through operation logs.

[0081] This embodiment configures a multi-level process template for building safety inspections in response to operation commands on an editable interface. The template parameters are adjusted according to these commands to generate corresponding safety inspection tasks. The multi-level process template includes modules for customer management, risk assessment, approval input, and supply chain management. Based on the safety inspection tasks, real-time inspection data from the construction site is collected via the mobile terminal's camera. This real-time inspection data includes at least one of images, videos, or sensor data. Violations, equipment malfunctions, and environmental risks within the real-time inspection data are identified, generating corresponding risk level results. These risk level results are then used to generate corresponding visual alarm signals, which are distributed to the mobile terminal that collected the real-time inspection data using LoRa spread spectrum technology. When the visual alarm signal is an emergency signal, [further action is taken]. The 5G / 4G network sends visual alarm signals to the monitoring center and monitors the response information of the corresponding mobile terminals. It updates the status of safety patrol tasks and records the operation logs of patrol personnel handling risk levels. Finally, if the mobile terminal does not respond to the visual alarm signal within a preset time, the alarm information is resent to the mobile terminal via backup communication, and an alarm message is sent to the monitoring center. By configuring multi-level process templates, real-time data acquisition, intelligent risk analysis, and closed-loop alarm management, this system generates patrol tasks using multi-level process templates, collects and analyzes construction site data in real time, issues alarm signals based on multi-network communication mechanisms, and combines backup communication to ensure response reliability. This solves the problems of cumbersome processes, delayed responses, and single communication methods inherent in traditional methods. It offers advantages such as improved patrol process efficiency, real-time risk identification and tiered alarms, and enhanced communication redundancy. It enables dynamic safety supervision throughout the entire building lifecycle, significantly improving patrol efficiency and risk handling capabilities, ensuring the accuracy and timeliness of building safety management, and reducing the accident rate.

[0082] Optionally, refer to Figure 2 Another embodiment of the present invention provides a security patrol method, based on the above. Figure 1 The illustrated embodiment, in response to operation instructions from the editable interface, configures a multi-level workflow template for building safety inspections, specifically steps S110-S140, wherein:

[0083] S110. Obtain a standardized process template library corresponding to the building type or engineering stage;

[0084] S120. In response to the operation instructions of the editable interface, add or delete process nodes in the standardized process template library, and configure corresponding trigger conditions for each modified process node.

[0085] S130. Bind an executor role to each process node and determine the operation permissions, task execution period and timeout alarm rules corresponding to the executor role.

[0086] S140. Import the triggering conditions, operation permissions, task execution period and timeout alarm rules into the standardized process template library to generate the multi-level process template adapted to the current project.

[0087] The standardized process template library includes risk management processes for residential buildings, commercial buildings, industrial plants, and different construction stages. This library is a pre-established database containing risk management processes for various building types and construction stages. It can be implemented using a relational database combined with classification tags, providing a basic framework for different projects. The triggering conditions include environmental thresholds, equipment status thresholds, and manual judgment rules. Triggering conditions refer to the environmental, equipment, or manual judgment standards that must be met when a process node is started. These can be implemented by combining a threshold setting module with a rule engine, transforming human experience into automatically executable conditional judgments. The executor roles refer to the job responsibilities associated with process nodes, which can be implemented using a role-permission matrix and RBAC model, clarifying the operational scope and timeliness requirements of different roles.

[0088] The process begins by retrieving a basic template matching the current building type from a standardized process template library, such as a template for the construction phase of an industrial plant. Then, process nodes are adjusted through an editable interface; for example, a high-altitude operation inspection node can be added to the risk assessment module. Trigger conditions are configured for this node, such as automatically triggering the inspection task when wind speed exceeds a set threshold. The node is then linked to a safety officer role, and the task is set to be completed within 24 hours, triggering an alarm if the timeout is exceeded. Finally, the configured node parameters are imported into the template library to create a customized template containing multi-level approval processes.

[0089] This embodiment enables dynamic addition and deletion of process nodes through an editable interface. Combined with trigger condition configuration and role permission binding, it allows the same template library to quickly generate different processes suitable for the main construction stage of commercial buildings or the equipment installation stage of industrial plants, while maintaining the consistency of core risk management logic. This embodiment solves the problem that fixed process templates cannot adapt to diverse project needs, and realizes rapid customization configuration based on standardized templates. It allows process nodes to be dynamically adjusted according to specific project characteristics, and ensures the standardization and timeliness of task execution by binding role permissions and timeliness rules.

[0090] Optionally, refer to Figure 3 Another embodiment of the present invention provides a security patrol method, based on the above. Figure 2 The embodiment shown configures corresponding trigger conditions for each modified process node, specifically steps S121-S123, wherein:

[0091] S121. Obtain historical inspection data for the current project and a database of risk cases for similar projects;

[0092] S122. Calculate the risk type and trigger probability based on the historical inspection data and the risk case library of similar projects using a time-series prediction model, so as to generate corresponding prediction results.

[0093] S123. Associate the prediction result with the triggering condition;

[0094] The step of importing the triggering conditions, operation permissions, task execution deadlines, and timeout alarm rules into the standardized process template library to generate the multi-level process template adapted to the current project includes steps S141-S142, wherein:

[0095] S141. Map the triggering conditions to the corresponding process nodes so that the process nodes are triggered automatically;

[0096] S142. When the triggering conditions are mapped to the corresponding process nodes, task execution permissions are assigned to different roles according to the operation permissions, so as to generate and determine the task execution period and timeout alarm rules corresponding to the task execution permissions, so as to generate the multi-level process template.

[0097] Among them, the time-series prediction model refers to a prediction model built based on time series analysis methods, which can be implemented using ARIMA models or LSTM neural networks, to capture the temporal correlation and periodic patterns of risk events. Trigger condition mapping refers to establishing a relationship between risk prediction results and the execution logic of process nodes, which can be implemented using rule engines or event-driven architectures to achieve automatic linkage between risk events and process actions. The similar project risk case library refers to a structured database storing historical risk events from different engineering projects, which can be built using relational databases or knowledge graph technologies to provide cross-project risk pattern references.

[0098] This process involves extracting the time, type, and handling records of risks from historical inspection data, and combining this with the risk statistical characteristics of similar projects in a case study library. A time-series prediction model is then used to calculate the probability of risk occurrence and the potential impact range at different construction stages. The prediction results are converted into conditional judgment rules. For example, when the probability of violations related to high-altitude operations in a certain area exceeds a threshold, the inspection process node for that area is automatically triggered. The triggering conditions are embedded in the execution logic of the process nodes. When sensor monitoring data or manual inspection results meet preset conditions, the inspection device automatically activates subsequent approval or rectification tasks. Simultaneously, task handlers are dynamically assigned based on role permissions. For example, high-risk alarms are automatically dispatched to the safety supervisor's account, with a two-hour response deadline set. After the deadline, an escalation alarm is sent to the next higher-level manager.

[0099] This embodiment establishes a data-driven trigger condition generation mechanism by integrating historical data and an external case library, enabling process nodes to respond in real time to changes in risk probability and eliminating the lag caused by manual recording. This embodiment achieves dynamic matching of task assignment and personnel capabilities by combining permissions with dynamically generated task execution deadlines. This embodiment can dynamically optimize process trigger rules based on actual engineering data, reducing the error rate of process configuration caused by human experience bias. Simultaneously, the automated trigger mechanism shortens risk response time, preventing the expansion of security risks caused by delays in manual operation. Furthermore, the dynamic binding of permissions and task deadlines ensures that high-risk tasks are prioritized for allocation to personnel with processing authority, improving task execution efficiency and accountability.

[0100] Optionally, refer to Figure 4 Another embodiment of the present invention provides a security patrol method, based on the above. Figure 1 The embodiment shown adjusts the template parameters of the multi-level process template according to the operation instructions to generate the corresponding security inspection task, specifically steps S150-S190, wherein:

[0101] S150. Obtain the current project's progress data, resource allocation table, historical risk distribution map, and material scheduling plan, and extract the first parameter from the multi-level process template;

[0102] S160. Determine the deviation between the current project's progress data and the contract performance period, the matching degree between the resource allocation table and the budget allocation ratio, and the correlation between the historical risk distribution map and the hidden danger weight coefficient.

[0103] S170. Determine the deviation threshold range of the first parameter based on the deviation value, the matching degree, and the correlation.

[0104] S180. Push the deviation threshold range of the first parameter to the editable interface, monitor the parameter adjustment operation of the editable interface, update the template parameters of the multi-level process template, and associate and bind the updated template parameters with the material scheduling plan.

[0105] S190, and logically verify the template parameters bound to the material scheduling plan, and generate a safety inspection task containing task priority, inspection route planning and resource allocation instructions after successful verification.

[0106] The first parameter includes the contract performance period, the risk weight coefficient, and the budget allocation ratio. Project progress data refers to the time nodes and workload data of the actual project completion status, which can be collected using progress management software to detect deviations between actual progress and the contract performance period. The resource allocation table records the allocation plan for manpower, equipment, and materials, and can be exported from an ERP system to assess the matching degree between the budget allocation ratio and actual resource usage. The first parameter refers to the standardized parameters preset in the process template, obtained from the process template through a parameter extraction algorithm, serving as a benchmark reference value for dynamic adjustment. The deviation threshold range refers to the reasonable range within which parameters can be dynamically adjusted, dynamically generated through a weighted calculation method and a Bayesian probability model, used to define the allowable parameter fluctuation range. The material scheduling plan refers to the time nodes involving material transportation and equipment allocation, using data from logistics management inspection devices, and ensuring synchronization of resource supply and task requirements by associating it with template parameters. Logical verification refers to verifying the logical consistency between parameters and the material plan through a rule engine to prevent task conflicts or resource waste caused by parameter errors.

[0107] The system includes several key features. First, by collecting project progress data and comparing it with the contract performance cycle, it can automatically detect whether a project is behind schedule or ahead of schedule, generating deviation values ​​as a basis for adjusting the contract period. Second, the matching degree calculation between the resource allocation table and the budget allocation ratio can identify whether resource allocation deviates from the original plan, thereby adjusting budget parameters. Third, the correlation analysis between historical risk distribution maps and hazard weight coefficients can dynamically correct the hazard weights in different areas, making the template parameters more closely match the actual risk distribution. Fourth, after the deviation threshold range is pushed to the editable interface, operators can adjust parameters based on quantitative indicators. For example, when the project progress deviation exceeds the threshold, it triggers automatic correction of the contract performance cycle parameters. Fifth, the updated template parameters establish a topological relationship with key nodes of the material scheduling plan through graph modeling technology. When the material supply status changes, triggers automatically verify whether the template parameters are suitable; if the verification fails, the parameters are readjusted. Finally, the template parameters, after logical verification, generate tasks containing priority sorting, inspection path optimization, and resource instructions, ensuring that inspection tasks remain synchronized with dynamic changes in the project.

[0108] This embodiment achieves intelligent calibration of template parameters through automated data association and dynamic calculation of deviation thresholds. By binding it to the topology of the material scheduling plan, it forms a closed-loop control of parameter adjustment and resource supply, solving the problem of plan rigidity caused by fixed parameters in traditional methods. This embodiment realizes dynamic matching between template parameters and actual engineering conditions, ensuring that the priority of inspection tasks, route planning, and resource allocation instructions can be automatically adjusted according to project progress, resource changes, and risk distribution, effectively avoiding resource waste and task execution deviations caused by parameter lag or errors.

[0109] Optionally, refer toFigure 5 The present invention also provides a security patrol method in one embodiment, based on the above. Figure 4 The illustrated embodiment determines the deviation threshold range of the first parameter based on the deviation value, the matching degree, and the correlation, including steps S171-S173, wherein:

[0110] S171. Generate an initial deviation threshold corresponding to the deviation value;

[0111] S172. Calculate the priority weight of resource allocation corresponding to the matching degree using a weighted calculation method, and map it to the initial deviation threshold to adjust and generate a second deviation threshold;

[0112] S173. Predict the probability distribution of potential hazards corresponding to the correlation using a Bayesian probability model, and integrate it into a second deviation threshold to determine the deviation threshold range of the first parameter.

[0113] The initial deviation threshold refers to the range of benchmark parameters generated based on the deviation between project progress and contract performance period. This can be implemented using linear interpolation or sliding window statistical methods to quantify contract performance constraints. The priority weight of resource allocation refers to a dynamic coefficient calculated based on the matching degree between the resource allocation table and the budget allocation ratio. This can be implemented using the analytic hierarchy process (AHP) or entropy weight method to reflect the impact of resource scarcity on the threshold. The hazard probability distribution refers to a risk occurrence probability model calculated using historical risk data and similar cases. This can be implemented using Monte Carlo simulation or Markov chains to predict the statistical patterns of hazard occurrence.

[0114] When a deviation occurs between the contract performance period and the project schedule, the average progress deviation over the past four weeks is first calculated using a sliding window statistical method to generate an initial deviation threshold range. Then, based on the relationship between equipment scheduling efficiency and budget allocation ratio in the resource allocation table, the priority weight of each resource item is calculated using the entropy weight method. This weight is multiplied by the upper limit of the initial deviation threshold to generate a second deviation threshold. Finally, collapse accident probability data for similar projects is extracted from a historical database, and the posterior probability of the current hazard occurring is calculated using a Bayesian model. This probability value is then mapped to the second deviation threshold range for dynamic expansion, forming the final deviation threshold range. This process achieves intelligent adjustment of the parameter range through the fusion of rigid contract performance constraints, dynamic resource scheduling weights, and statistical patterns of hazard occurrence.

[0115] This embodiment integrates resource priority weights and hazard probability distributions, enabling the threshold range to automatically shrink based on resource scarcity and dynamically expand by incorporating hazard statistical patterns. This avoids misjudgments while improving risk detection sensitivity. This embodiment can generate a deviation threshold range that dynamically matches the actual project status, making the triggering conditions for contract performance monitoring, resource allocation adjustments, and hazard warnings in safety inspection tasks more aligned with the actual project situation. For example, during construction in the rainy season, this method automatically widens the environmental monitoring threshold range by integrating historical collapse probability data, avoiding frequent false alarms due to weather factors. Simultaneously, it dynamically adjusts the resource allocation deviation threshold based on the pumping equipment scheduling priority, ensuring timely monitoring of critical equipment.

[0116] Optionally, refer to Figure 6 Another embodiment of the present invention provides a security patrol method, based on the above. Figure 4 The embodiment shown associates and binds the updated template parameters with the material scheduling plan, including steps S181-S183, wherein:

[0117] S181. Using graph modeling technology, establish a topological relationship between the updated template parameters and key nodes in the material scheduling plan;

[0118] S182. Embed a real-time monitoring trigger in the topology relationship. When the status of a key node in the material scheduling plan changes, the dynamic verification of the template parameters is automatically triggered.

[0119] S183, and adjust the template parameters in real time based on the results of dynamic verification.

[0120] Among these, graph modeling technology refers to expressing the relationship between parameters and resource scheduling through a network structure composed of nodes and edges. This can be implemented using knowledge graphs or graph databases, such as using Neo4j to construct the attributes and association rules of parameters and resource nodes. Real-time monitoring triggers are state-aware modules embedded in the topology, which can be implemented using an event-driven architecture combined with a rule engine, such as configuring Apache Kafka to listen for state change events of resource nodes. Dynamic verification refers to logically verifying parameters based on changes in resource state. This can be implemented by combining verification algorithms with real-time data stream processing technology, such as using the Flink stream processing engine to execute parameter verification logic.

[0121] In the material scheduling plan of the engineering project, there is a correlation between the concrete supply node and the construction progress threshold in the formwork parameters. When a concrete transport vehicle arrives late due to a malfunction, a real-time monitoring trigger captures the status change event of that node and immediately triggers the parameter verification process. The inspection device calls a preset verification rule library, such as comparing the delay time with the construction progress threshold, and finds that the original parameter setting for the concrete pouring window period cannot meet the adjusted delivery time. A new progress threshold is calculated through a dynamic verification algorithm, the construction sequence time in the formwork parameters is automatically updated, and the adjusted parameters are synchronized to the associated inspection task generation module to replan the safety inspection route.

[0122] Compared to existing technologies, traditional parameter adjustments rely on manual identification of material changes and parameter relationships. For example, managers need to manually verify transportation delay records and construction schedules. Existing technologies use linear correlation methods, such as setting simple time correlation formulas in spreadsheets, which cannot handle complex relationships involving the cross-influence of multiple nodes. This embodiment establishes a multi-dimensional correlation network through graph modeling, simultaneously associating multiple nodes such as concrete supply, scaffolding erection progress, and worker shift schedules, enabling dynamic parameter adjustments across dimensions.

[0123] Through the aforementioned technical means, this embodiment solves the problem of adjustment lag caused by insufficient correlation between template parameters and material scheduling. For example, it automatically adjusts the frequency of safety patrols and key areas when material supply is abnormal. A real-time monitoring trigger mechanism avoids response delays from manual monitoring; for example, parameter verification is completed within 5 minutes of equipment failure. Dynamic parameter adjustment ensures real-time coordination between safety patrol tasks and material scheduling plans; for example, it automatically updates high-altitude operation patrol routes based on changes in steel arrival times, optimizing resource allocation efficiency.

[0124] Optionally, refer to Figure 7 Another embodiment of the present invention provides a security patrol method, based on the above. Figure 1 The illustrated embodiment identifies violations, equipment malfunctions, and environmental risks in the real-time inspection data and generates corresponding risk level results, including steps S310-S340, wherein:

[0125] S310. Call the pre-trained violation recognition model to perform frame-level analysis on the photos or videos in the real-time patrol data to generate corresponding analysis results;

[0126] S320. When the analysis result indicates a violation, extract the facial features and violation timestamp of the violator to generate structured data containing employee ID and violation type.

[0127] S330. Obtain equipment operation data and environmental monitoring data at the construction site;

[0128] S340. Input the structured data, the equipment operation data and the environmental monitoring data into the risk level assessment model to generate the corresponding risk level result.

[0129] The pre-trained violation recognition model refers to a deep learning model built using multi-dimensional training data. It can be implemented using a composite training method combining convolutional neural networks with safety helmet wearing detection, high-altitude work safety belt binding status recognition, and hot work permit verification rules. This model is used to extract violation features from images or videos. The structured data refers to violation records integrated through standardized fields. This can be achieved by blurring facial features and associating them with employee IDs and timestamps, ensuring traceability of violations while protecting privacy. The risk level assessment model is a comprehensive analysis algorithm that integrates multi-source data. It can be implemented using a dynamic weighting method for equipment operating parameters and environmental monitoring indicators, used to quantitatively assess the impact of risk events.

[0130] The violation identification model uses a convolutional neural network to detect violations frame-by-frame in the video stream, identifying actions such as not wearing a safety helmet or seatbelt, and verifies compliance with the hot work permit database. When a violation is detected, the inspection device automatically captures the relevant footage, extracts the facial area of ​​the violator, blurs it, and retains the employee's ID and timestamp to generate a structured record. Equipment operation data is collected in real time via IoT sensors, and environmental monitoring data includes parameters such as temperature, humidity, and gas concentration. After standardization, this data is input into the risk level assessment model. The model calculates a comprehensive risk value based on preset weighting coefficients and outputs a low, medium, or high risk assessment result.

[0131] Existing methods rely on manual visual inspection of violations, which cannot correlate with real-time device and environmental data, and violation records are mostly unstructured text. This embodiment achieves multi-dimensional data fusion analysis through an automated model, eliminating subjective errors in manual recording while preserving the traceability of behavior. Existing technologies typically store unprocessed facial images directly, posing a risk of privacy leakage; this embodiment employs blurring technology to ensure personal information security.

[0132] Through the aforementioned technical means, this embodiment can automatically identify complex risk factors at construction sites, perform real-time correlation analysis of personnel behavior, equipment status, and environmental parameters, and generate standardized risk level assessment results. The structured data storage method satisfies the traceability requirements of safety supervision while preventing the leakage of sensitive information. The integrated processing of multi-source data effectively improves the accuracy of risk assessment, providing a reliable basis for subsequent emergency response.

[0133] Optionally, refer to Figure 8 Another embodiment of the present invention provides a security patrol method, based on the above.Figure 7 The embodiment shown generates a corresponding visual alarm signal from the risk level result, specifically through steps S410-S440, wherein:

[0134] S410. Match the preset alarm template library according to the risk level results;

[0135] S420: Perform spatial enhancement processing on alarm signals, overlay the coordinates of the hidden danger location onto the BIM model of the construction site, and mark the association relationship with the affected equipment in the editable interface;

[0136] S430. If it is an environmental risk alarm, display links to historical similar accident handling cases and emergency material inventory status to the corresponding display position on the editable interface.

[0137] S440, and generate alarm icons of different colors and different flashing frequencies on the editable interface according to the risk level results.

[0138] The alarm template library refers to a pre-established collection of standardized alarm formats covering three types: equipment failure, environmental exceedance, and personnel violation. It can be implemented using XML or JSON data structures and is used to automatically adapt alarm content to different risk types. Spatial enhancement processing refers to the technology of fusing two-dimensional hazard coordinates with three-dimensional BIM models. This can be achieved using a GIS coordinate conversion interface and is used to accurately locate risk points in a three-dimensional scene. Hazard location coordinates refer to the latitude and longitude data of the hazard location obtained through mobile terminal GPS or indoor positioning inspection devices. This can be achieved using Bluetooth beacons or UWB positioning technology and is used to associate alarm signals with physical spatial locations. The BIM model refers to a three-dimensional building information model of the construction site, which can be built using Revit or Tekla software and is used to visualize the construction scene structure. Historical similar accident handling case links refer to hyperlinks to historical accident handling records similar to the current environmental risk. This can be achieved using database association query technology and is used to quickly retrieve reference cases. Emergency material inventory status refers to the real-time quantity information of protective equipment, fire-fighting equipment, and other materials currently in stock in the project. This can be obtained through IoT sensors or inventory management inspection device interfaces and is used to assess the adequacy of emergency response resources. Alarm icons with different colors and flashing frequencies are visual identifiers that are dynamically adjusted according to the risk level. They can be implemented using RGB color value encoding and timer controls to distinguish alarm priorities.

[0139] Once the risk level result is generated, the inspection device first matches the corresponding equipment failure, environmental exceedance, or personnel violation template from the alarm template library to ensure a consistent alarm content format. The coordinates of the hazard location are obtained through the positioning module and spatially matched with the component coordinates in the BIM model to form alarm markers in the 3D scene. These markers are then displayed as lines connecting the affected equipment or areas in the editable interface. For environmental risk alarms, the inspection device automatically retrieves similar incident handling records from the database, generates clickable case links, and simultaneously calls the real-time data interface of the materials management inspection device to display emergency material inventory levels. Alarm icons are colored red, orange, and yellow according to the risk level, with the flashing frequency increasing as the risk level rises, allowing operators to quickly identify the severity of the alarm through visual characteristics.

[0140] Compared to existing technologies, traditional methods rely on manual labeling of hazard locations and display alarm information only as text or a single-color icon, failing to link 3D spatial coordinates and historical response cases. This embodiment achieves precise 3D positioning of alarm signals by overlaying hazard coordinates onto a BIM model, solving the problem of traditional 2D plans failing to intuitively display the scope of risk impact. Furthermore, the embedded links to historical cases and material inventory status within the environmental risk alarms avoid response delays caused by manual queries, providing direct data support for emergency response. Dynamically adjusted alarm icons, through a combination of color and flashing frequency, overcome the technical deficiency of single visual identifiers in distinguishing priorities.

[0141] This embodiment achieves precise spatial positioning of alarm signals at the construction site, making the correlation between risk points and affected equipment intuitively visible; by associating with historical similar accident cases and real-time material inventory data, it provides operators with a quick reference for handling; through the dynamic combination of color and flashing frequency, alarm information of different risk levels is clearly distinguishable in the visualization interface, making it easier to prioritize the handling of high-risk events.

[0142] The present invention also proposes an inspection device, the inspection device comprising: a memory, a processor, and a security inspection program stored in the memory and executable on the processor, the security inspection program being configured to implement the security inspection method as described above.

[0143] It is worth noting that since the inspection device of the present invention is based on the above-described security inspection method, the embodiments of the inspection device of the present invention include all the technical solutions of all embodiments of the above-described security inspection method, and the technical effects achieved are exactly the same, so they will not be repeated here.

[0144] The present invention also proposes an inspection device, which includes the inspection apparatus as described in the above embodiments.

[0145] It is worth noting that since the inspection device of the present invention is based on the above-mentioned inspection apparatus, the embodiments of the inspection device of the present invention include all the technical solutions of all the embodiments of the above-mentioned inspection apparatus, and the technical effects achieved are exactly the same, so they will not be repeated here.

[0146] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0147] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0148] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0149] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A security patrol method, characterized in that, The security patrol method includes: In response to the operation instructions of the editable interface, a multi-level process template for building safety inspection is configured, and the template parameters of the multi-level process template are adjusted according to the operation instructions to generate a corresponding safety inspection task. The multi-level process template includes modules for customer management, risk assessment, approval input, and supply chain management. According to the safety patrol task, real-time patrol data of the construction site is collected by the shooting device of the patrol personnel's mobile terminal. The real-time patrol data includes at least one of images, videos or sensor data. Identify violations, equipment malfunctions, and environmental risks in the real-time inspection data, and generate corresponding risk level results; The risk level results are used to generate corresponding visual alarm signals, and the visual alarm signals are sent to the mobile terminals that collect the real-time patrol data based on LoRa spread spectrum technology. When the visual alarm signal is an emergency signal, the visual alarm signal is sent to the monitoring center through the 5G / 4G network. Monitor the response information of the corresponding mobile terminal, update the status of the security patrol task, and record the operation log of the patrol personnel in handling the risk level results; If the mobile terminal does not respond to the visual alarm signal within a preset time, the alarm information will be sent to the mobile terminal again through the backup communication method, and an alarm message will be sent to the monitoring center. The step of adjusting the template parameters of the multi-level process template according to the operation instructions to generate the corresponding security inspection task is as follows: Obtain the current project's progress data, resource allocation table, historical risk distribution map, and material scheduling plan, and extract the first parameter from the multi-level process template. The first parameter includes the contract performance period, the hidden danger weight coefficient, and the budget allocation ratio. Determine the deviation between the current project's progress data and the contract performance period, the matching degree between the resource allocation table and the budget allocation ratio, and the correlation between the historical risk distribution map and the hidden danger weight coefficient; The deviation threshold range of the first parameter is determined based on the deviation value, the matching degree, and the correlation. The deviation threshold range of the first parameter is pushed to the editable interface, and the parameter adjustment operation of the editable interface is monitored. The template parameters of the multi-level process template are updated, and the updated template parameters are associated and bound with the material scheduling plan. The template parameters bound to the material scheduling plan are logically verified, and a safety inspection task containing task priority, inspection route planning and resource allocation instructions is generated after successful verification. Determining the deviation threshold range of the first parameter based on the deviation value, the matching degree, and the correlation includes: Generate an initial deviation threshold corresponding to the deviation value; The priority weight of resource allocation corresponding to the matching degree is calculated by weighted calculation method and mapped to the initial deviation threshold in order to adjust and generate the second deviation threshold. The probability distribution of potential hazards corresponding to the correlation is predicted by a Bayesian probability model and fused into a second deviation threshold to determine the deviation threshold range of the first parameter.

2. The security patrol method as described in claim 1, characterized in that, The operation commands in response to the editable interface configure a multi-level workflow template for building safety inspections, specifically as follows: Obtain a standardized process template library corresponding to the building type or engineering stage, the standardized process template library including risk management processes for residential buildings, commercial buildings, industrial plants and different construction stages; In response to the operation instructions of the editable interface, process nodes are added or deleted in the standardized process template library, and corresponding trigger conditions are configured for each modified process node. The trigger conditions include environmental thresholds, equipment status thresholds and manual judgment rules. For each process node, an executor role is bound, and the operation permissions, task execution period, and timeout alarm rules corresponding to the executor role are determined. The triggering conditions, operation permissions, task execution deadlines, and timeout alarm rules are imported into the standardized process template library to generate the multi-level process template adapted to the current project.

3. The security patrol method as described in claim 2, characterized in that, The specific steps involve configuring corresponding trigger conditions for each modified process node as follows: Obtain historical inspection data for the current project and a database of risk cases from similar projects; The risk type and trigger probability are calculated by using a time-series prediction model on the historical inspection data and the risk case library of similar projects to generate corresponding prediction results. The prediction result is associated with the triggering condition; The step of importing the triggering conditions, operation permissions, task execution deadlines, and timeout alarm rules into the standardized process template library to generate the multi-level process template adapted to the current project includes: The triggering conditions are mapped to the corresponding process nodes so that the process nodes are triggered automatically. When the triggering conditions are mapped to the corresponding process nodes, task execution permissions are assigned to different roles according to the operation permissions, so as to generate and determine the task execution period and timeout alarm rules corresponding to the task execution permissions, and generate the multi-level process template.

4. The security patrol method as described in claim 1, characterized in that, The step of associating and binding the updated template parameters with the material scheduling plan specifically involves: By using graph modeling technology, a topological relationship is established between the updated template parameters and the key nodes in the material scheduling plan; Real-time monitoring triggers are embedded in the topology relationship, and dynamic verification of the template parameters is automatically triggered when the status of key nodes in the material scheduling plan changes. And adjust the template parameters in real time based on the results of dynamic verification.

5. The security patrol method as described in claim 1, characterized in that, The process of identifying violations, equipment malfunctions, and environmental risks in the real-time inspection data, and generating corresponding risk level results, includes: The pre-trained violation recognition model is invoked to perform frame-level analysis on the photos or videos in the real-time patrol data to generate corresponding analysis results. When the analysis result indicates a violation, the facial features and violation timestamp of the violator are extracted to generate structured data containing the employee's employee number and violation type. Obtain equipment operation data and environmental monitoring data at the construction site; The structured data, equipment operation data, and environmental monitoring data are input into the risk level assessment model to generate the corresponding risk level results.

6. The security patrol method as described in claim 5, characterized in that, The step of generating a corresponding visual alarm signal from the risk level result is specifically as follows: The alarm template library is matched with the risk level results. The alarm template library includes equipment failure alarm templates, environmental exceedance alarm templates, and personnel violation alarm templates. Spatial enhancement processing is performed on alarm signals, overlaying the coordinates of the hidden danger location onto the BIM model of the construction site, and marking the association with the affected equipment in the editable interface; In the event of an environmental risk alarm, links to historical similar accident handling cases and the status of emergency supplies inventory will be displayed at the corresponding display position on the editable interface. Based on the risk level results, alarm icons with different colors and flashing frequencies are generated on the editable interface.

7. A patrol device, characterized in that, The inspection device includes: a memory, a processor, and a security inspection program stored in the memory and executable on the processor, the security inspection program being configured to implement the security inspection method as described in any one of claims 1 to 6.

8. An inspection device, characterized in that, Includes the patrol device as described in claim 7.

Images