A Smart Fire Safety Inspection System and Method Based on AI and AR Integration

Abstract

This invention relates to the field of smart fire protection technology, and in particular to a smart fire protection inspection system and method based on the fusion of AI and AR. The method includes the following steps: collecting multi-source associated data related to building fire protection, and cleaning and standardizing the multi-source associated data to form a unified format fire protection inspection basic dataset; overlaying and fusing the fire protection inspection basic dataset with information from the actual building scene, and constructing an AR real-scene 3D model containing the relationship between fire protection facilities and building structure using augmented reality technology; based on a preset fire protection code knowledge base, using artificial intelligence algorithms to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model, identifying and marking potential compliance issues; retrieving potential compliance issue points through augmented reality devices, and performing a cross-verification operation of virtual and real fusion in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues.

Description

Technical Field

[0001] This invention relates to the field of smart fire protection technology, and in particular to a smart fire protection inspection system and method based on the fusion of AI and AR. Background Technology

[0002] With the acceleration of urbanization, buildings are becoming increasingly larger and more complex, and the types and quantities of fire protection facilities are constantly increasing, placing higher demands on the professionalism, efficiency, and accuracy of fire protection inspections. Traditional fire protection inspections rely on manual processes, which have the following problems: The data sources involved, such as building structure data, fire protection facility parameters, and fire lane dimensions, are scattered (including paper drawings, electronic documents, and on-site measurement data). The formats are inconsistent, the redundancy is high, and there are outliers. Manual sorting and preprocessing are inefficient and prone to data omissions or errors, which affects the reliability of the basic data for verification. Relying on two-dimensional drawings and on-site visual observation makes it difficult to intuitively present the spatial relationship between fire protection facilities and building structure. Judgment of key compliance indicators such as fire separation distance and facility installation height depends on personnel experience, which is highly subjective and prone to misjudgment and omission. At the same time, fire protection regulations are numerous and frequently updated, and manual search and matching are inefficient, making it difficult to ensure compliance and timeliness of the review. The rectification suggestions are mostly in text description, making it difficult for construction personnel to intuitively understand the implementation path, resulting in low rectification efficiency and poor results; the data throughout the process is stored in a scattered manner, making it difficult to trace the source, and it is difficult to trace the responsible party and operation process after a problem occurs, which is not conducive to subsequent operation and maintenance management and responsibility identification.

[0003] Existing fire safety inspection systems only achieve electronic data management and do not deeply integrate AI and AR technologies, making it impossible to automatically analyze compliance indicators and perform virtual-real integrated verification and rectification display. On the other hand, simple AR visualization or AI data analysis technologies lack adaptability to the entire fire safety inspection process and are unable to solve the systemic problems of traditional inspections. Summary of the Invention

[0004] Therefore, it is necessary to provide a smart fire protection inspection system and method based on the integration of AI and AR to address the above-mentioned technical problems. This system and method can effectively eliminate redundant and abnormal data by using standardized preprocessing and intelligent cleaning algorithms to address the pain points of scattered and disordered multi-source data. The data processing efficiency is significantly improved compared to manual processing, and the accuracy of basic data is significantly improved. This lays a high-quality data foundation for the inspection work and avoids inspection deviations caused by data errors.

[0005] This invention provides a smart fire safety inspection method based on the fusion of AI and AR, the method comprising: Collect multi-source correlated data related to building fire protection, and clean and standardize the multi-source correlated data to form a basic dataset for fire protection inspection in a unified format; By overlaying and integrating the basic dataset of fire safety inspection with information from actual building scenes, an AR real-scene 3D model containing the relationship between fire safety facilities and building structure is constructed using augmented reality technology. Based on a pre-set fire safety code knowledge base, artificial intelligence algorithms are used to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model to identify and mark potential compliance issues. By using augmented reality devices to retrieve potential compliance issues, a cross-verification operation that blends the virtual and real elements is performed in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues. Based on the compliance issues, the corresponding fire safety regulations are matched to generate targeted rectification suggestions, and the implementation path of the rectification suggestions is visualized using augmented reality technology. Data and information from the entire fire safety inspection process are collected, standardized electronic inspection files are generated, and a full-process data traceability link is established to enable the query and traceability of fire safety inspection results.

[0006] In one embodiment, the collection of multi-source correlated data related to building fire protection, followed by cleaning and standardization preprocessing of the multi-source correlated data to form a unified format fire safety inspection basic dataset, includes: By scanning drawings, collecting sensor data, taking on-site photos, and connecting with systems, we collect multi-source data including fire protection facility parameters, building structure data, and fire lane dimensions. Data deduplication and outlier removal algorithms were used to clean the multi-source data, and the field formats were standardized according to the unified data standards of the fire protection industry to generate a structured basic dataset.

[0007] In one embodiment, the process of overlaying and fusing the basic fire safety inspection dataset with on-site building scene information to construct an AR real-scene 3D model containing the relationship between fire safety facilities and building structure using augmented reality technology includes: The basic dataset is precisely matched and overlaid with the latitude, longitude, and elevation information of the actual scene; Based on the AR real-time rendering engine, the system associates the installation location and model of fire protection facilities with the structure of building beams, columns, and walls to construct a realistic 3D model.

[0008] In one embodiment, the method of automatically comparing and analyzing multiple compliance indicators in an AR real-world 3D model based on a preset fire safety code knowledge base and using artificial intelligence algorithms to identify and mark potential compliance issues includes: Extract core compliance indicator thresholds, including fire separation distance, facility installation height, and passage width, from the fire safety code knowledge base; A deep learning comparison algorithm is used to match the indicators and thresholds in the AR model one by one, and mark suspected non-compliance points and related regulatory clauses.

[0009] In one embodiment, the step of retrieving potential compliance issue points using augmented reality devices and performing cross-verification (hybridization) within a real-world building scene to confirm the authenticity and specific spatial location information of the potential compliance issues includes: By using AR devices to retrieve marked problem point models, a virtual-real overlay display can be achieved in real-world scenarios; Laser ranging and image comparison tools are used to verify whether the problem actually exists, and the three-dimensional spatial coordinates of the problem and the surrounding environment information are recorded.

[0010] In one embodiment, the step of matching corresponding fire safety regulations based on compliance issues, generating targeted rectification suggestions, and visually demonstrating the implementation path of the rectification suggestions using augmented reality technology includes: Based on the identified compliance issues, the fire safety regulations clauses are matched, and a plan including rectification measures, material selection, and compliance standards is developed. In the AR real-world environment, the rectification construction sequence, key nodes, and final effect are visually presented using dynamic arrows and highlighted annotations.

[0011] In one embodiment, the process of collecting data information from the entire fire safety inspection process, generating standardized electronic inspection files, and establishing a full-process data traceability link includes: Integrate information from the entire process, including data collection, model building, compliance analysis, verification records, and rectification results, and generate electronic verification files in a standardized format; Add timestamps, operator and equipment number identifiers to the data at each stage, and build a chain traceability link. The chain traceability link supports querying and tracing by file number and time range.

[0012] This invention also provides a smart fire safety inspection system based on the fusion of AI and AR, the system comprising: The data standardization module is used to collect multi-source associated data related to building fire protection, and to clean and standardize the multi-source associated data to form a unified format fire protection inspection basic dataset. The modeling and fusion module is used to overlay and fuse the basic dataset of fire protection inspection with the information of the actual building scene, and to build an AR real-scene 3D model containing the relationship between fire protection facilities and building structure through augmented reality technology; The indicator analysis module is used to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model based on a preset fire protection code knowledge base and using artificial intelligence algorithms to identify and mark potential compliance issues. The verification module is used to retrieve potential compliance issues through augmented reality devices and perform cross-verification operations that blend virtual and real elements in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues. The visualization module is used to match the corresponding fire safety regulations based on compliance issues, generate targeted rectification suggestions, and use augmented reality technology to visualize the implementation path of the rectification suggestions. The management module is used to collect data and information from the entire fire safety inspection process, generate standardized electronic inspection files, and establish a full-process data traceability link to enable the querying and traceability of fire safety inspection results. The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the intelligent fire protection inspection method based on AI and AR fusion as described above.

[0013] The present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the intelligent fire protection inspection method based on the fusion of AI and AR as described above.

[0014] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the intelligent fire protection inspection method based on the fusion of AI and AR as described above.

[0015] The aforementioned intelligent fire safety inspection system and method based on the integration of AI and AR addresses the pain points of scattered and disorganized multi-source data. Through standardized preprocessing and intelligent cleaning algorithms, it effectively eliminates redundant and abnormal data, significantly improving data processing efficiency compared to manual methods and substantially increasing the accuracy of basic data. This lays a high-quality data foundation for the inspection work and avoids inspection deviations caused by data errors. The AR real-scene 3D model intuitively presents the spatial relationship between fire protection facilities and building structures. Combined with AI algorithms, it automatically compares compliance indicators without relying on human experience, resulting in higher accuracy in compliance identification. At the same time, it automatically matches updated fire protection code clauses, solving the problems of low efficiency and poor adaptability of manual retrieval, ensuring the compliance and timeliness of the inspection. AR virtual-real cross-verification enables accurate correspondence between model problems and actual scenarios. This approach avoids misjudgments caused by the disconnect between models and actual conditions, resulting in more accurate problem location. The visualized rectification path reduces the understanding cost for construction personnel in a dynamic manner, significantly improving rectification efficiency and ensuring the effective implementation of rectification measures. It collects data from all stages to generate standardized electronic archives, combined with a chain-like traceability system, clearly defining the responsible parties and operational processes at each stage. This meets the compliance and regulatory requirements of fire safety inspections, providing solid data support for subsequent operation and maintenance management and responsibility identification, and solving the drawbacks of traditional inspection data being scattered and difficult to trace. For the first time, it achieves deep integration of AI and AR technologies throughout the entire fire safety inspection process, rather than a single technology application, completely changing the traditional inspection model of manual labor plus two-dimensional drawings. This promotes the transformation of fire safety inspections from manual to intelligent driving, adapting to the fire safety inspection needs of modern complex buildings and possessing broad industry promotion value. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this 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 some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 A flowchart of the intelligent fire protection inspection method based on the fusion of AI and AR provided by this invention; Figure 2 A module diagram of the intelligent fire protection inspection system based on the fusion of AI and AR provided by the present invention; Figure 3 An internal structural diagram of the computer device provided by the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. 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.

[0019] The following is combined with Figures 1 to 3 This invention describes a smart fire safety inspection system and method based on the fusion of AI and AR.

[0020] In one embodiment, a smart fire safety inspection method based on the fusion of AI and AR includes the following steps: Step S100: Collect multi-source correlation data related to building fire protection, and clean and standardize the multi-source correlation data to form a fire protection inspection basic dataset in a unified format. Step S200: Overlay and fuse the basic dataset of fire protection inspection with the information of the actual building scene, and construct an AR real-scene 3D model containing the relationship between fire protection facilities and building structure through augmented reality technology; Step S300: Based on the preset fire safety code knowledge base, use artificial intelligence algorithms to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model, identify and mark potential compliance issues; Step S400: Using augmented reality devices, potential compliance issues are retrieved, and a cross-verification operation that blends virtual and real elements is performed in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues. Step S500: Match the corresponding fire safety regulations based on the compliance issues, generate targeted rectification suggestions, and use augmented reality technology to visualize the implementation path of the rectification suggestions. Step S600: Collect data and information from the entire fire safety inspection process, generate standardized electronic inspection files, and establish a full-process data traceability link to enable the fire safety inspection results to be queryable and traceable.

[0021] The aforementioned intelligent fire safety inspection method based on the integration of AI and AR addresses the pain points of scattered and disorganized multi-source data. Through standardized preprocessing and intelligent cleaning algorithms, it effectively eliminates redundant and abnormal data, significantly improving data processing efficiency compared to manual methods and substantially increasing the accuracy of basic data. This lays a high-quality data foundation for the inspection work and avoids inspection deviations caused by data errors. The AR real-scene 3D model intuitively presents the spatial relationship between fire protection facilities and building structures. Combined with AI algorithms, it automatically compares compliance indicators without relying on human experience, resulting in higher accuracy in compliance identification. Simultaneously, it automatically matches updated fire protection regulations, solving the problems of low efficiency and poor adaptability of manual retrieval, ensuring the compliance and timeliness of the inspection. AR virtual-real cross-verification achieves accurate correspondence between model issues and actual scenarios. To avoid misjudgments caused by the disconnect between models and actual conditions, the system achieves higher accuracy in problem location. A visualized rectification path, presented dynamically, reduces the understanding cost for construction personnel, significantly improving rectification efficiency and ensuring the effective implementation of rectification measures. Data from all stages is collected to generate standardized electronic archives, which, combined with a chain-like traceability system, clearly define the responsible parties and operational processes at each stage, meeting the compliance and regulatory requirements of fire safety inspections. This provides solid data support for subsequent operation and maintenance management and liability identification, addressing the shortcomings of traditional inspection data being scattered and difficult to trace. For the first time, AI and AR technologies are deeply integrated into the entire fire safety inspection process, rather than being a single technology application. This completely changes the traditional inspection model based on manual labor and two-dimensional drawings, promoting the transformation of fire safety inspections from manual to intelligent, adapting to the fire safety inspection needs of modern complex buildings, and possessing broad industry promotion value.

[0022] In one embodiment, multi-source correlated data related to building fire protection are collected, and the multi-source correlated data is cleaned and standardized preprocessed to form a basic dataset for fire protection inspection in a unified format, including the following steps: Step S110 involves collecting multi-source data, including fire protection facility parameters, building structure data, and fire lane dimensions, through methods such as drawing scanning, sensor acquisition, on-site photography, and system integration.

[0023] Specifically, high-precision scanners are used to scan paper drawings such as building fire protection design drawings, as-built drawings, and facility layout drawings to generate clear electronic images; mature OCR text recognition technology is used to extract structured information such as fire protection facility models, dimensions, and installation locations from the drawings. Information that is not successfully recognized is marked and manually entered to ensure the completeness of information collection. Deploy an IoT sensor network, including ranging sensors, temperature and humidity sensors, and smoke sensors adapted to fire scenarios, to collect real-time data on fire lane dimensions, facility operating parameters, and building environment. Encrypt the data and transmit it to the data center through stable network transmission technology to ensure the security and real-time performance of data transmission. High-definition drones equipped with positioning modules were used to take aerial photos of the building's external fire protection facilities to ensure accurate positioning of the shooting location; AR shooting equipment adapted to on-site operations was used to record the condition of the building's internal facilities and simultaneously collect relevant spatial information of the shooting location to provide support for subsequent data matching. It interfaces with BIM systems, fire protection operation and maintenance management systems, and urban fire protection remote monitoring systems through standardized API interfaces, and transmits data using secure encrypted transmission protocols to obtain relevant information such as three-dimensional data of building structures, factory parameters of facilities, and historical operation and maintenance records, ensuring the comprehensiveness of data sources.

[0024] Step S120: Data deduplication and outlier removal algorithms are used to clean the multi-source data, and the field format is standardized according to the unified data standard of the fire protection industry to generate a structured basic dataset.

[0025] Specifically, a mature hash algorithm is used to calculate a unique identifier for each piece of data. Entries with the same identifier are compared, and records with better data integrity are retained. For structured data, deduplication is performed using a combination of core fields as a composite primary key, effectively eliminating duplicate records. We use industry-standard outlier detection algorithms to process numerical data, and combine this with knowledge of the fire protection industry to manually review the marked outliers. Once confirmed to be invalid, they are removed. For non-numerical data, we use regular expressions to match preset formats and remove invalid data that does not match the format, thus ensuring data validity. A unified field specification was developed based on relevant general standards in the fire protection industry, clarifying the standards for field names, data types, units of measurement, time formats, etc.; an extraction algorithm adapted to drawing data was used to extract facility-related information from scanned drawings and map it to standardized fields; finally, a structured basic dataset was generated, supporting batch querying and retrieval, improving the convenience of data use.

[0026] In one embodiment, the fire safety inspection dataset is overlaid and fused with on-site building scene information to construct an AR real-scene 3D model containing the relationship between fire safety facilities and building structure using augmented reality technology, including the following steps: Step S210 involves accurately matching and overlaying the basic dataset with the latitude, longitude, and elevation information of the actual scene.

[0027] Specifically, real-time dynamic positioning technology with high positioning accuracy is used to obtain data such as the latitude, longitude, and elevation of the building. Multiple benchmark control points are set at key locations of the building, and the three-dimensional coordinates of the control points are recorded. Laser scanning technology is used to scan the interior and exterior scenes of the building to generate a high-density three-dimensional terrain and building outline model, providing an accurate scene basis for data matching. The coordinates in the basic dataset are converted into a coordinate system consistent with the actual scene, and a professional coordinate calibration algorithm is used for calibration to ensure that the data corresponds accurately with the actual spatial location and the calibration error is controlled at a low level. A mature point cloud registration algorithm is used to register the calibrated basic dataset with the point cloud model generated by laser scanning, so that the parameters of fire protection facilities are accurately aligned with the actual physical entities, resulting in better registration effect.

[0028] Step S220: Based on the AR real-time rendering engine, associate the installation location and model of fire protection facilities with the structure of building beams, columns and walls to construct a real-world 3D model.

[0029] Specifically, a real-time rendering engine adapted for AR applications is used to build a 3D scene framework, supporting real-time rendering of lighting, shadows, and material textures to ensure smooth rendering; mainstream AR development SDKs are integrated to achieve real-time positioning and posture tracking between mobile devices and real-world scenes, ensuring the stability of virtual-real fusion. Extract data such as the installation location, model, and connection relationship of fire protection facilities, as well as the dimensions, materials, and spatial locations of building beams, columns, and walls from the basic dataset, and create full-scale 3D model instances in the rendering engine; establish logical relationships between facilities and building structures through professional model association algorithms, and store the associated data in the model attribute panel for easy subsequent querying and analysis; Detail-level optimization technology is used to improve rendering efficiency, and the model accuracy is dynamically adjusted according to the observation distance to balance rendering effect and running efficiency; collision detection is optimized to avoid problems such as model penetration when the virtual and real scenes are superimposed; the final model supports multiple AR devices and has low latency when superimposed with the real scene to ensure a good user experience.

[0030] In one embodiment, based on a pre-defined fire safety code knowledge base, artificial intelligence algorithms are used to automatically compare and analyze multiple compliance indicators in an AR real-world 3D model to identify and mark potential compliance issues, including the following steps: Step S310: Extract core compliance indicator thresholds, including fire separation distance, facility installation height, and passage width, from the fire protection code knowledge base.

[0031] Specifically, an ontological approach is adopted, using core fire protection industry standards as the data source to extract key information such as entities, attributes, and relationships from the standards. This information is then formally described using a standardized knowledge description language and stored in an appropriate database. Web crawler technology is used to regularly capture updated standards information, and text classification algorithms are used to identify updated clauses and automatically replace old clauses, ensuring the timeliness of the knowledge base. By employing efficient information extraction algorithms, core compliance indicators such as fire separation distance, facility installation height, and passage width are extracted from the knowledge base. The threshold range and applicable scenario labels of each indicator are clarified, and a standardized indicator threshold table is generated to provide a basis for subsequent comparative analysis.

[0032] Step S320: Using a deep learning comparison algorithm, the indicators in the AR model are matched with the thresholds one by one, and suspected non-compliance points and related regulatory clauses are marked.

[0033] Specifically, advanced target detection algorithms are used to extract actual indicator values ​​from AR models, resulting in high detection accuracy and faster extraction speed, meeting the needs of batch processing; a high-performance deep learning model is used to build a comparison model, and compliance matching measures the degree of fit between quantifiable indicators and regulatory requirements.

[0034] The formula for calculating compliance matching degree is: in, Indicates the compliance matching degree (value from 0 to 1). For compliance, Suspected non-compliance This is non-compliant; This represents the actual value of the indicator extracted from the AR model; Mean ( When there is no upper limit value ( This is the lower limit value. This is the upper limit value. (for the optimal value) Represents the feature vector of the applicable scenario (one-hot encoded form); , The model weight parameters are obtained by training and fitting through a large amount of historical verification data to ensure the accuracy of model analysis. This represents the model bias term, used to adjust the baseline of the algorithm output.

[0035] In one embodiment, potential compliance issues are identified using augmented reality devices, and a cross-verification operation is performed in a real-world building setting to confirm the authenticity and specific spatial location of the potential compliance issues. This includes the following steps: Step S410: Use an AR device to retrieve the marked problem point model and display it in a real-world scene using a virtual-real overlay.

[0036] Specifically, stable AR smart devices that support high-speed networks are selected. These devices have built-in high-precision positioning, attitude tracking, and high-definition shooting modules to meet on-site verification requirements. The AR models of problem points are streamed and loaded via a high-speed network connection to the data center, resulting in low loading latency and ensuring smooth operation. AR devices automatically scan reference control points and calibrate the model's posture using mature feature matching algorithms to ensure minimal overlay deviation; they also support manual fine-tuning of the model's position and scaling to achieve precise alignment between the model and the real-world scene.

[0037] Step S420: Use laser ranging and image comparison tools to verify whether the problem actually exists, and record the three-dimensional spatial coordinates of the problem and the surrounding environment information.

[0038] Specifically, the AR device integrates a high-precision laser ranging module, which measures the actual distance to the problem point and compares it with the model index value. The data consistency is judged based on the difference. If the difference exceeds the reasonable range, the data is collected again. Real-world scene images are captured, and image feature points are extracted using an efficient image feature matching algorithm. These images are then compared with the rendered images from the model, and scene consistency is determined based on the matching degree. For concealed issues, an auxiliary detection module can be used for auxiliary verification to improve the accuracy of the verification. The system automatically records the three-dimensional coordinates of the problem points, multi-angle on-site photos, notes from the inspectors, and on-site environmental data. All information is stored in the verification record table to ensure the completeness and relevance of the information.

[0039] In one embodiment, based on the compliance issues, corresponding fire safety regulations are matched to generate targeted rectification suggestions. Augmented reality technology is then used to visualize the implementation path of these suggestions, including the following steps: Step S510: Match the identified compliance issues with the fire safety regulations and develop a plan that includes corrective measures, material selection, and compliance standards.

[0040] Specifically, based on the identified problem type, semantic retrieval algorithms are used to quickly match the corresponding rectification requirements in the knowledge base, and feasible rectification directions are selected in combination with the actual situation of the real-world scenario to ensure the adaptability of the rectification suggestions; A comprehensive rectification plan is generated using a scientific decision-making algorithm, including specific rectification measures, appropriate material selection, clear compliance standards, reasonable construction period and cost estimation, etc. After the plan is generated, it is pushed to the terminal of professional personnel for review, modification and improvement to form the final plan, ensuring the feasibility and professionalism of the plan.

[0041] Step S520: Visually present the rectification construction sequence, key nodes and final effect in the AR real scene using dynamic arrows and highlighted annotations.

[0042] Specifically, in the AR real-world scene, the construction sequence is marked with intuitive dynamic arrows, and the speed of the arrows is adapted to the construction rhythm; key construction nodes are marked with eye-catching flashing marks, and the construction requirements of the nodes are clearly marked to facilitate the understanding of construction personnel. It supports switching between multiple views: "Current Status - Under Construction - After Rectification". The post-rectification view uses AR rendering technology to display the final effect. Clicking on key parts allows you to view the core indicator data after rectification. It supports real-time measurement and verification, helping construction personnel to predict the rectification effect in advance.

[0043] In one embodiment, data information from the entire fire safety inspection process is collected, standardized electronic inspection files are generated, and a full-process data traceability chain is established, including the following steps: Step S610: Integrate the information from the entire process, including data collection, model building, compliance analysis, verification records, and rectification results, and generate electronic verification files in a standardized format.

[0044] Specifically, it comprehensively integrates data from the entire process, including data collection, model building, compliance analysis, verification records, and rectification results, and systematically organizes the data according to the classification logic of "project-stage-data type" to ensure clear data classification; In accordance with relevant management regulations in the fire protection industry, electronic inspection files are generated using a common file format. The files include complete content such as a cover, table of contents, data details, inspection conclusions, and electronic signature pages. The files also have a built-in efficient indexing function, supporting keyword search and quick navigation, thus improving the ease of use of the files.

[0045] Step S620: Add timestamps, operator and equipment number identifiers to the data of each link, and build a chain traceability link. The chain traceability link supports querying and tracing by file number and time range.

[0046] Specifically, add complete identification information to the data packets at each stage, including timestamps accurate to milliseconds, unique operator IDs, device numbers, and tamper-proof hash values ​​to ensure that the data can be accurately located; Based on a secure and reliable blockchain consortium blockchain architecture, data from each stage is stored as independent blocks. Each block contains the hash value of the previous block, and participating nodes synchronize and share the ledger to ensure that the data cannot be tampered with. It provides multiple terminal query portals, supporting queries by file number, time range, project name, and other conditions; it clearly displays the entire data flow trajectory, and clicking on any link allows viewing the original data and detailed processing logs, achieving full-chain traceability and meeting the needs of supervision and accountability.

[0047] The intelligent fire protection inspection system based on the fusion of AI and AR provided by the present invention will be described below. The intelligent fire protection inspection system based on the fusion of AI and AR described below can be referred to in correspondence with the intelligent fire protection inspection method based on the fusion of AI and AR described above.

[0048] In one embodiment, a smart fire protection inspection system based on the fusion of AI and AR includes a data organization module, a modeling and fusion module, an indicator analysis module, a verification module, a visualization module, and a management module. The data standardization module is used to collect multi-source correlated data related to building fire protection, and to clean and standardize the multi-source correlated data to form a basic dataset for fire protection inspection in a unified format. The modeling and fusion module is used to overlay and fuse the basic dataset of fire protection inspection with the information of the actual building scene, and to build an AR real-scene 3D model containing the relationship between fire protection facilities and building structure through augmented reality technology; The indicator analysis module is used to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model based on a preset fire protection code knowledge base and using artificial intelligence algorithms to identify and mark potential compliance issues. The verification module is used to retrieve potential compliance issues through augmented reality devices and perform cross-verification operations that blend virtual and real elements in the actual building scene to confirm the authenticity and specific spatial location information of potential compliance issues. The visualization module is used to match the corresponding fire safety regulations based on compliance issues, generate targeted rectification suggestions, and use augmented reality technology to visualize the implementation path of the rectification suggestions. The management module is used to collect data and information from the entire fire safety inspection process, generate standardized electronic inspection files, and build a full-process data traceability link to enable the query and traceability of fire safety inspection results.

[0049] Figure 3 This example illustrates a schematic diagram of the physical structure of an electronic device, which can be a smart terminal. Its internal structure diagram can be as follows: Figure 3 As shown, the electronic device includes a processor, memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a smart fire inspection method based on the fusion of AI and AR.

[0050] Those skilled in the art will understand that Figure 3The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the electronic device to which the present invention is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0051] On the other hand, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements a smart fire protection inspection method based on the fusion of AI and AR.

[0052] On another front, a computer program product or computer program is provided, comprising computer instructions stored in a computer storage medium. The processor of an electronic device reads the computer instructions from the computer storage medium, and when the processor executes the computer instructions, it implements a smart fire safety inspection method based on the fusion of AI and AR.

[0053] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0054] By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0055] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0056] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A smart fire safety inspection method based on the fusion of AI and AR, characterized in that, The method includes: Collect multi-source correlated data related to building fire protection, and clean and standardize the multi-source correlated data to form a basic dataset for fire protection inspection in a unified format; By overlaying and integrating the basic dataset of fire safety inspection with information from actual building scenes, an AR real-scene 3D model containing the relationship between fire safety facilities and building structure is constructed using augmented reality technology. Based on a pre-set fire safety code knowledge base, artificial intelligence algorithms are used to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model to identify and mark potential compliance issues. By using augmented reality devices to retrieve potential compliance issues, a cross-verification operation that blends the virtual and real elements is performed in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues. Based on the compliance issues, the corresponding fire safety regulations are matched to generate targeted rectification suggestions, and the implementation path of the rectification suggestions is visualized using augmented reality technology. Data and information from the entire fire safety inspection process are collected, standardized electronic inspection files are generated, and a full-process data traceability link is established to enable the query and traceability of fire safety inspection results.

2. The intelligent fire safety inspection method based on AI and AR fusion according to claim 1, characterized in that, The process involves collecting multi-source correlated data related to building fire protection, cleaning and standardizing the data to form a unified formatted basic dataset for fire safety inspection, including: By scanning drawings, collecting sensor data, taking on-site photos, and connecting with systems, we collect multi-source data including fire protection facility parameters, building structure data, and fire lane dimensions. Data deduplication and outlier removal algorithms were used to clean the multi-source data, and the field formats were standardized according to the unified data standards of the fire protection industry to generate a structured basic dataset.

3. The intelligent fire safety inspection method based on AI and AR fusion according to claim 2, characterized in that, The process of overlaying and fusing basic fire safety inspection datasets with on-site building scene information, and constructing an AR real-scene 3D model containing the relationship between fire safety facilities and building structure using augmented reality technology, includes: The basic dataset is precisely matched and overlaid with the latitude, longitude, and elevation information of the actual scene; Based on the AR real-time rendering engine, the system associates the installation location and model of fire protection facilities with the structure of building beams, columns, and walls to construct a realistic 3D model.

4. The intelligent fire protection inspection method based on AI and AR fusion according to claim 3, characterized in that, The system, based on a pre-set fire safety code knowledge base, uses artificial intelligence algorithms to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model, identifying and marking potential compliance issues, including: Extract core compliance indicator thresholds, including fire separation distance, facility installation height, and passage width, from the fire safety code knowledge base; A deep learning comparison algorithm is used to match the indicators and thresholds in the AR model one by one, and mark suspected non-compliance points and related regulatory clauses.

5. The intelligent fire safety inspection method based on AI and AR fusion according to claim 4, characterized in that, The process of retrieving potential compliance issue points using augmented reality devices and performing cross-verification (fusing virtual and real elements) within a real-world building setting to confirm the authenticity and specific spatial location information of the potential compliance issues includes: By using AR devices to retrieve marked problem point models, a virtual-real overlay display can be achieved in real-world scenarios; Laser ranging and image comparison tools are used to verify whether the problem actually exists, and the three-dimensional spatial coordinates of the problem and the surrounding environment information are recorded.

6. The intelligent fire safety inspection method based on AI and AR fusion according to claim 5, characterized in that, The process involves matching corresponding fire safety regulations to compliance issues, generating targeted rectification suggestions, and visually demonstrating the implementation path of these suggestions using augmented reality technology. This includes: Based on the identified compliance issues, the fire safety regulations clauses are matched, and a plan including rectification measures, material selection, and compliance standards is developed. In the AR real-world environment, the rectification construction sequence, key nodes, and final effect are visually presented using dynamic arrows and highlighted annotations.

7. The intelligent fire protection inspection method based on AI and AR fusion according to claim 6, characterized in that, The process of collecting data information from the entire fire safety inspection process, generating standardized electronic inspection files, and establishing a full-process data traceability link includes: Integrate information from the entire process, including data collection, model building, compliance analysis, verification records, and rectification results, and generate electronic verification files in a standardized format; Add timestamps, operator and equipment number identifiers to the data at each stage, and build a chain traceability link. The chain traceability link supports querying and tracing by file number and time range.

8. A smart fire safety inspection system based on the fusion of AI and AR, characterized in that, The system includes: The data standardization module is used to collect multi-source associated data related to building fire protection, and to clean and standardize the multi-source associated data to form a unified format fire protection inspection basic dataset. The modeling and fusion module is used to overlay and fuse the basic dataset of fire protection inspection with the information of the actual building scene, and to build an AR real-scene 3D model containing the relationship between fire protection facilities and building structure through augmented reality technology; The indicator analysis module is used to automatically compare and analyze multiple compliance indicators in the AR real-scene 3D model based on a preset fire protection code knowledge base and using artificial intelligence algorithms to identify and mark potential compliance issues. The verification module is used to retrieve potential compliance issues through augmented reality devices and perform cross-verification operations that blend virtual and real elements in the actual building scene to confirm the authenticity and specific spatial location information of the potential compliance issues. The visualization module is used to match the corresponding fire safety regulations based on compliance issues, generate targeted rectification suggestions, and use augmented reality technology to visualize the implementation path of the rectification suggestions. The management module is used to collect data information from the entire fire safety inspection process, generate standardized electronic inspection files, and establish a full-process data traceability link to enable the query and traceability of fire safety inspection results.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images