Road disease detection system and method based on multi-source data fusion

Abstract

The invention discloses a road disease detection system and method based on multi-source data fusion. The system comprises a data acquisition module, a data preprocessing module, a disease detection module, a three-dimensional modeling module, a data analysis module and a maintenance module. The data acquisition module is used for acquiring road surface data and sending the acquired data to the data preprocessing module; the road surface data comprises road data, road disease data, road vehicle data, non-motor vehicle data and pedestrian data, compared with the prior art, more comprehensive and more accurate detection of road diseases is realized by fusing multi-source data; by means of multi-dimensional data acquisition of radar data, video data, three-dimensional point cloud data, geographical location information and the like, the limitation of a single data source is overcome; the data preprocessing module can effectively remove noise and abnormal values, accurately classify and mark diseases, and fuse data collected by different sensors, so that the quality and integrity of the data are improved.

Description

Technical Field

[0001] The present invention relates to the field of road detection, and in particular to a road damage detection system and method based on multi-source data fusion. Background Art

[0002] In today's society, with the rapid acceleration of urbanization, the construction scale of urban roads continues to expand, and its maintenance work has become increasingly critical and arduous. As an important infrastructure of urban transportation, the good condition of urban roads is of vital importance to ensuring traffic safety, improving traffic efficiency, promoting urban economic development and improving the quality of life of residents. The timely and accurate detection and repair of road diseases play an indispensable role in this process. Because it can not only effectively prevent the occurrence of traffic accidents and ensure the travel safety of pedestrians and vehicles, but also significantly improve the service life of roads and reduce the high cost of road maintenance.

[0003] Traditional road disease detection methods mainly rely on manual inspections. Workers need to conduct on-site inspections along the road, relying on their naked eyes and experience to determine whether there are road diseases and the type and extent of the diseases. However, this method has many obvious disadvantages. First, the efficiency of manual inspections is extremely low. Faced with a long road network, detection work often takes a lot of time and manpower. Secondly, since the test results depend largely on the subjective judgment of the staff, they are easily affected by factors such as personal experience and fatigue, making it difficult to guarantee the accuracy and consistency of the test results. Furthermore, the detection range of manual inspections is very limited, and it is difficult to cover every corner of the road. It is easy to miss some potential diseases, which poses a hidden danger to road safety.

[0004] In recent years, with the rapid development of computer technology and the continuous innovation of sensor technology, road damage detection methods based on machine vision and deep learning have gradually emerged and become a hot research area. These emerging technical means have brought new ideas and methods to road damage detection with their significant advantages such as high efficiency, accuracy and objectivity. They can automatically process and analyze a large amount of images and data, quickly and accurately identify various types of road damage, greatly improve the efficiency and accuracy of detection, and provide strong support for the timely maintenance and repair of roads. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the manual chilling speed is slow and cannot realize automated production. When the chilling iron is manually placed, the chilling iron is not firmly adsorbed, and the chilling iron will be displaced and fall off during the movement of the tooling, causing production pauses and defective products. To address this problem, a road disease detection system and method based on multi-source data fusion are proposed.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A road disease detection system and method based on multi-source data fusion, including a data acquisition module, a data preprocessing module, a disease detection module, a three-dimensional modeling module, a data analysis module and a maintenance module; The data acquisition module is used to collect road surface data and send the collected data to the data preprocessing module for processing; the road surface data includes road data, road disease data, road vehicle data, non-motor vehicle data and pedestrian data; The data preprocessing module preprocesses the road data through data cleaning, data integration, data reduction and data transformation, classifies the road data, and forms estimated data for estimating diseases and current data for diagnosing diseases; and sends both the estimated data and current data to the disease detection module and the three-dimensional modeling module; the preprocessing includes data cleaning, data labeling and data fusion.

[0007] The defect detection module gives the corresponding defect level, defect type and defect repair plan based on the current defect data, and sends the defect repair plan to the maintenance module; defect types include pavement crack detection, pavement pothole detection and pavement settlement detection The 3D modeling module is used to construct a road damage model. Through the constructed surface model and underground model, the location of road damage can be determined and the efficiency of road damage simulation and analysis can be improved. The three-dimensional modeling module adopts three-dimensional point cloud processing technology to perform three-dimensional reconstruction on the road after disease detection and generate a high-precision three-dimensional model.

[0008] The road damage model includes a surface model and an underground model. The surface model is used to construct the three-dimensional geometric shape and texture information of the road surface, accurately presenting the road's flatness, slope, road surface markings and other features; The underground model is used to describe the distribution and condition of the underground structure of the road, including the location, depth and direction of underground pipelines, as well as the composition of the roadbed and potential underground diseases; The data analysis module, combined with 3D modeling, compares and analyzes the current data of the disease and the estimated data to obtain the predicted results of road diseases, confirm the accuracy of the predicted results, and provide the disease level, disease type, disease range, disease repair plan and repair period, and send them to the maintenance module; The maintenance module is used to notify maintenance personnel to maintain relevant roads in a timely manner.

[0009] As a further preferred embodiment of the present invention, the data acquisition module includes a radar data unit, a video data unit, a three-dimensional point cloud data unit and a geographic location information unit; Wherein, the radar information collection unit is used to collect radar data information of the road; The video data acquisition unit is used to acquire video data of the road; The three-dimensional point cloud data acquisition unit is used to acquire three-dimensional point cloud data of the road; The geographical location information acquisition unit is used to acquire geographical location information data of the road.

[0010] As a further preferred embodiment of the present invention, the data preprocessing module includes a data cleaning unit, a data labeling unit and a data fusion unit; Wherein, the data cleaning unit is used to remove noise and outliers; The data labeling unit is used to classify and label diseases; The data fusion unit is used to fuse the data collected by different sensors to improve the accuracy and completeness of the data.

[0011] As a further preferred embodiment of the present invention, the disease detection module includes a road crack detection unit, a road pothole detection unit and a road settlement detection unit; The pavement crack detection unit is used to perform pavement crack detection on current data of the disease; The road pothole detection unit is used to perform road pothole detection on current data of the disease; The road surface settlement detection unit is used to perform road surface settlement detection on current data of the disease.

[0012] As a further preferred embodiment of the present invention, the three-dimensional modeling module includes a surface model unit and an underground model unit.

[0013] As a further preferred embodiment of the present invention, the data analysis module includes a disease distribution statistics unit, a disease development trend analysis unit and a road maintenance suggestion generation unit; The disease distribution statistics unit is used to perform detailed statistics and analysis on the occurrence frequency and distribution of road diseases detected in different sections and locations; The disease development trend analysis unit is used to predict the development direction and speed of the disease by using mathematical models and algorithms through comprehensive analysis of current data, historical data and prediction results of the disease; The road maintenance suggestion generating unit is used to intelligently generate targeted and feasible road maintenance suggestions based on the distribution statistics of the diseases and the development trend analysis, combined with the road usage and traffic flow factors.

[0014] A road damage detection method based on multi-source data fusion includes the following steps: S1. Collect historical data of different types of roads within 1-2 years, including the surface conditions, types of defects, scope of defects, vehicle flow, pedestrian flow, location, depth and direction of underground pipelines, roadbed composition and potential underground defects of the road; S2. Construct a 3D model of the road based on historical data and 3D point cloud data; S3. Based on historical data, through deep learning, establish an analysis and prediction model to provide support for the disease detection module and data analysis module; S4, collecting current road data from multiple angles through a data collection module, and performing data preprocessing to form current data and estimated data; S5, importing current data and estimated data into the analysis and prediction model and the three-dimensional model; S6. Based on the analysis and prediction model and the three-dimensional model, the disease detection module detects the disease condition of the current road, forms a maintenance report, and sends it to the maintenance module; S7. Based on the analysis and prediction model and the three-dimensional model, the data analysis module analyzes the current data and makes a prediction, obtains the prediction result of road damage, gives the damage grade, damage type, damage range, damage repair plan and repair period, and sends them to the maintenance module; S8. The maintenance module performs maintenance on relevant roads in a timely manner according to the maintenance report; monitors relevant roads according to the prediction results, and performs maintenance on relevant roads when the damage maintenance level or time is reached.

[0015] As a further preferred embodiment of the present invention, the multi-angle acquisition includes radar acquisition, video acquisition, three-dimensional laser scanner acquisition and global positioning system.

[0016] The road damage detection system and method based on multi-source data fusion proposed in the present invention have the following beneficial effects compared with the prior art: 1. The present invention achieves a more comprehensive and accurate detection of road diseases by integrating multi-source data; 2. Multi-dimensional data collection such as radar data, video data, 3D point cloud data and geographic location information is used to overcome the limitations of a single data source; 3. The data preprocessing module can effectively remove noise and outliers, accurately classify and label diseases, and fuse data collected by different sensors to improve the quality and integrity of the data; 4. The disease detection module can accurately detect a variety of common road diseases, such as road cracks, potholes and settlement, providing detailed and accurate disease information for road maintenance; 5. The 3D modeling module can not only build a 3D model of the road surface, showing the road's flatness, slope, markings and other features, but also depict the road's underground structure, revealing potential problems such as underground pipelines and roadbeds, providing an intuitive and in-depth perspective for the overall evaluation and planning of the road; 6. The data analysis module provides a scientific basis for decision-making for road management departments through statistics on disease distribution, analysis of development trends, and generation of road maintenance recommendations. It helps to reasonably arrange maintenance resources, take preventive measures in advance, reduce road maintenance costs, and extend the service life of roads, thereby ensuring traffic safety and improving the operating efficiency and service quality of urban roads. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a structural block diagram of the road damage detection and three-dimensional modeling system based on multi-source data fusion of the present invention; Figure 2 It is a structural block diagram of the data acquisition module of the present invention; Figure 3 It is a structural block diagram of the data preprocessing module of the present invention; Figure 4 This is a structural block diagram of the disease detection module of the present invention; Figure 5 It is a structural block diagram of the three-dimensional modeling module of the present invention; Figure 6 This is a structural block diagram of the data analysis module of the present invention. DETAILED DESCRIPTION

[0018] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] Embodiment 1: Combination Figure 1-6 A road disease detection system and method based on multi-source data fusion includes a data acquisition module, a data preprocessing module, a disease detection module, a three-dimensional modeling module, a data analysis module and a maintenance module.

[0020] The data acquisition module is used to collect road surface data and send the collected data to the data preprocessing module for processing; the road surface data includes road data, road disease data, road vehicle data, non-motor vehicle data and pedestrian data.

[0021] like Figure 2As shown, the data acquisition module includes a radar data unit, a video data unit, a three-dimensional point cloud data unit and a geographic location information unit; wherein the radar information acquisition unit is used to collect radar data information of the road; the radar information acquisition unit is used to accurately collect radar data information of the road by virtue of its advanced technology and sophisticated equipment. By emitting electromagnetic waves and receiving echoes, it is possible to detect the structural features inside the road, such as the thickness of the road layer, potential cavities or loose areas, etc., providing key basic data for a comprehensive understanding of the road conditions.

[0022] The video data acquisition unit is used to collect video data of the road; the video data acquisition unit uses high-resolution camera equipment and intelligent image acquisition system to comprehensively collect video data of the road. It can capture various conditions of the road surface in real time, including road damage, clarity of markings, traffic flow and other intuitive information, and show the real-time status of the road from a dynamic perspective.

[0023] The 3D point cloud data acquisition unit is used to collect 3D point cloud data of the road; the 3D point cloud data acquisition unit uses high-precision laser scanning equipment and advanced point cloud generation technology to accurately collect 3D point cloud data of the road. It can carefully depict the spatial geometry of the road, including complex 3D features such as ups and downs, slopes, and curves of the road, providing rich spatial data for building an accurate road model.

[0024] The geographic location information acquisition unit is used to collect geographic location information data of roads; it uses the global positioning system (GPS) and geographic information system (GIS) technology to accurately collect geographic location information data of roads. It can clarify the specific location, direction, and relationship with the surrounding geographical environment of the road, providing an important geographic reference for the positioning, planning, and management of the road.

[0025] The data acquisition module includes a variety of advanced sensors such as radar, video, 3D laser scanner and global positioning system, which are used to collect multi-source data of the road in all directions, including radar data that can reflect the internal structure of the road, video data that intuitively displays the surface condition of the road, 3D point cloud data that accurately depicts the spatial form of the road, and information that clearly defines the geographical location of the road.

[0026] The data preprocessing module preprocesses the road data through data cleaning, data integration, data reduction and data transformation, classifies the road data, and forms estimated data for estimating diseases and current data for diagnosing diseases; and sends both the estimated data and current data to the disease detection module and the three-dimensional modeling module; the preprocessing includes data cleaning, data labeling and data fusion.

[0027] like Figure 3As shown, the data preprocessing module includes a data cleaning unit, a data labeling unit and a data fusion unit.

[0028] During the data collection process, due to the interference of various external factors, a large amount of noise data may be introduced, which will seriously affect the subsequent analysis and processing results. At the same time, the existence of outliers may also lead to wrong judgments and conclusions. Therefore, the data cleaning unit mainly removes the noise and outliers in the data through specific algorithms and filtering mechanisms to improve the purity and reliability of the data and provide a clean and reliable data basis for subsequent processing.

[0029] The data annotation unit mainly classifies and annotates possible defects accurately and in detail based on professional standards and experience, such as cracks, fissures, looseness and other defect types, and annotates the severity of the monitored defects according to the database and surface and underground models, providing a clear basis for subsequent analysis and processing; in the face of complex and diverse road defect conditions, the data annotation unit can accurately classify the defects according to factors such as type, severity, and location based on strict standards and specifications, and give each defect instance a clear and unambiguous label, providing accurate guidance for subsequent defect detection and analysis.

[0030] The data fusion unit focuses on organically integrating the multi-dimensional data collected from different sensors, giving full play to the advantages of each data source, thereby greatly improving the accuracy and integrity of the data. Since different sensors have their own characteristics and limitations when collecting data, through the processing of the data fusion unit, the advantages of each sensor can be fully utilized to make up for the shortcomings of a single data source. This can greatly improve the accuracy and integrity of the data, and provide more comprehensive, accurate and reliable data support for subsequent disease detection, three-dimensional modeling and data analysis.

[0031] The defect detection module is used to conduct comprehensive and accurate defect detection on the pre-processed data, covering key aspects such as pavement crack detection, pavement pothole detection, and pavement settlement detection. The corresponding defect level, defect type, and defect repair plan are given based on the current defect data, and the defect repair plan is sent to the maintenance module; the defect types include pavement crack detection, pavement pothole detection, and pavement settlement detection.

[0032] like Figure 4 As shown, the disease detection module includes a road crack detection unit, a road pothole detection unit and a road settlement detection unit.

[0033] The pavement crack detection unit is used to perform pavement crack detection on the current data of the disease; it can accurately identify the types of cracks, such as longitudinal cracks, transverse cracks, mesh cracks, etc., and accurately measure key parameters such as the length, width and depth of the cracks, providing an important basis for evaluating the safety and service life of the road.

[0034] The road pothole detection unit uses highly sensitive sensing equipment and intelligent image processing technology to detect potholes on the road based on the current data of the disease; potholes on the road not only affect driving comfort, but may also cause damage to the vehicle. The detection unit can quickly capture the location and shape of the potholes, accurately calculate the area and depth of the potholes, and thus provide accurate information for timely repair and improvement of road conditions.

[0035] The pavement settlement detection unit combines high-precision measuring instruments and professional data analysis models to detect pavement settlement based on the current data of the disease. Pavement settlement is a relatively serious road disease that may cause damage to the road structure and traffic safety hazards. The detection unit can accurately determine the area and degree of settlement, analyze the development trend of settlement, and provide strong support for the formulation of scientific and reasonable repair plans and preventive measures.

[0036] The disease detection module innovatively adopts a deep learning algorithm, which can deeply explore the hidden features in the preprocessed data and classify them efficiently, thereby realizing automatic detection and accurate identification of road diseases, greatly improving the efficiency and accuracy of detection.

[0037] The three-dimensional modeling module is used to carry out comprehensive and detailed three-dimensional modeling of the road after disease detection, determine the location of road diseases and improve the efficiency of road disease simulation analysis by constructing surface models and underground models; the three-dimensional modeling module includes a surface model unit and an underground model unit.

[0038] The 3D modeling module uses cutting-edge 3D point cloud processing technology to perform high-precision 3D reconstruction of roads that have been inspected for defects. It can accurately restore every detail of the road and generate a highly accurate and realistic 3D model.

[0039] like Figure 5 As shown, the road damage model includes a surface model and an underground model. The surface model is used to construct the three-dimensional geometric shape and texture information of the road surface, and accurately present the flatness, slope, road surface markings and other characteristics of the road.

[0040] The surface model unit integrates advanced 3D modeling technology and high-resolution image acquisition equipment to construct fine 3D geometry and realistic texture information of the road surface. It can accurately present key features such as road flatness, slope, and road markings through precise measurement and data processing. The display of flatness can reflect the comfort of road driving, the depiction of slope helps to understand the resistance and energy consumption of vehicle driving, and clear road markings provide an important guarantee for traffic safety.

[0041] The underground model is used to depict the distribution and condition of the underground structure of the road. The underground model unit includes the location, depth and direction of underground pipelines, as well as the composition of the roadbed and potential underground diseases. It combines geological detection technology, underground imaging equipment and professional modeling software to carefully depict the complex distribution and detailed conditions of the underground structure of the road. It can accurately locate the exact location of underground pipelines, accurately measure their depth and clearly depict their direction, and can also analyze the composition of the roadbed and potential underground diseases in detail. A clear understanding of underground pipelines can avoid damage to them during road construction and maintenance; and understanding the composition of the roadbed and underground diseases can provide early warning of possible serious problems such as road collapse, providing strong support for the long-term stability and safety of roads.

[0042] Data analysis module: The data analysis module is used to conduct in-depth and systematic data analysis on the disease detection results and 3D modeling results, including disease distribution statistics, disease development trend analysis and road maintenance suggestion generation. By combining 3D modeling, the current data and estimated data of the disease are compared and analyzed to obtain the predicted results of road diseases, and the accuracy of the predicted results is confirmed. The disease level, disease type, disease range, disease repair plan and repair period are given and sent to the maintenance module. The data analysis module has powerful functions and can conduct comprehensive statistical analysis on the disease detection results, clearly understand the distribution of diseases in different regions and their development trends over time, and thus provide scientific and powerful support for road maintenance decisions.

[0043] like Figure 6 As shown, the data analysis module includes a disease distribution statistics unit, a disease development trend analysis unit and a road maintenance suggestion generation unit.

[0044] The disease distribution statistics unit combines powerful data processing capabilities and precise statistical methods to conduct comprehensive, in-depth and detailed statistics and analysis of the frequency and distribution of detected road diseases in different sections and locations. It can carefully sort out the number, regularity and concentration of disease occurrences in each section and location, and provide clear and accurate basic information for subsequent analysis and decision-making by generating detailed data reports and intuitive charts.

[0045] The disease development trend analysis unit is used to conduct a comprehensive analysis of the current data, historical data and prediction results of the disease, and use mathematical models and algorithms to predict the development direction and speed of the disease; it can not only mine potential laws and trends from massive data, but also proactively evaluate the future evolution of the disease, providing key decision-making basis for taking preventive and repair measures in advance.

[0046] The road maintenance suggestion generation unit combines intelligent algorithms and rich road engineering knowledge to intelligently generate targeted and feasible road maintenance suggestions based on the distribution statistics and development trend analysis of defects, combined with road usage and traffic flow factors. It can formulate reasonable maintenance plans based on different defect types, distribution and development trends, combined with the daily use intensity and traffic flow characteristics of the road, including maintenance time nodes, specific construction methods, required materials and equipment, and corresponding budget arrangements, thereby effectively improving the efficiency and quality of road maintenance and ensuring the safety and smoothness of roads.

[0047] The maintenance module is used to notify maintenance personnel to maintain relevant roads in a timely manner.

[0048] Embodiment 2: A road damage detection method based on multi-source data fusion, comprising the following steps: S1. Collect historical data within 1-2 years for different types of roads. The historical data include the road surface conditions, types of defects, damage range, vehicle flow, pedestrian flow, location, depth and direction of underground pipelines, roadbed composition and potential underground defects.

[0049] S2. Based on historical data and combined with 3D point cloud data, a 3D model of the road is constructed. By meticulously restoring the surface morphology and underground structure of the road, a realistic and accurate 3D model is generated, providing intuitive and comprehensive visualization support for the overall evaluation, planning and design of the road.

[0050] S3. Based on historical data, through deep learning, fully process historical data, rely on a large number of manual annotations of disease conditions, categories and severity, establish an analysis and prediction model to provide support for the disease detection module and data analysis module; S4. The data acquisition module collects current road data from multiple angles and performs data preprocessing to form current data and estimated data; the multiple angles include radar acquisition, video acquisition, 3D laser scanner acquisition and global positioning system. With the help of a series of advanced and accurate sensors and acquisition equipment, various road-related data are collected from all directions and multiple angles. These data come from a wide range of sources, covering information that can reflect the internal structural characteristics of the road, images that intuitively show the surface conditions of the road, point cloud data that accurately depicts the spatial form of the road, and positioning information that clearly defines the geographical location of the road, etc., providing comprehensive and detailed original materials for subsequent processing and analysis.

[0051] The data preprocessing module uses processing technology and algorithms to conduct in-depth processing of the collected raw data. First, through data cleaning technology, the noise and outliers in the data are removed to ensure the purity and reliability of the data; then, using professional data annotation methods, the possible diseases are accurately classified and detailedly labeled to provide clear guidance for subsequent disease detection; finally, through data fusion technology, the data collected from different channels and different types of sensors are organically integrated to give full play to their respective advantages, thereby significantly improving the accuracy and completeness of the data and laying a solid foundation for subsequent analysis and processing.

[0052] S5. Import the current data and estimated data into the analysis and prediction model and the three-dimensional model.

[0053] S6. Based on the analysis and prediction model and the three-dimensional model, the disease detection module detects the disease condition of the current road, forms a maintenance report, and sends it to the maintenance module.

[0054] The road damage category is determined through data preprocessing, historical data analysis and comparison, and the damage situation is restored based on the database and surface and underground models, while the degree of damage caused by road damage is clarified.

[0055] S7. Based on the analysis and prediction model and the three-dimensional model, the data analysis module analyzes the current data and makes predictions, obtains the prediction results of road damage, gives the damage level, damage type, damage range, damage repair plan and repair period, and sends them to the maintenance module.

[0056] Through the intuitive presentation of diseases constructed by data models and three-dimensional models, the disease grade, type, scope and recommended data for disease repair are obtained for reference by decision makers.

[0057] By fully integrating multiple functions such as statistical analysis, trend prediction and intelligent decision-making, the system conducts in-depth and systematic analysis of the disease detection results and 3D modeling results. It can comprehensively count the distribution of diseases, accurately predict the development trend of diseases, and combine multiple factors such as the actual use of roads and traffic flow to intelligently generate highly targeted and feasible road maintenance suggestions, providing scientific and powerful support for the decision-making of road management departments, thereby effectively ensuring the safety, stability and efficient operation of roads.

[0058] S8. The maintenance module performs maintenance on relevant roads in a timely manner according to the maintenance report; monitors relevant roads according to the prediction results, and performs maintenance on relevant roads when the damage maintenance level or time is reached.

[0059] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A road damage detection system based on multi-source data fusion, characterized in that: It includes data acquisition module, data preprocessing module, disease detection module, 3D modeling module, data analysis module and maintenance module; The data acquisition module is used to collect road surface data and send the collected data to the data preprocessing module for processing; the road surface data includes road data, road disease data, road vehicle data, non-motor vehicle data and pedestrian data; The data preprocessing module preprocesses the road data through data cleaning, data integration, data reduction and data transformation, classifies the road data, and forms estimated data for estimating diseases and current data for diagnosing diseases; and sends both the estimated data and current data to the disease detection module and the three-dimensional modeling module; the preprocessing includes data cleaning, data labeling and data fusion.

2. The defect detection module gives the corresponding defect level, defect type and defect repair plan based on the current defect data, and sends the defect repair plan to the maintenance module; defect types include pavement crack detection, pavement pothole detection and pavement settlement detection The 3D modeling module is used to construct a road damage model. Through the constructed surface model and underground model, the location of road damage can be determined and the efficiency of road damage simulation and analysis can be improved. The three-dimensional modeling module adopts three-dimensional point cloud processing technology to perform three-dimensional reconstruction on the road after disease detection and generate a high-precision three-dimensional model.

3. The road damage model includes a surface model and an underground model. The surface model is used to construct the three-dimensional geometric shape and texture information of the road surface, accurately presenting the road's flatness, slope, road markings and other features; The underground model is used to describe the distribution and condition of the underground structure of the road, including the location, depth and direction of underground pipelines, as well as the composition of the roadbed and potential underground diseases; The data analysis module, combined with 3D modeling, compares and analyzes the current data of the disease and the estimated data to obtain the predicted results of road diseases, confirm the accuracy of the predicted results, and provide the disease level, disease type, disease range, disease repair plan and repair period, and send them to the maintenance module; The maintenance module is used to notify maintenance personnel to maintain relevant roads in a timely manner.

4. The road damage detection system based on multi-source data fusion according to claim 1 is characterized in that: The data acquisition module includes a radar data unit, a video data unit, a three-dimensional point cloud data unit and a geographic location information unit; Wherein, the radar information collection unit is used to collect radar data information of the road; The video data acquisition unit is used to acquire video data of the road; The three-dimensional point cloud data acquisition unit is used to acquire three-dimensional point cloud data of the road; The geographical location information acquisition unit is used to acquire geographical location information data of the road.

5. The road damage detection system based on multi-source data fusion according to claim 1 is characterized in that: The data preprocessing module includes a data cleaning unit, a data labeling unit and a data fusion unit; Wherein, the data cleaning unit is used to remove noise and outliers; The data labeling unit is used to classify and label diseases; The data fusion unit is used to fuse the data collected by different sensors to improve the accuracy and completeness of the data.

6. The road damage detection system based on multi-source data fusion according to claim 1 is characterized in that: The disease detection module includes a road crack detection unit, a road pothole detection unit and a road settlement detection unit; The pavement crack detection unit is used to perform pavement crack detection on current data of the disease; The road pothole detection unit is used to perform road pothole detection on current data of the disease; The road surface settlement detection unit is used to perform road surface settlement detection on current data of the disease.

7. The road damage detection system based on multi-source data fusion according to claim 1 is characterized in that: The three-dimensional modeling module includes a surface model unit and an underground model unit.

8. The road damage detection system based on multi-source data fusion according to claim 1 is characterized in that: The data analysis module includes a disease distribution statistics unit, a disease development trend analysis unit and a road maintenance suggestion generation unit; The disease distribution statistics unit is used to perform detailed statistics and analysis on the occurrence frequency and distribution of road diseases detected in different sections and locations; The disease development trend analysis unit is used to predict the development direction and speed of the disease by using mathematical models and algorithms through comprehensive analysis of current data, historical data and prediction results of the disease; The road maintenance suggestion generating unit is used to intelligently generate targeted and feasible road maintenance suggestions based on the distribution statistics of the diseases and the development trend analysis, combined with the road usage and traffic flow factors.

9. A road damage detection system detection method based on multi-source data fusion according to any one of claims 1 to 6, characterized in that: The following steps are involved: S1. Collect historical data of different types of roads within 1-2 years, including the surface conditions, types of defects, scope of defects, vehicle flow, pedestrian flow, location, depth and direction of underground pipelines, roadbed composition and potential underground defects of the road; S2. Construct a 3D model of the road based on historical data and 3D point cloud data; S3. Based on historical data, through deep learning, establish an analysis and prediction model to provide support for the disease detection module and data analysis module; S4, collecting current road data from multiple angles through a data collection module, and performing data preprocessing to form current data and estimated data; S5, importing current data and estimated data into the analysis and prediction model and the three-dimensional model; S6. Based on the analysis and prediction model and the three-dimensional model, the disease detection module detects the disease condition of the current road, forms a maintenance report, and sends it to the maintenance module; S7. Based on the analysis and prediction model and the three-dimensional model, the data analysis module analyzes the current data and makes a prediction, obtains the prediction result of road damage, gives the damage grade, damage type, damage range, damage repair plan and repair period, and sends them to the maintenance module; S8. The maintenance module performs maintenance on relevant roads in a timely manner according to the maintenance report; monitors relevant roads according to the prediction results, and performs maintenance on relevant roads when the damage maintenance level or time is reached.

10. The detection method according to claim 1, characterized in that: The multiple angles include radar acquisition, video acquisition, 3D laser scanner acquisition and global positioning system.

Images