Road disease detection method and system based on large model

Abstract

The invention discloses a road disease detection method and system based on a large model, belongs to the technical field of large models and digital twinning, and realizes intelligent analysis of road detection data, intelligent disease identification and automatic generation of a detection report by combining various detection devices and utilizing a large model technology. The digital twinborn technology is utilized to realize the functions of three-dimensional modeling of disease data, analog simulation of disease development and remote control of detection equipment, so that three-dimensional display and positioning of road disease detection data are realized. According to the method, the accuracy and efficiency of disease detection can be greatly improved, the intelligent level of road maintenance is improved, the labor cost and the maintenance cost are also remarkably reduced, the road safety is guaranteed, major and medium repair is delayed, and the service life of the road is prolonged.

Description

Technical Field

[0001] The present invention relates to the field of large model and digital twin technology, and specifically to a road disease detection method and system based on large model and digital twin technology. Background Art

[0002] In recent years, my country has made remarkable achievements in the construction of transportation infrastructure. The criss-crossing transportation network is densely spread across the country, providing inexhaustible power for the country's development. The development of road disease detection is of great significance to improving traffic safety, reducing maintenance costs, and extending the service life of roads. How to timely discover and repair road problems through efficient and accurate detection technology to ensure the safety of drivers, optimize resource allocation, and improve road management efficiency is a problem that needs to be solved at present. Summary of the invention

[0003] The technical task of the present invention is to address the above shortcomings and provide a road disease detection method and system based on a large model, which can greatly improve the accuracy and efficiency of disease detection, not only improve the intelligent level of road maintenance, but also significantly reduce labor costs and maintenance costs, ensure road safety, and extend the service life of roads.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] A road disease detection method based on a large model, combined with various detection equipment, uses large model technology to realize intelligent analysis of road detection data, intelligent identification of diseases, and automatic generation of detection reports; uses digital twin technology to realize three-dimensional modeling of disease data, simulation of disease development, and remote control of detection equipment; thereby realizing three-dimensional display and three-dimensional positioning of road disease detection data, and simulation of disease development trends.

[0006] The application of big models and digital twin technology in road disease detection can greatly improve the accuracy and efficiency of disease detection. Big model technology can accurately identify small anomalies and potential defects in road detection data by processing large amounts of data, while digital twins can accurately locate and comprehensively analyze diseases by creating three-dimensional models of roads and diseases, providing a scientific basis for road maintenance. The application of these technologies not only improves the level of intelligent road maintenance, but also significantly reduces labor costs and maintenance costs, ensures road safety, postpones major and medium repairs, and extends the service life of roads.

[0007] Furthermore, the implementation of this method includes:

[0008] Data acquisition module, used to collect detailed inspection data of road surface and internal structure;

[0009] The disease database is responsible for collecting, storing and managing road disease data obtained from various sensors and detection equipment;

[0010] The data middle platform module is responsible for data cleaning, integration, feature extraction and preliminary analysis, converting raw data into structured information that can be used by machine learning models, and providing a data foundation for subsequent digital twin modeling and large model standard training and promotion;

[0011] The large model standard training and promotion module uses the data processed by the data middle platform module to train, verify and deploy machine learning models;

[0012] Large model application module, used for intelligent analysis of test data, intelligent identification of diseases and automatic generation of test reports;

[0013] Digital twin application modules, including 3D disease modeling and positioning, disease development trend simulation and equipment remote control;

[0014] The business management and control platform module is responsible for integrating various functional modules into a unified platform to ensure efficient collaboration and information flow between modules, while optimizing the overall performance of the system; it is also responsible for docking and data exchange with external systems, providing a friendly user interface, receiving user instructions, displaying analysis results, and collecting user feedback for continuous improvement of the system.

[0015] Furthermore, the data acquisition module supports various types of detection equipment, including visual inspection, laser, 3D ground penetrating radar, infrared camera and other equipment, and can be mounted on various equipment such as inspection vehicles and drones.

[0016] Furthermore, the road damage data includes images, videos, and sensor data, and covers information including road surface detection data and underground structure detection data.

[0017] Furthermore, the large model standard training and inference module,

[0018] Big models refer to deep learning models that require large amounts of data and computing resources to identify, classify, and predict road damage;

[0019] The large model standard training and promotion module relies on the data provided by the data middle platform module to train the model, and its output model is again applied to the processing and analysis of new data to form a closed loop.

[0020] Furthermore, the large model application module,

[0021] Intelligent data analysis: Based on large model training and learning, intelligent noise reduction of detection data is achieved, as well as intelligent elimination of interference from underground steel bars, pipelines and other objects;

[0022] Intelligent disease identification: Based on massive disease data learning, intelligent disease identification of detection data can be achieved, disease automatic labeling can be achieved, and the accuracy of disease identification can be improved;

[0023] Automatically generate inspection reports: Utilize the trained road disease detection model to automatically analyze the processed data provided by the data center module, identify the type, severity and specific location of road diseases, and generate a detailed inspection report. The inspection report includes disease summary analysis, disease cause analysis, and disease treatment recommendations;

[0024] The digital twin application module,

[0025] 3D modeling of defects: By collecting detailed information about the road, including road surface conditions and structural composition, the digital twin module can create a high-precision 3D model of the road and its defects, accurately locate the location of the defects, and provide a basis for subsequent defect treatment;

[0026] Simulation and remote control: Based on digital twin technology, the development of road diseases under different conditions is simulated, the development trend of diseases is predicted, and the operating status of detection equipment and detection routes are adjusted in real time through remote control systems.

[0027] The present invention also claims a road disease detection system based on a large model, including a data acquisition layer, a data calculation layer and a data application layer.

[0028] The data collection layer supports various detection devices to collect detailed data on the road surface and internal structure;

[0029] The data computing layer includes the disease data storage module, the data middle platform module and the large model standard training and promotion module;

[0030] The data application layer includes large model function modules, digital twin modules, and management and control modules;

[0031] The system realizes road damage detection through the above-mentioned large model-based road damage detection method.

[0032] Furthermore, the system includes a large model standard training and promotion platform, which conducts learning and training of disease data based on road disease data and outputs relevant API interfaces.

[0033] The present invention also claims a large model-based road disease detection device, comprising: at least one memory and at least one processor;

[0034] The at least one memory is used to store a machine-readable program;

[0035] The at least one processor is used to call the machine-readable program to implement the above method.

[0036] The present invention also claims protection for a computer-readable medium having computer instructions stored thereon, which, when executed by a processor, causes the processor to perform the above method.

[0037] Compared with the prior art, the road disease detection method and system based on a large model of the present invention have the following beneficial effects:

[0038] The present invention can effectively improve detection efficiency and accuracy, automatically process a large amount of road data, quickly identify potential defects, and significantly improve detection efficiency. By applying large models and digital twin technology, while ensuring high efficiency, it can also maintain a high degree of detection accuracy and reduce missed detections and misjudgments. It can monitor road conditions around the clock, and combined with real-time data analysis, it can predict the development trend of defects, issue early warnings in a timely manner, and help relevant departments respond quickly. It has beneficial effects in optimizing maintenance decisions and resource allocation, extending the service life of roads, and improving road safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is an architecture diagram of a road damage detection system based on a large model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0041] The embodiment of the present invention provides a road disease detection method based on a large model. In combination with various detection equipment, the large model technology is used to realize intelligent analysis of road detection data, intelligent identification of diseases, and automatic generation of detection reports; digital twin technology is used to realize functions such as three-dimensional modeling of disease data, simulation of disease development, and remote control of detection equipment; thereby realizing three-dimensional display and three-dimensional positioning of road disease detection data.

[0042] Big model technology uses massive data to train complex machine learning models, which can capture very subtle patterns and features in the data. It can be applied to a variety of fields such as speech recognition, image recognition, and natural language processing, and achieve high-precision prediction and classification through deep learning algorithms.

[0043] Digital twin technology creates a complete digital model of a physical entity in a virtual space, and achieves simulation and management of the entire life cycle of a real object. It can receive the status data of a physical entity in real time, update it synchronously, and optimize the performance and operation strategy of the physical entity through data analysis and simulation. It is widely used in smart manufacturing, smart city construction, traffic management and other fields, and significantly improves the intelligence level of system management and decision-making efficiency through highly simulated digital models.

[0044] Road disease detection technology uses laser scanners, optical cameras, infrared and microwave sensors and other equipment to collect detailed data on the road surface and internal structure. By preprocessing and extracting features from the collected data, an accurate three-dimensional model is established to identify the disease conditions of the entire road surface and interior. Based on the identification results, a detailed inspection report and disposal suggestions are automatically generated to improve the efficiency and accuracy of road maintenance work.

[0045] This method realizes the three-dimensional display of road disease detection data, and automatically generates detection reports and disposal plans based on disease data. It includes data acquisition layer, data calculation layer and data application layer. It supports the combination of various road detection hardware equipment and large model training system to realize disease data training and learning. It has significant application value in improving road maintenance efficiency, extending road service life, and improving road safety. It also has far-reaching significance in optimizing maintenance decisions and promoting technological innovation.

[0046] The implementation of this method includes:

[0047] The data acquisition module is used to collect detailed data on the road surface and internal structure; it supports various types of detection equipment, including visual detection, laser, 3D ground penetrating radar, infrared camera and other equipment, and can be mounted on various equipment such as inspection vehicles and drones. Collect detailed data on the road surface and internal structure.

[0048] The disease database is responsible for collecting, storing and managing road disease detection data obtained from various sensors and detection equipment. These data include images, videos, sensor readings, etc., which may cover information such as road conditions, cracks, potholes, etc. The disease database provides raw data for the data middle platform module and is the starting point of the entire data processing process. It ensures the security, reliability and accessibility of the data and supports subsequent data analysis and model training.

[0049] The data center module is responsible for data cleaning, integration, feature extraction and preliminary analysis, converting raw data into structured information that can be used by machine learning models, and providing a data foundation for subsequent digital twin modeling and large model standard training and promotion. The data center module processes the data in the disease database and converts it into a consistent format and structure so that the large model standard training and promotion module can effectively use it. It ensures data quality, improves the efficiency and accuracy of data analysis, and serves as a bridge between data storage and model training.

[0050] The large model training and promotion module uses the data processed by the data middle platform module to train, verify and deploy machine learning models; the "large model" here refers to deep learning, large language, and multimodal models that may require a large amount of data and computing resources, and are used to identify, classify and predict road defects and generate inspection reports.

[0051] The large model standard training module relies on the data provided by the data middle platform module to train the model, and its output model is applied again to the processing and analysis of new data to form a closed loop. This module is the core of the system's intelligent decision-making and directly affects the accuracy and efficiency of disease detection.

[0052] The large model application module is used for intelligent analysis of detection data, intelligent disease identification and automatic generation of detection reports. This module is based on the data output of the data center module and the large model standard training and promotion module, and converts abstract data into specific business results. It is a key step in realizing the application value of the system. It includes:

[0053] Intelligent data analysis: Based on large model training and learning, intelligent noise reduction of detection data is achieved, as well as intelligent elimination of interference from underground steel bars, pipelines and other objects;

[0054] Intelligent disease identification: Based on massive disease data learning, intelligent disease identification of detection data can be achieved, disease automatic labeling can be achieved, and the accuracy of disease identification can be improved;

[0055] Automatically generate inspection reports: Using the trained road disease detection model, this module can automatically analyze the processed data provided by the data center module, identify the type, severity and specific location of road diseases, and generate a detailed inspection report; the inspection report includes disease summary analysis, disease cause analysis, and disease treatment recommendations.

[0056] Digital twin application module: The digital twin module complements the large model function module. It not only provides more intuitive and detailed data support for the large model, but also can dynamically display and further analyze the prediction results of the large model in three-dimensional form. Specifically include:

[0057] 3D modeling of defects: By collecting detailed information about the road, including road surface conditions and structural composition, the digital twin module can create a high-precision 3D model of the road and its defects, accurately locate the location of the defects, and provide a basis for subsequent defect treatment;

[0058] Simulation and remote control: Based on digital twin technology, the development of road diseases under different conditions can be simulated, the development trend of diseases can be predicted, and the operating status of detection equipment and detection routes can be adjusted in real time through remote control systems.

[0059] The system control platform module coordinates the work of each module and is responsible for docking and data exchange with external systems to ensure that the system can adapt to the ever-changing needs and technical environment. Including:

[0060] System integration and optimization: The management and control module is responsible for integrating the large model functional modules and digital twin modules into a unified platform to ensure efficient collaboration and information flow among modules, while optimizing the overall performance of the system.

[0061] User interaction and feedback: As the interface for users to interact with the system, the management and control module provides a friendly user interface, receives user instructions, displays analysis results, and collects user feedback for continuous improvement of the system.

[0062] An embodiment of the present invention further provides a large model-based road damage detection system, which implements road damage detection through the large model-based road damage detection method described in the above embodiment.

[0063] like Figure 1 As shown, the system includes:

[0064] 1. Data collection layer.

[0065] The data collection layer supports various types of inspection equipment, including visual inspection, laser, 3D ground penetrating radar, infrared camera and other equipment, which can be mounted on inspection vehicles, drones and other equipment to collect detailed data on the road surface and internal structure.

[0066] 2. Data computing layer.

[0067] The data computing layer includes a disease data storage module, a data middle platform module, and a large model standard training and promotion module. The disease data storage module is responsible for collecting and managing raw data, the data middle platform module processes these data, and finally the large model standard training and promotion module uses the processed data to train and deploy models to achieve intelligent detection of road diseases.

[0068] Disease data storage module:

[0069] This module is responsible for collecting, storing and managing road damage data obtained from various sensors and detection equipment. These data include images, videos, sensor data, etc., and may include information such as road surface detection data and underground structure detection data.

[0070] As the data foundation of the system, the disease data storage module provides raw data for the data middle platform module and is the starting point of the entire data processing flow. It ensures the security, reliability and accessibility of the data and supports subsequent data analysis and model training.

[0071] Data middle platform module:

[0072] The data center module is located between the disease data storage module and the large model training module, and serves as the hub for data processing and management. It is responsible for data cleaning, integration, feature extraction and preliminary analysis, converting raw data into structured information that can be used by machine learning models, and providing a data foundation for subsequent digital twin modeling.

[0073] The data center module processes the data from the disease data storage module and converts it into a consistent format and structure so that the large model standard training module can effectively use it. It ensures data quality, improves the efficiency and accuracy of data analysis, and serves as a bridge between data storage and model training.

[0074] Large model standard training and promotion module:

[0075] The large model training and promotion module uses the processed data from the data center module to train, verify and deploy machine learning models. The "large model" here refers to a deep learning model that may require a large amount of data and computing resources to identify, classify and predict road damage.

[0076] The large model standard training module relies on the data provided by the data middle platform module to train the model, and its output model is applied again to the processing and analysis of new data to form a closed loop. This module is the core of the system's intelligent decision-making and directly affects the accuracy and efficiency of disease detection.

[0077] The system has a large model standard training and promotion platform, which can conduct learning and training based on road disease data and output relevant API interfaces.

[0078] 3. Data application layer.

[0079] The data application layer outputs road inspection reports, disposal plans, inspection plans, as well as three-dimensional disease modeling, disease development simulation, and equipment remote control based on road inspection data, including large models, digital twin-related application functions, and road inspection system management and control modules.

[0080] Large model function module:

[0081] Intelligent data analysis: Based on large model training and learning, intelligent noise reduction of detection data is achieved, as well as intelligent elimination of interference from underground steel bars, pipelines and other objects;

[0082] Intelligent disease identification: Based on massive disease data learning, intelligent disease identification of detection data can be achieved, disease automatic labeling can be achieved, and the accuracy of disease identification can be improved;

[0083] Automatically generate inspection reports: Utilize the trained road disease detection model to automatically analyze the processed data provided by the data center module, identify the type, severity and specific location of road diseases, and generate a detailed inspection report. The inspection report includes disease summary analysis, disease cause analysis, and disease treatment recommendations;

[0084] This module is based on the data and API output of the data middle platform module and the large model standard training and promotion module, converting abstract data into specific business application results, which is a key step in realizing the application value of the system.

[0085] Digital Twin Module:

[0086] 3D modeling of defects: By collecting detailed information about the road, including road conditions, structural composition, etc., the digital twin module can create a high-precision 3D model of the road and its defects, accurately locate the location of the defect, and provide a basis for subsequent defect treatment.

[0087] Simulation and remote control: Based on digital twin technology, the development of road diseases under different conditions can be simulated, the development trend of diseases can be predicted, and the operating status and detection route of detection equipment can be adjusted in real time through the remote control system.

[0088] The digital twin module complements the large model functional module. It not only provides more intuitive and detailed data support for the large model, but also can dynamically display and further analyze the prediction results of the large model in three-dimensional form.

[0089] Management and control module:

[0090] System integration and optimization: The management and control module is responsible for integrating the large model functional modules and digital twin modules into a unified platform to ensure efficient collaboration and information flow among modules, while optimizing the overall performance of the system.

[0091] User interaction and feedback: As the interface for users to interact with the system, the management and control module provides a friendly user interface, receives user instructions, displays analysis results, and collects user feedback for continuous improvement of the system.

[0092] The control module is the "brain" of the system. It not only coordinates the work of various modules, but is also responsible for docking and data exchange with external systems to ensure that the system can adapt to changing needs and technical environments.

[0093] The embodiment of the present invention also provides a road disease detection device based on a large model, comprising: at least one memory and at least one processor;

[0094] The at least one memory is used to store a machine-readable program;

[0095] The at least one processor is used to call the machine-readable program to implement the road damage detection method based on a large model described in the above embodiment.

[0096] The embodiment of the present invention further provides a computer-readable medium, on which computer instructions are stored, and when the computer instructions are executed by a processor, the processor executes the road disease detection method based on a large model described in the above embodiment. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code that implements the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0097] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0098] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer by a communication network.

[0099] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0100] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0101] The present invention is shown and described in detail above through the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art can know that the code review methods in the above different embodiments can be combined to obtain more embodiments of the present invention, and these embodiments are also within the protection scope of the present invention.

Claims

1. A road disease detection method based on a large model, characterized in that: Combining various types of detection equipment, large model technology is used to realize intelligent analysis of road detection data, intelligent identification of defects, and automatic generation of detection reports. Digital twin technology is used to realize three-dimensional modeling of defect data, simulation of defect development, and remote control of detection equipment, thereby realizing three-dimensional display and three-dimensional positioning of road defect detection data.

2. A road damage detection method based on a large model according to claim 1, characterized in that: The implementation of this method includes: Data collection module, used to collect detailed data on the road surface and internal structure; The disease database is responsible for collecting, storing and managing road disease detection data obtained from various sensors and detection equipment; The data middle platform module is responsible for data cleaning, integration, feature extraction and preliminary analysis, converting raw data into structured information that can be used by machine learning models, and providing a data foundation for subsequent digital twin modeling and large model standard training and promotion; The large model standard training and promotion module uses the data processed by the data middle platform module to train, verify and deploy machine learning models; Large model application module, used for intelligent analysis of test data, intelligent identification of diseases and automatic generation of test reports; Digital twin application modules, including 3D modeling display and positioning of diseases, simulation of disease development trends and remote control of equipment; The business management and control platform module is responsible for integrating various functional modules into a unified platform to ensure efficient collaboration and information flow between modules, while optimizing the overall performance of the system; it is also responsible for docking and data exchange with external systems, providing a friendly user interface, receiving user instructions, displaying analysis results, and collecting user feedback for continuous improvement of the system.

3. A road disease detection method based on a large model according to claim 2, characterized in that: The data acquisition module supports various types of detection equipment, including visual inspection, laser, 3D ground penetrating radar, and infrared camera equipment, and can be mounted on various types of equipment such as inspection vehicles and drones.

4. A road disease detection method based on a large model according to claim 2, characterized in that: Road disease data include images, videos, and sensor data, and cover information including road surface detection data and underground structure detection data.

5. A road disease detection method based on a large model according to claim 2 or 4, characterized in that: The large model standard training and inference module, Big models refer to deep learning models that require large amounts of data and computing resources to identify, classify, and predict road damage; The large model standard training and promotion module relies on the data provided by the data middle platform module to train the model, and its output model is again applied to the processing and analysis of new data to form a closed loop.

6. A road disease detection method based on a large model according to claim 2, characterized in that: The large model application module, Intelligent data analysis: Based on large model training and learning, intelligent noise reduction of detection data and intelligent elimination of interference from underground steel bars and pipelines can be achieved; Intelligent disease identification: Based on massive disease data learning, intelligent disease identification of detection data can be achieved, disease automatic labeling can be achieved, and the accuracy of disease identification can be improved; Automatically generate inspection reports: Utilize the trained road disease detection model to automatically analyze the processed data provided by the data center module, identify the type, severity and specific location of road diseases, and generate a detailed inspection report. The inspection report includes disease summary analysis, disease cause analysis, and disease treatment recommendations; The digital twin application module, 3D modeling of defects: By collecting detailed information about the road, including road surface conditions and structural composition, a high-precision 3D model of the road and its defects is created to accurately locate the location of the defect, providing a basis for subsequent defect treatment; Simulation and remote control: Based on digital twin technology, the development of road diseases under different conditions is simulated, the development trend of diseases is predicted, and the operating status of detection equipment and detection routes are adjusted in real time through remote control systems.

7. A road disease detection system based on a large model, characterized in that: Including data collection layer, data calculation layer and data application layer, The data collection layer supports various detection devices to collect detailed data on the road surface and internal structure; The data computing layer includes the disease data storage module, the data middle platform module and the large model standard training and promotion module; The data application layer includes large model function modules, digital twin modules, and management and control modules; The system realizes road damage detection through the large model-based road damage detection method described in any one of claims 1 to 6.

8. The road disease detection system based on large model according to claim 1 is characterized in that: The system includes a large model standard training and promotion platform, which conducts learning and training of disease data based on road disease data and outputs relevant API interfaces.

9. A road disease detection device based on a large model, characterized in that: include: at least one memory and at least one processor; The at least one memory is used to store a machine-readable program; The at least one processor is used to call the machine-readable program to implement the method described in any one of claims 1 to 6.

10. A computer-readable medium, characterized in that The computer readable medium stores computer instructions, and when the computer instructions are executed by a processor, the processor executes the method according to any one of claims 1 to 6.

Images