Intelligent analysis method and system for building defects

Abstract

The present invention relates to the field of building detection technology, and specifically to a method and system for intelligent analysis of building defects. The method for intelligent analysis of building defects comprises: collecting photo data and video data of buildings through a data acquisition module, and integrating the collected photo data and video data into a database; preprocessing the data in the database; uploading the preprocessed data to a cloud server; based on the data stored in the cloud server, an artificial intelligence model performs analysis and modeling; based on the defect analysis results, the artificial intelligence model generates an engineering supervision consulting report; after checking the engineering supervision consulting report, the user provides feedback to the artificial intelligence model, and the artificial intelligence model automatically adjusts the artificial intelligence model algorithm according to the user's feedback and the real-time data collected by the data acquisition module.

Description

Technical Field

[0001] The present invention relates to the technical field of building detection, and in particular to a method and system for intelligent analysis of building defects. Background Art

[0002] In high-rise buildings, peeling of plaster or tiles can pose a threat to pedestrians, and existing detection methods may not be able to detect these potential dangers quickly and effectively.

[0003] High-altitude building survey: High-precision inspection of high-rise buildings requires the construction of external wall scaffolding or high platforms. The inspection personnel working at high altitudes poses a challenge to the feasibility and safety of inspection. Accuracy issues: Traditional inspection methods, such as visual inspection and manual measurement, are susceptible to human error, resulting in low accuracy. Insufficient viewing angle: The inspection of high-rise buildings has a narrow viewing angle, limited accuracy, and is prone to omissions. Light influence: Visual inspection is easily affected by environmental factors such as light and weather, which may lead to missed inspections or false inspections. Environmental protection time and cost: Building external wall scaffolding or high platforms for close-range inspection and measurement requires a lot of manpower and time, and the cost is high. Safety risks: There are certain safety risks in the inspection of the external walls of high-rise buildings, especially when using hanging baskets or scaffolding. Experience limitations: The experience and skills of inspectors vary, resulting in inconsistent and fluctuating inspection results. Human errors: Human negligence or fatigue may affect the comprehensiveness and accuracy of the inspection. Data processing: The processing and analysis of large amounts of inspection data requires time and resources, and requires high data accuracy. Report storage: There are challenges in storing and managing old inspection reports, especially for multi-building properties or multiple projects, where reports are easily confused or improperly stored. Regular inspection data: Regular inspection and maintenance require reference to previous data, but sometimes reports may be lost and cannot be compared continuously. Lack of standardization of reports: Different people, regions or countries have different inspection standards, resulting in poor comparability of inspection results. Report format: The lack of a unified data reporting format affects information sharing and communication. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method and system for intelligent analysis of building defects, aiming to solve the problem that the existing detection methods may not be able to quickly and effectively discover these potential dangers.

[0005] To achieve the above object, the present invention adopts the following technical solution:

[0006] A building defect intelligent analysis method, comprising:

[0007] Collect photo data and video data of the building through the data acquisition module, and integrate the collected photo data and video data into the database;

[0008] Preprocess the data in the database;

[0009] Upload the preprocessed data to the cloud server;

[0010] Based on the data stored in the cloud server, the artificial intelligence model performs analysis and modeling;

[0011] Based on the defect analysis results, the AI ​​model generates an engineering supervision consulting report;

[0012] After checking the engineering supervision consulting report, the user provides feedback to the artificial intelligence model, and the artificial intelligence model automatically adjusts the artificial intelligence model algorithm based on the user's feedback and the real-time data collected by the data acquisition module.

[0013] Furthermore, the method of preprocessing the data in the database includes: removing invalid or abnormal data, filling missing data, and standardizing the data.

[0014] Furthermore, the artificial intelligence model uses computer vision technology to label the building defects in the data, and uses the Neural Architecture Search algorithm through a deep learning model based on a residual network and a two-stage detection network to identify the type of building defects, and classify the data according to the type of building defects. The classification standards are designed through color coding or severity grading, and machine learning and large language models are used to make maintenance recommendations for different building defect types and locations, and provide relevant technical documents or reference materials. The algorithm is used to calculate the estimated maintenance costs based on the building defect type, location and required maintenance materials.

[0015] Furthermore, when training the artificial intelligence model, a large number of labeled samples provided by professional measurement consulting companies are used.

[0016] Furthermore, the AI ​​model uses image processing technology to extract the size information of building defects from the inspection images; at the same time, it automatically identifies and blurs facial features to protect privacy.

[0017] Furthermore, the AI ​​model applies a sorting algorithm to generate a priority list based on the type and size of building defects, and continuously optimizes the sorting logic based on user feedback.

[0018] Furthermore, drone scanning or structured light technology is used to collect three-dimensional data of buildings; the collected data is converted into an interactive 3D model and integrated with a geographic information system to mark the specific location of each detected building defect, while providing an interface that allows users to directly upload and embed 2D drawings into the 3D model.

[0019] Furthermore, the corresponding engineering supervision consulting report is automatically generated according to the language preference set by the user; or the engineering supervision consulting report is automatically translated into multiple languages ​​using natural language processing technology.

[0020] Furthermore, the user reviews the engineering supervision consulting report through the cloud server, and after confirming the accuracy of the engineering supervision consulting report, the user can directly add an electronic signature to the engineering supervision consulting report to authenticate the engineering supervision consulting report.

[0021] A building defect intelligent analysis system, based on the above-mentioned building defect intelligent analysis method, comprises:

[0022] A data acquisition module, used to collect photo data and video data of buildings;

[0023] A preprocessing module, used for preprocessing the collected data;

[0024] Cloud server, used to store pre-processed data;

[0025] Analysis and modeling module, used to analyze data stored in cloud servers;

[0026] Report generation module, used to generate engineering supervision consulting reports based on analysis results;

[0027] The interactive module is used for users to check the engineering supervision consulting report and feedback their own needs to the analysis and modeling module.

[0028] The invention discloses a method and system for intelligent analysis of building defects, which has the following beneficial effects:

[0029] The building defect intelligent system not only provides data and reports, but also has a guiding function. It can use artificial intelligence to provide engineering and supervision personnel with maintenance decisions, relevant maintenance cost estimates, and recommended action guidelines based on project requirements and actual conditions. The building defect intelligent system can continuously optimize and improve its performance through artificial intelligence machine learning. Through continuous learning and practice, the building defect intelligent system can better adapt to various complex and changeable building / infrastructure and other maintenance and engineering supervision environments. The building defect intelligent system uses cloud computing technology, allowing users to access the platform through the Internet at any time and in different places, improving the flexibility and convenience of work. Through automation and artificial intelligence processing, the platform can greatly improve work efficiency while reducing labor costs and time costs. Due to the use of artificial intelligence, manual data processing and human interpretation are reduced, so errors caused by human factors will also be reduced accordingly. The building defect intelligent system can provide real-time artificial intelligence analysis results and high-risk defect alarm functions, allowing users to promptly discover and deal with high-risk and high-threat problems, improve the safety and quality of buildings, and ensure the safety of building structures and the public. The building defect intelligent system has the function of data visualization, which presents various defects and complex data in the form of charts or graphics, allowing users to understand the data more intuitively and make more effective maintenance and repair decisions. The building defect intelligent system also provides communication and collaboration tools to facilitate communication and collaboration between users, supervisors, engineers and decision makers. Through standardized reports and guidance automatically generated by professional artificial intelligence, the platform can help users improve the standardization and standardization of maintenance and engineering supervision. The building defect intelligent system can be customized and expanded according to the needs of user feedback to meet the specific needs of different building, infrastructure, civil engineering and other engineering supervision projects and companies. Because there is no need to build external wall grids or platforms, and by reducing the use of paper reports and documents, the platform helps achieve environmental protection, carbon reduction and energy conservation goals. The building defect intelligent system provides a real-time feedback mechanism, allowing users and supervisors to adjust strategies and plans in time according to feedback, discover highly dangerous defects and locations, so as to make rapid crisis relief responses, reduce dangers in real time, and reduce casualties. Improve the efficiency of external wall inspection through professional artificial intelligence, optimize preventive maintenance resources, improve the accuracy of condition surveys, and ensure building and public safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 1 is a flow chart of a method for intelligent analysis of building defects according to an embodiment of the present invention;

[0031] Figure 2 It is a structural block diagram of the building defect intelligent analysis system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0033] like Figure 1 As shown, the present invention provides a building defect intelligent analysis method, comprising:

[0034] Collect photo data and video data of the building through the data acquisition module, and integrate the collected photo data and video data into the database;

[0035] Preprocess the data in the database;

[0036] Upload the preprocessed data to the cloud server;

[0037] Based on the data stored in the cloud server, the artificial intelligence model performs analysis and modeling;

[0038] Based on the defect analysis results, the AI ​​model generates an engineering supervision consulting report;

[0039] After checking the engineering supervision consulting report, the user provides feedback to the artificial intelligence model, and the artificial intelligence model automatically adjusts the artificial intelligence model algorithm based on the user's feedback and the real-time data collected by the data acquisition module.

[0040] The building defect intelligent system not only provides data and reports, but also has a guiding function. It can use artificial intelligence to provide engineering and supervision personnel with maintenance decisions, relevant maintenance cost estimates, and recommended action guidelines based on project requirements and actual conditions. The building defect intelligent system can continuously optimize and improve its performance through artificial intelligence machine learning. Through continuous learning and practice, the building defect intelligent system can better adapt to various complex and changeable building / infrastructure and other maintenance and engineering supervision environments. The building defect intelligent system uses cloud computing technology, allowing users to access the platform through the Internet at any time and in different places, improving the flexibility and convenience of work. Through automation and artificial intelligence processing, the platform can greatly improve work efficiency while reducing labor costs and time costs. Due to the use of artificial intelligence, manual data processing and human interpretation are reduced, so errors caused by human factors will also be reduced accordingly. The building defect intelligent system can provide real-time artificial intelligence analysis results and high-risk defect alarm functions, allowing users to promptly discover and deal with high-risk and high-threat problems, improve the safety and quality of buildings, and ensure the safety of building structures and the public. The building defect intelligent system has the function of data visualization, which presents various defects and complex data in the form of charts or graphics, allowing users to understand the data more intuitively and make more effective maintenance and repair decisions. The building defect intelligent system also provides communication and collaboration tools to facilitate communication and collaboration between users, supervisors, engineers and decision makers. Through standardized reports and guidance automatically generated by professional artificial intelligence, the platform can help users improve the standardization and standardization of maintenance and engineering supervision. The building defect intelligent system can be customized and expanded according to the needs of user feedback to meet the specific needs of different building, infrastructure, civil engineering and other engineering supervision projects and companies. Because there is no need to build external wall grids or platforms, and by reducing the use of paper reports and documents, the platform helps achieve environmental protection, carbon reduction and energy conservation goals. The building defect intelligent system provides a real-time feedback mechanism, allowing users and supervisors to adjust strategies and plans in time according to feedback, discover highly dangerous defects and locations, so as to make rapid crisis relief responses, reduce dangers in real time, and reduce casualties. Improve the efficiency of external wall inspection through professional artificial intelligence, optimize preventive maintenance resources, improve the accuracy of condition surveys, and ensure building and public safety.

[0041] During the entire process, the building defect intelligent analysis system needs to ensure data security and privacy protection. It may adopt encryption, access control and other technologies to protect user data security through encrypted storage and cloud security protocols.

[0042] The intelligent building defect analysis system combines large language models (LLM) (such as SenseChat) with retrieval augmented generation (RAG) technology to enhance the platform's real-time response capabilities, support seamless detection and data management across devices and regions, and allow users to query and analyze according to their needs.

[0043] Furthermore, the method of preprocessing the data in the database includes: removing invalid or abnormal data, filling missing data, and standardizing the data.

[0044] Furthermore, the artificial intelligence model uses computer vision technology to label the building defects in the data, and uses a deep learning model based on a residual network and a two-stage detection network to use a neural architecture search algorithm (NAS) to identify the type of building defects, and classify the data according to the type of building defects. The classification standards are designed through color coding or severity grading, and maintenance suggestions are made for different building defect types and locations through machine learning and large language models. Relevant technical documents or reference materials are provided, and the estimated maintenance costs are calculated using algorithms based on the building defect type, location and required maintenance materials.

[0045] Types of building defects include: cracks, subsoil cracks, web cracks, crazing cracks, peeling, paint, efflorescence, looseness / separation, water stains / dirt, leaks, mold growth, corrosion or rust, loose or missing tiles, loose / bulging wall tiles, patches from previous repairs, broken glass, detached joint sealants, deteriorated mortar joints, plant growth, and deformation / unevenness.

[0046] Machine learning and large language models provide guidance and decision support for engineering supervision based on analysis results, such as predicting potential risks and problems and proposing preventive and remedial measures.

[0047] Furthermore, when training the AI ​​model, a large number of labeled samples provided by professional measurement consulting companies are used to improve the accuracy and efficiency of the analysis.

[0048] Furthermore, the AI ​​model uses image processing technology to extract the size information of building defects from the inspection images; at the same time, it automatically identifies and blurs facial features to protect privacy.

[0049] Furthermore, the AI ​​model applies a sorting algorithm to generate a priority list based on the type and size of building defects, and continuously optimizes the sorting logic based on user feedback. The platform can provide a real-time feedback mechanism, so that users and supervisors can adjust strategies and plans in a timely manner based on feedback, discover highly dangerous defects and locations, so as to make rapid crisis relief responses, reduce dangers in real time, and reduce casualties.

[0050] Furthermore, drone scanning or structured light technology is used to collect three-dimensional data of buildings; the collected data is converted into an interactive 3D model, and the geographic information system is integrated to mark the specific location of each detected building defect, and an interface is provided to allow users to directly upload and embed 2D drawings into the 3D model. The intelligent building defect analysis system can build a 3D building model based on defect data and support 2D / 3D graphic overlays, making it easier for non-technical personnel to understand the defect location, further enhancing practicality.

[0051] Furthermore, the corresponding engineering supervision consulting report is automatically generated according to the language preference set by the user; or the engineering supervision consulting report is automatically translated into multiple languages ​​using natural language processing technology.

[0052] Furthermore, after reviewing the project supervision consulting report through the cloud server and confirming its accuracy, the user can directly add an electronic signature to the project supervision consulting report to certify the project supervision consulting report, which greatly saves the time spent on delivering reports.

[0053] like Figure 2 As shown, the present invention provides a building defect intelligent analysis system, comprising:

[0054] A data acquisition module, used to collect photo data and video data of buildings;

[0055] A preprocessing module, used for preprocessing the collected data;

[0056] Cloud server, used to store pre-processed data;

[0057] Analysis and modeling module, used to analyze data stored in cloud servers;

[0058] Report generation module, used to generate engineering supervision consulting reports based on analysis results;

[0059] The interactive module is used for users to check the engineering supervision consulting report and feedback their own needs to the analysis and modeling module.

[0060] The above description is only a preferred embodiment of the present invention and does not limit the technical scope of the present invention. Therefore, any slight modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for intelligent analysis of building defects, characterized in that: include: Collect photo data and video data of the building through the data acquisition module, and integrate the collected photo data and video data into the database; Preprocess the data in the database; Upload the preprocessed data to the cloud server; Based on the data stored in the cloud server, the artificial intelligence model performs analysis and modeling; Based on the defect analysis results, the AI ​​model generates an engineering supervision consulting report; After checking the engineering supervision consulting report, the user provides feedback to the artificial intelligence model, and the artificial intelligence model automatically adjusts the artificial intelligence model algorithm based on the user's feedback and the real-time data collected by the data acquisition module.

2. The intelligent building defect analysis method according to claim 1, characterized in that: Ways to preprocess data in a database include removing invalid or abnormal data, filling missing data, and standardizing the data.

3. The intelligent building defect analysis method according to claim 1, characterized in that: The artificial intelligence model uses computer vision technology to label building defects in the data, and uses the Neural Architecture Search algorithm through a deep learning model based on a residual network and a two-stage detection network to identify the type of building defects. The data is classified according to the type of building defects, and classification standards are designed through color coding or severity grading. Through machine learning and large language models, repair and maintenance suggestions are made for different building defect types and locations, and relevant technical documents or reference materials are provided. The algorithm is used to calculate the estimated repair costs based on the building defect type, location and required repair materials.

4. The intelligent building defect analysis method according to claim 3 is characterized in that: When training artificial intelligence models, a large number of labeled samples provided by professional measurement consulting companies are used.

5. The intelligent building defect analysis method according to claim 3 is characterized in that: The AI ​​model uses image processing technology to extract the size information of building defects from the inspection images; it also automatically identifies and blurs facial features to protect privacy.

6. The intelligent analysis method for building defects according to claim 5 is characterized in that: The AI ​​model applies a sorting algorithm to generate a priority list based on the type and size of building defects, and continuously optimizes the sorting logic based on user feedback.

7. The intelligent building defect analysis method according to claim 1, characterized in that: Use drone scanning or structured light technology to collect three-dimensional data of buildings; convert the collected data into an interactive 3D model and integrate the geographic information system to mark the specific location of each detected building defect, while providing an interface that allows users to upload 2D drawings directly and embed them into the 3D model.

8. The intelligent building defect analysis method according to claim 1, characterized in that: Automatically generate the corresponding engineering supervision consulting report according to the language preference set by the user; or use natural language processing technology to automatically translate the engineering supervision consulting report into multiple languages.

9. The intelligent building defect analysis method according to claim 1, characterized in that: Users review the engineering supervision consulting report through the cloud server. After confirming the accuracy of the engineering supervision consulting report, they can directly add an electronic signature to the engineering supervision consulting report to certify the engineering supervision consulting report.

10. An intelligent building defect analysis system, characterized in that: The intelligent building defect analysis method according to any one of claims 1 to 9 is characterized by comprising: A data acquisition module, used to collect photo data and video data of buildings; A preprocessing module, used for preprocessing the collected data; Cloud server, used to store pre-processed data; Analysis and modeling module, used to analyze data stored on cloud servers; Report generation module, used to generate engineering supervision consulting reports based on analysis results; The interactive module is used for users to check the engineering supervision consulting report and feedback their own needs to the analysis and modeling module.

Images