Roadbed disease detection method and system, electronic equipment and product

Abstract

The invention belongs to the technical field of road detection, and aims to provide a roadbed disease detection method and system, electronic equipment and a product. The method comprises the following steps: collecting multi-source detection data of a specified roadbed in real time and carrying out data fusion processing on the multi-source detection data to obtain fused detection data; performing crack detection processing on the roadbed surface image to obtain crack characteristic data of the specified roadbed, performing surface settlement analysis processing on the laser point cloud data to obtain settlement characteristic data of the specified roadbed, and performing deep disease analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information to obtain the subgrade settlement characteristic data of the specified roadbed. Obtaining deep disease characteristic data of the specified roadbed; according to the crack characteristic data, the settlement characteristic data and the deep-layer disease characteristic data, constructing a disease three-dimensional model of the specified roadbed; and carrying out visualization processing on the disease three-dimensional model. According to the invention, omnibearing disease detection on the surface and the deep layer of the roadbed can be realized, and the intelligent degree is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of road detection, and particularly relates to a subgrade disease detection method, system, electronic device and product. Background Art

[0002] With the rapid development of highways, railways and urban infrastructure, the subgrade, as the basic bearing structure of roads and tracks, its health status directly affects the safety and service life of road traffic. However, due to the combined action of natural factors (such as rain erosion, geological activities) and human factors (such as overloaded vehicles, improper construction), various diseases are prone to occur in the subgrade, including surface cracks, settlements, potholes, and deep - structure cavities and material looseness, etc. If these diseases are not detected and repaired in time, it will lead to a decline in the bearing capacity of the subgrade and even cause major safety accidents. Therefore, rapid and accurate detection and analysis of subgrade diseases are crucial for road maintenance and management.

[0003] Traditional subgrade detection relies on manual operation, that is, mainly relying on engineering personnel to conduct on - site evaluation of road surface cracks, settlements and other diseases through means such as visual inspection, knocking and measurement. This method has a low equipment cost and is easy to operate, but its disadvantages are also very obvious: the detection results rely on the experience of engineering personnel, with strong subjectivity, and it is difficult to guarantee the detection accuracy and reliability; at the same time, manual detection has low efficiency, is difficult to meet the needs of large - area subgrade detection, and is even more unable to deeply detect deep - layer diseases.

[0004] To overcome the problems existing in manual detection, currently, technologies such as optical image detection, lidar detection, and ground - penetrating radar and ultrasonic detection technologies have emerged. Among them, optical image detection technology uses high - definition cameras or drones to take road surface images, and identifies diseases such as road surface cracks and spalling through image - processing algorithms (such as edge detection, deep - learning models). This technology has high detection accuracy for surface diseases and is suitable for large - area rapid screening; lidar detection technology mainly collects high - precision three - dimensional point - cloud data on the subgrade surface, which can identify deformation diseases such as settlements and depressions; ground - penetrating radar and ultrasonic technologies can detect cavities, cracks and other abnormal features in the deep structure of the subgrade. Among them, ground - penetrating radar obtains underground information by analyzing the reflection signals of electromagnetic waves and is suitable for detecting hidden dangers inside the subgrade; ultrasonic waves calculate the disease depth and range through echo delay time and signal amplitude.

[0005] The above - mentioned existing technologies have played an important role in the field of subgrade disease detection. However, in the process of using the existing technologies, the inventor found that there are at least the following problems in the existing technologies, making it difficult to meet the actual needs: Existing single technical means usually target specific types of diseases. For example, optical images and lidar are mainly used for surface crack detection, while ground penetrating radar and ultrasonic waves are more suitable for deep structure detection. Each detection means is independent of each other, lacking a unified detection scheme, unable to cover surface and deep diseases simultaneously, and the detection range is limited. In addition, the detection data in the existing technology usually needs to be processed later by professionals, which is time-consuming and complex to operate, difficult to meet the needs of rapid diagnosis and real-time analysis, and the intelligent level is insufficient. Summary of the Invention

[0006] The present invention aims to solve the above technical problems at least to a certain extent. The present invention provides a subgrade disease detection method, system, electronic device and product.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a subgrade disease detection method, including: Collecting multi-source detection data of a specified subgrade in real time; wherein, the multi-source detection data includes a subgrade surface image collected by a camera, lidar point cloud data collected by a lidar, ultrasonic detection data collected by an ultrasonic sensor, and electromagnetic wave reflection information collected by a ground penetrating radar; Performing data fusion processing on the multi-source detection data to obtain fused detection data; Performing crack detection processing on the subgrade surface image in the fused detection data to obtain crack feature data of the specified subgrade; Performing surface settlement analysis processing on the lidar point cloud data in the fused detection data to obtain settlement feature data of the specified subgrade; Performing deep disease analysis processing on the ultrasonic detection data and electromagnetic wave reflection information in the fused detection data to obtain deep disease feature data of the specified subgrade; Constructing a disease three-dimensional model of the specified subgrade according to the crack feature data, the settlement feature data, and the deep disease feature data; Performing visualization processing on the disease three-dimensional model.

[0008] In a possible design, performing data fusion processing on the multi-source detection data to obtain fused detection data includes: Obtaining the sampling time of each sensor in the multi-source detection data, and obtaining a reference time according to the sampling time of each sensor; wherein each sensor includes a lidar, a camera, an ultrasonic sensor, and a ground penetrating radar, and the reference time is: ; In the formula, T k is thek The sampling time of each sensor, K is the total number of the sensors; Use the reference time to perform time synchronization processing on the multi-source detection data to obtain time-synchronized multi-source detection data; Perform spatial alignment processing on the time-synchronized multi-source detection data to obtain fused detection data.

[0009] In a possible design, perform crack detection processing on the roadbed surface image in the fused detection data to obtain the crack feature data of the specified roadbed, including: Input the roadbed surface image in the fused detection data into a pre-trained crack detection model, so as to obtain crack image information in the roadbed surface image based on the crack detection model; wherein, the crack detection model adopts a convolutional neural network model; Calculate the crack feature data of the specified roadbed according to the crack image information; wherein, the crack feature data includes crack length and crack width, and the crack length is: ; In the formula, ( x i , y i ) is the actual coordinate of the i th pixel point of the center line of the crack image information, x i = u i · R , y i = v i ·R , ( u i , v i ) is the pixel coordinate of the i th pixel point of the center line of the crack image information, R is the proportional coefficient between the pixel size and the actual size of the pixel points in the crack image information, ( x i+1 , y i+1 ) is the actual coordinate of the i + 1th pixel point of the center line of the crack image information, n is the total number of pixel points in the center line of the crack image information; The crack width is: ; In the formula, W i is the actual width of the i th pixel point of the center line of the crack image information, W i =△ u i ·R , △ u i is the pixel width of the i th pixel point of the center line of the crack image information in the normal direction of the center line of the crack image information.

[0010] In a possible design, the RANSAC algorithm is used to fit the laser point cloud data only on the reference plane, and then based on the reference plane position information obtained by the fitting, the settlement characteristic data of the specified subgrade is calculated.

[0011] In a possible design, the deep disease characteristic data includes disease depth data and material abnormal area information; correspondingly, the ultrasonic detection data and electromagnetic wave reflection information in the fused detection data are subjected to deep disease analysis and processing to obtain the deep disease characteristic data of the specified subgrade, including: Obtain the ultrasonic echo delay data in the ultrasonic detection data, and calculate the disease depth data of the specified subgrade according to the ultrasonic detection data; wherein, the disease depth data is: d = v·t’ / 2; In the formula, v is the propagation speed of the ultrasonic wave emitted by the ultrasonic sensor in the specified subgrade, t’ is the echo time of the ultrasonic detection data; Perform Fourier transform analysis on the electromagnetic wave reflection information to obtain spectral characteristic information; Obtain the material abnormal area information of the specified subgrade according to the spectral characteristic information.

[0012] In a possible design, a point cloud reconstruction algorithm is used to generate a three-dimensional disease model.

[0013] In a possible design, after generating the three-dimensional disease model, the method further includes: Obtain the historical three-dimensional disease model of the specified subgrade, and perform disease expansion trend prediction processing on the specified subgrade according to the historical three-dimensional disease model and the three-dimensional disease model to obtain the disease characteristic prediction data at a future specified time.

[0014] Second aspect, the present invention provides a subgrade disease detection system for implementing a subgrade disease detection method as described in any one of the above; the subgrade disease detection system includes: A data acquisition module for real-time acquisition of multi-source detection data of a specified subgrade; wherein, the multi-source detection data includes a subgrade surface image collected by a camera, lidar point cloud data collected by a lidar, ultrasonic detection data collected by an ultrasonic sensor, and electromagnetic wave reflection information collected by a ground penetrating radar; A data fusion module, communicatively connected to the data acquisition module, for performing data fusion processing on the multi-source detection data to obtain fused detection data; A road damage feature recognition module, communicatively connected to the data fusion module, for performing crack detection processing on the subgrade surface image in the fused detection data to obtain crack feature data of the specified subgrade; for performing surface settlement analysis processing on the lidar point cloud data in the fused detection data to obtain settlement feature data of the specified subgrade; and further for performing deep disease analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information in the fused detection data to obtain deep disease feature data of the specified subgrade; A model construction module, communicatively connected to the road damage feature recognition module, for constructing a three-dimensional disease model of the specified subgrade according to the crack feature data, the settlement feature data, and the deep disease feature data; A visualization processing module, communicatively connected to the model construction module, for performing visualization processing on the three-dimensional disease model.

[0015] Third aspect, the present invention provides an electronic device, including: A memory for storing computer program instructions; and, A processor for executing the computer program instructions to complete the operations of a subgrade disease detection method as described in any one of the above.

[0016] Fourth aspect, the present invention provides a computer program product including a computer program or instructions, and the computer program or the instructions, when executed by a computer, implement a subgrade disease detection method as described in any one of the above.

[0017] The beneficial effects of the present invention are: The present invention discloses a subgrade disease detection method, system, electronic device and product, which can realize the all-round disease detection of the subgrade surface and deep layer, and has a high degree of intelligence. Specifically, the present invention adopts a multi-sensor collaborative detection method to collaboratively collect the surface and deep layer disease data of the specified subgrade, and uses the multi-source data fusion technology to comprehensively analyze the detection results of multiple sensors, and then extracts the disease characteristics based on the fused detection data to obtain crack characteristic data, settlement characteristic data and deep layer disease characteristic data. Finally, a disease three-dimensional model is constructed based on this to quantify the surface and deep layer disease characteristics. In this process, the present invention uses a variety of sensors such as cameras, lidars, ultrasonic sensors and ground penetrating radars for disease detection, which can achieve a comprehensive coverage of the detection range. In addition, in the implementation process of the present invention, different types of disease characteristics are identified and classified based on the fused detection data, and a disease three-dimensional model is constructed and visualized based on this, realizing the real-time processing and visual display of the detection results, and further providing intuitive analysis results for maintenance personnel, improving the intelligence level, and providing a solid guarantee for road safety management.

[0018] Other beneficial effects of the present invention will be further described in the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a flowchart of a subgrade disease detection method in an embodiment; Figure 2 is a block diagram of a subgrade disease detection system in an embodiment; Figure 3 is a block diagram of an electronic device in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the present invention in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the drawings is only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. It should be noted here that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation to the present invention.

[0021] Embodiment 1: This embodiment discloses a subgrade disease detection method, which can be but is not limited to being executed by a computer device or virtual machine with certain computing resources, such as being executed by an electronic device such as a personal computer, a smart phone, a personal digital assistant or a wearable device, or being executed by a virtual machine.

[0022] Such as Figure 1As shown, a subgrade disease detection method may, but is not limited to, include the following steps: S1. Collect multi-source detection data of a specified subgrade in real time; among them, the multi-source detection data includes the subgrade surface image collected by a camera, the lidar point cloud data collected by a lidar, the ultrasonic detection data collected by an ultrasonic sensor, and the electromagnetic wave reflection information collected by a ground penetrating radar; it should be understood that the camera is a high-resolution camera, and the subgrade surface image collected by it is used to assist in subgrade crack detection; the ultrasonic detection data is used to indicate deep structural disease information such as cracks and cavities in the subgrade, and it includes information such as acoustic wave propagation time and signal amplitude; the ground penetrating radar realizes data collection by transmitting electromagnetic wave signals to the subgrade and receiving the reflected signals, and the electromagnetic wave reflection information includes its echo time and intensity, which can be used to reflect abnormal conditions of subgrade internal materials such as water content change information in the subgrade. In the implementation process, the multi-source detection data can be quickly transmitted and processed through a 5G (fifth-generation mobile communication technology) network or edge computing devices, which is conducive to quickly performing data analysis and processing.

[0023] In this embodiment, after collecting the multi-source detection data of the specified subgrade, the multi-source detection data is also denoised. Specifically, for the subgrade surface image, image enhancement technology such as the CLAHE (Contrast Limited Adaptive Histogram Equalization) algorithm is applied to improve the image quality. For the lidar point cloud data, filtering algorithms such as radius filtering or statistical filtering are used to remove the noise points. For the ultrasonic detection data and the electromagnetic wave reflection information, the Kalman filtering method is used for data smoothing processing, and then the preprocessed multi-source detection data is obtained, and subsequent data processing and analysis are performed based on the preprocessed multi-source detection data.

[0024] S2. Perform data fusion processing on the multi-source detection data to obtain the fused detection data. It should be noted that since the data collected by different sensors has differences in time and space, before disease feature recognition, the multi-source detection data is preprocessed by data fusion processing. The fused detection data can reflect the three-dimensional position and shape of the disease in a unified spatial coordinate system, thereby making full use of the advantages of different sensors, ensuring data consistency, avoiding the problem of inaccurate disease positions and ranges corresponding to the detection results, and at the same time further improving the subsequent analysis accuracy and facilitating the enhancement of disease positioning ability.

[0025] Specifically, in step S2, performing data fusion processing on the multi-source detection data to obtain the fused detection data includes: S201. Obtain the sampling time of each sensor in the multi-source detection data, and obtain the reference time according to the sampling time of each sensor; wherein, each sensor includes a lidar, a camera, an ultrasonic sensor, and a ground penetrating radar, and the reference time is: ; In the formula, T k is the sampling time of the k th sensor, K is the total number of the sensors; It should be noted that the calculation scheme of this reference time can be directly applied to the application scenarios where the sampling frequencies of the multi-source detection data are the same. If the sampling frequencies of the multi-source detection data are different, the detection data can be pre-processed by data interpolation or downsampling, and then the above calculation formula can be applied to calculate the reference time.

[0026] S202. Use the reference time to perform time synchronization processing on the multi-source detection data to obtain the time-synchronized multi-source detection data; it should be noted that using the reference time to perform time synchronization processing on the multi-source detection data, that is, using the reference time to perform time alignment processing on all detection data, can ensure the consistency of the detection data in time.

[0027] It should be noted that each sensor will record a timestamp when collecting data, indicating the sampling time when the data is collected. In this embodiment, the reference time is obtained based on the sampling time of each sensor, and the reference time is used to perform time synchronization processing on the multi-source detection data, that is, calculate the difference between the sampling time of the multi-source detection data and the reference time respectively, and then interpolate or compensate these time differences to make all detection data unified to the same time point.

[0028] S203. Perform spatial alignment processing on the time-synchronized multi-source detection data to obtain the fused detection data. It should be noted that the spatial alignment processing is to convert the coordinate systems of multiple sensors so that various detection data are unified to a global coordinate system. The representation of the detection data of a certain sensor in the preset global coordinate system: P = R 1· P raw + R 2. In the formula, R 1 represents the rotation matrix, which is used to describe the direction change of the coordinate system where a certain sensor is located relative to the global coordinate system, R 2 represents the translation matrix, which is used to describe the displacement between the origin of the coordinate system where a certain sensor is located and the origin of the global coordinate system, P raw represents the representation of the detection data of a certain sensor in its own coordinate system.

[0029] S3. Perform crack detection processing on the roadbed surface image in the fused detection data to obtain the crack feature data of the specified roadbed.

[0030] Specifically, in step S3 of this embodiment, performing crack detection processing on the roadbed surface image in the fused detection data to obtain the crack feature data of the specified roadbed includes: S301. Input the roadbed surface image in the fused detection data into a pre-trained crack detection model to obtain crack image information in the roadbed surface image based on the crack detection model; wherein, the crack detection model uses a convolutional neural network model; specifically, during the image processing of the convolutional neural network, features such as crack edges and textures can be extracted in advance, and then downsampling processing is performed to retain the main features and reduce the computational complexity, and then the ReLU activation function is used to enhance the non-linear expression ability of the model, and finally the crack image information is obtained, which is output in the form of a binary segmentation map for subsequent calculation of crack length, etc.

[0031] S302. Calculate the crack feature data of the specified roadbed according to the crack image information; wherein, the crack feature data includes crack length and crack width, and the crack length is: ; In the formula, ( x i , y i ) is the actual coordinate of the i -th pixel point on the center line of the crack image information, x i = u i · R , y i = v i ·R , ( u i , v i ) is the pixel coordinate of the i -th pixel point on the center line of the crack image information, R is the proportionality coefficient between the pixel size and the actual size of the pixel points in the crack image information, ( x i+1 , y i+1 ) is the actual coordinate of the i + 1-th pixel point on the center line of the crack image information,n is the total number of pixel points in the center line of the crack image information; The crack width is: ; In the formula, W i is the actual width of the i th pixel point of the center line of the crack image information, W i =△ u i ·R , △ u i is the pixel width of the i th pixel point of the center line of the crack image information in the normal direction of the center line of the crack image information. It should be noted that for the pixel points on the center line of the crack image information, calculate the crack width in its normal direction and summarize the widths of each pixel point to obtain the crack width, which can simplify the calculation difficulty of the crack width.

[0032] Furthermore, the crack feature data may further include the crack depth, and the crack depth is: H= △ I / (r·cosθ ); In the formula, △ I is the light intensity change information of the crack image information, r is a preset optical constant, θ is the light incident angle of the camera.

[0033] S4. Perform surface settlement analysis processing on the laser point cloud data in the fused detection data to obtain the settlement feature data of the specified roadbed.

[0034] Specifically, in step S4 of this embodiment, the RANSAC algorithm (RAndom SAmple Consensus) is used to only fit the laser point cloud data to the reference plane, and then based on the reference plane position information obtained by the fitting, the offset depth is calculated, that is, the settlement feature data of the specified roadbed.

[0035] S5. Perform deep disease analysis processing on the ultrasonic detection data and electromagnetic wave reflection information in the fused detection data to obtain the deep disease feature data of the specified roadbed.

[0036] Specifically, in this embodiment, the deep disease characteristic data includes disease depth data and material abnormal area information; correspondingly, in step S5, ultrasonic detection data and electromagnetic wave reflection information in the fused detection data are subjected to deep disease analysis and processing to obtain the deep disease characteristic data of the specified roadbed, including: S501. Obtain the ultrasonic echo delay data in the ultrasonic detection data, and calculate the disease depth data of the specified roadbed according to the ultrasonic detection data; wherein, the disease depth data is: d = v·t’ / 2; In the formula, v is the propagation speed of the ultrasonic wave emitted by the ultrasonic sensor in the specified roadbed, t’ is the echo time of the ultrasonic detection data, that is, the time difference from the ultrasonic wave emission to the reception of the ultrasonic detection data, and this parameter is recorded by the ultrasonic sensor; S502. Perform Fourier transform analysis on the electromagnetic wave reflection information to obtain spectral characteristic information; wherein, the spectral characteristic information is: ; In the formula, f ( t ) is the electromagnetic wave reflection information, which is a function of time t , ω is the frequency of the electromagnetic wave reflection information, i is the imaginary symbol.

[0037] It should be understood that the electromagnetic wave reflection signal is a time-domain signal, and the obtained spectral characteristic information is a frequency-domain signal, which contains the spectral characteristics of the electromagnetic wave reflection signal. Abnormalities such as cracks and cavities inside the foundation material will cause changes in the spectral characteristics of the electromagnetic wave reflection signal (such as a decrease in frequency peak and amplitude). Based on this, the material abnormality identification of the specified roadbed can be realized through the spectral characteristic information.

[0038] S503. Obtain the material abnormal area information of the specified roadbed according to the spectral characteristic information. Specifically, since material abnormality will cause the amplitude of the reflection signal to weaken, in the implementation process, the spectral change data in the spectral characteristic information can be obtained to confirm the material abnormality situation, and the material abnormal depth can be calculated according to the echo time of the electromagnetic wave reflection signal and the propagation speed of the electromagnetic wave sent by the ground penetrating radar in the specified roadbed (the calculation method is the same as that of the disease depth), and then further combined with the scanning path of the ground penetrating radar to locate the spatial range of the abnormal area.

[0039] It should be noted that in this embodiment, after obtaining the crack feature data, the settlement feature data, and the deep disease feature data, the above analysis results can be further integrated into a report in a format such as HTML (HyperText Markup Language), which is convenient for users to view and provides a reference for maintenance decision-making.

[0040] S6. Based on the crack feature data, the settlement feature data, and the deep disease feature data, a three-dimensional disease model of the specified roadbed is constructed. It should be understood that based on the three-dimensional disease model, information such as the disease location, depth, range, and type of the specified roadbed can be obtained. Specifically, in this embodiment, the three-dimensional model of the specified roadbed is reconstructed in advance according to the laser point cloud data, ultrasonic detection data, and electromagnetic wave reflection information. The point cloud interpolation algorithm is used to generate the roadbed surface, and the crack feature data is superimposed on the three-dimensional model as the surface texture feature of the three-dimensional model. Then, according to the deep disease feature data, a three-dimensional structure of the disease area is constructed using voxels or parametric geometric models (such as spheres and cuboids), and the three-dimensional structure of the disease area and the settlement feature data are integrated into the three-dimensional model, thereby forming a complete three-dimensional disease model.

[0041] Specifically, in step S6 of this embodiment, based on the crack feature data, the settlement feature data, and the deep disease feature data, a point cloud reconstruction algorithm such as the Poisson Surface Reconstruction algorithm is used to generate a three-dimensional disease model.

[0042] In the prior art, there is generally a lack of the function of intelligently predicting the future development trend of diseases based on the current disease situation. At the same time, the prior art usually only provides single detection results and fails to comprehensively evaluate historical data, disease expansion trends, and the overall condition of the roadbed, making it difficult to provide a comprehensive decision-making basis for road maintenance. Therefore, in step S6 of this embodiment, after generating the three-dimensional disease model, the method further includes: A. Obtain the historical three-dimensional disease model of the specified roadbed, and based on the historical three-dimensional disease model and the three-dimensional disease model, perform disease expansion trend prediction processing on the specified roadbed to obtain the predicted disease feature data at a future specified time. Specifically, in this embodiment, a time series prediction model (such as LSTM) can be used to implement the disease expansion trend prediction processing on the specified roadbed, and the predicted disease feature data includes data such as crack expansion speed and settlement aggravation trend.

[0043] In this embodiment, based on the result obtained by predicting the disease expansion trend of the specified roadbed, the disease change trend can be further displayed in a three-dimensional view. For example, in the three-dimensional view, the predicted data of the disease characteristics at a specified future time can be marked with a dotted line.

[0044] It should be noted that in this embodiment, by combining a time series analysis model, the prediction of the disease expansion trend is realized, which can further provide an early warning for road maintenance and assist in formulating a maintenance plan.

[0045] S7. Visualize the three-dimensional disease model. Specifically, by visualizing the three-dimensional disease model, an interactive three-dimensional view can be generated, and information such as the disease location can be displayed.

[0046] This embodiment can realize the comprehensive disease detection of the roadbed surface and deep layer, with a high degree of intelligence. Specifically, this embodiment adopts a multi-sensor collaborative detection method to collaboratively collect the surface and deep layer disease data of the specified roadbed, and uses multi-source data fusion technology to comprehensively analyze the detection results of multiple sensors. Then, based on the fused detection data, disease feature extraction is carried out to obtain crack feature data, settlement feature data, and deep layer disease feature data. Finally, a three-dimensional disease model is constructed based on this to quantify the surface and deep layer disease characteristics. In this process, this embodiment uses a variety of sensors such as cameras, lidar, ultrasonic sensors, and ground penetrating radar for disease detection, which can achieve a comprehensive coverage of the detection range. In addition, during the implementation of this embodiment, by identifying and classifying different types of disease characteristics from the fused detection data, a three-dimensional disease model is constructed and visualized, realizing the real-time processing and visual display of the detection results, thereby providing intuitive analysis results for maintenance personnel and improving the intelligent level, providing a solid guarantee for road safety management.

[0047] This embodiment can be widely used in scenarios such as regular maintenance of highway roadbeds, health detection of high-speed railway roadbeds, disease detection of airport runways, and evaluation and diagnosis of bridge foundation structures.

[0048] Embodiment 2: This embodiment discloses a roadbed disease detection system for implementing the roadbed disease detection method in Embodiment 1; as Figure 2 shown, the roadbed disease detection system includes: A data acquisition module for real-time acquisition of multi-source detection data of a specified roadbed; wherein, the multi-source detection data includes roadbed surface images collected by a camera, lidar point cloud data collected by lidar, ultrasonic detection data collected by an ultrasonic sensor, and electromagnetic wave reflection information collected by a ground penetrating radar. A data fusion module, communicatively connected to the data acquisition module, for performing data fusion processing on the multi-source detection data to obtain fused detection data; A road hazard feature recognition module, communicatively connected to the data fusion module, for performing crack detection processing on the roadbed surface image in the fused detection data to obtain crack feature data of the specified roadbed; for performing surface settlement analysis processing on the laser point cloud data in the fused detection data to obtain settlement feature data of the specified roadbed; and further for performing deep disease analysis processing on the ultrasonic detection data and electromagnetic wave reflection information in the fused detection data to obtain deep disease feature data of the specified roadbed; A model construction module, communicatively connected to the road hazard feature recognition module, for constructing a three-dimensional disease model of the specified roadbed according to the crack feature data, the settlement feature data, and the deep disease feature data; A visualization processing module, communicatively connected to the model construction module, for performing visualization processing on the three-dimensional disease model.

[0049] It should be noted that for the working process, working details, and technical effects of the roadbed disease detection system provided in this Embodiment 2, reference can be made to Embodiment 1, and details will not be elaborated here.

[0050] Embodiment 3: Based on Embodiment 1 or 2, this embodiment discloses an electronic device, which may be a smart phone, a tablet computer, a notebook computer, or a desktop computer, etc. The electronic device may be referred to as a user terminal, a portable terminal, a desktop terminal, etc. As Figure 3 shown, the electronic device includes: A memory, for storing computer program instructions; and, A processor, for executing the computer program instructions to complete the operations of any one of the roadbed disease detection methods described in Embodiment 1.

[0051] Specifically, the processor 301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). The processor 301 may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake state, also known as the CPU (Central Processing Unit); the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 301 may be integrated with a GPU (Graphics Processing Unit), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen.

[0052] The memory 302 may include one or more computer-readable storage media, and the computer-readable storage media may be non-transitory. The memory 302 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In some embodiments, the non-transitory computer-readable storage media in the memory 302 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 301 to implement the subgrade disease detection method provided in Embodiment 1 of the present application.

[0053] In some embodiments, the terminal may optionally further include: a communication interface 303 and at least one peripheral device. The processor 301, the memory 302, and the communication interface 303 may be connected through a bus or signal lines. Each peripheral device may be connected to the communication interface 303 through a bus, signal lines, or a circuit board. Specifically, the peripheral device includes at least one of a radio frequency circuit 304, a display screen 305, and a power supply 306.

[0054] The communication interface 303 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 301 and the memory 302. In some embodiments, the processor 301, the memory 302, and the communication interface 303 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 301, the memory 302, and the communication interface 303 may be implemented on a separate chip or circuit board, and this embodiment does not limit this.

[0055] The radio frequency circuit 304 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 304 communicates with the communication network and other communication devices through electromagnetic signals.

[0056] The display screen 305 is used to display the UI (User Interface). The UI can include any combination of graphics, text, icons, and videos.

[0057] The power supply 306 is used to supply power to each component in the electronic device.

[0058] Embodiment 4: Based on any one of Embodiments 1 to 3, this embodiment discloses a computer program product, including a computer program or instruction, and the computer program or the instruction, when executed by a computer, implements a subgrade disease detection method as described in any one of Embodiments 1. Wherein, the computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0059] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present invention is not limited to any specific combination of hardware and software.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting subgrade diseases, characterized in that, Including: Collecting multi-source detection data of a specified roadbed in real time; wherein, the multi-source detection data includes roadbed surface images collected by a camera, lidar point cloud data collected by a lidar, ultrasonic detection data collected by an ultrasonic sensor, and electromagnetic wave reflection information collected by a ground penetrating radar; Performing data fusion processing on the multi-source detection data to obtain fused detection data; Performing crack detection processing on the roadbed surface image in the fused detection data to obtain crack feature data of the specified roadbed; Performing surface settlement analysis processing on the lidar point cloud data in the fused detection data to obtain settlement feature data of the specified roadbed; Performing deep damage analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information in the fused detection data to obtain deep damage feature data of the specified roadbed; Constructing a three-dimensional damage model of the specified roadbed according to the crack feature data, the settlement feature data, and the deep damage feature data; Performing visualization processing on the three-dimensional damage model.

2. The subgrade disease detection method according to claim 1, characterized in that Performing data fusion processing on the multi-source detection data to obtain fused detection data, including: Obtaining the sampling time of each sensor in the multi-source detection data, and obtaining a reference time according to the sampling time of each sensor; wherein, each sensor includes a lidar, a camera, an ultrasonic sensor, and a ground penetrating radar, and the reference time is: ； In the formula, T k is the sampling time of the k th sensor, K is the total number of the said sensors; Performing time synchronization processing on the multi-source detection data using the reference time to obtain time-synchronized multi-source detection data; Performing spatial alignment processing on the time-synchronized multi-source detection data to obtain fused detection data.

3. A subgrade disease detection method according to claim 1, characterized in that Performing crack detection processing on the roadbed surface image in the fused detection data to obtain crack feature data of the specified roadbed, including: Inputting the roadbed surface image in the fused detection data into a pre-trained crack detection model to obtain crack image information in the roadbed surface image based on the crack detection model; wherein, the crack detection model adopts a convolutional neural network model; Calculating the crack feature data of the specified roadbed according to the crack image information; wherein, the crack feature data includes crack length and crack width, and the crack length is: ； wherein, ( x i , y i ) is the actual coordinate of the i th pixel point of the center line of the crack image information, x i = u i ·R , y i = v i ·R ), ( u i , v i ) is the pixel coordinate of the i th pixel point of the center line of the crack image information, R is the proportionality coefficient between the pixel size and the actual size of the pixel points in the crack image information, ( x i+1 , y i+1 ) is the actual coordinate of the i + 1th pixel point of the center line of the crack image information, n is the total number of pixel points in the center line of the crack image information; The crack width is: ； In the formula, W i is the actual width of the i th pixel point of the center line of the crack image information, W i =△ u i ·R where △ u i is the pixel width of the i th pixel point of the center line of the crack image information in the normal direction of the center line of the crack image information.

4. A subgrade disease detection method according to claim 1, characterized in that, Using the RANSAC algorithm to only fit the lidar point cloud data on the reference plane, and then calculating the settlement feature data of the specified roadbed based on the position information of the fitted reference plane.

5. A subgrade disease detection method according to claim 1, characterized in that The deep damage feature data includes damage depth data and material anomaly area information; correspondingly, performing deep damage analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information in the fused detection data to obtain the deep damage feature data of the specified roadbed, including: Obtaining the ultrasonic echo delay data in the ultrasonic detection data, and calculating the damage depth data of the specified roadbed according to the ultrasonic detection data; wherein, the damage depth data is: d = v·t’ / 2； Wherein, v is the propagation speed of the ultrasonic wave emitted by the ultrasonic sensor in the specified roadbed; t’ is the echo time of the ultrasonic detection data; Performing Fourier transform analysis processing on the electromagnetic wave reflection information to obtain spectrum feature information; Obtaining the material anomaly area information of the specified roadbed according to the spectrum feature information.

6. The subgrade disease detection method according to claim 1, wherein, Generate a 3D model of the disease using a point cloud reconstruction algorithm.

7. A subgrade disease detection method according to claim 1, characterized in that, After generating the 3D model of the disease, the method further includes: Obtain the historical 3D model of the disease of the specified roadbed, and based on the historical 3D model of the disease and the 3D model of the disease, perform a prediction process on the disease expansion trend of the specified roadbed to obtain the predicted data of the disease characteristics at a future specified time.

8. A subgrade disease detection system, characterized in that, For implementing a roadbed disease detection method according to any one of claims 1 to 7; the roadbed disease detection system includes: A data acquisition module for real-time acquisition of multi-source detection data of a specified roadbed; wherein, the multi-source detection data includes roadbed surface images collected by a camera, lidar point cloud data collected by a lidar, ultrasonic detection data collected by an ultrasonic sensor, and electromagnetic wave reflection information collected by a ground penetrating radar; A data fusion module, communicatively connected to the data acquisition module, for performing data fusion processing on the multi-source detection data to obtain the fused detection data; A road damage feature recognition module, communicatively connected to the data fusion module, for performing crack detection processing on the roadbed surface image in the fused detection data to obtain the crack feature data of the specified roadbed; for performing surface settlement analysis processing on the lidar point cloud data in the fused detection data to obtain the settlement feature data of the specified roadbed; and also for performing deep disease analysis processing on the ultrasonic detection data and the electromagnetic wave reflection information in the fused detection data to obtain the deep disease feature data of the specified roadbed; A model construction module, communicatively connected to the road damage feature recognition module, for constructing a 3D model of the disease of the specified roadbed based on the crack feature data, the settlement feature data, and the deep disease feature data; A visualization processing module, communicatively connected to the model construction module, for performing visualization processing on the 3D model of the disease.

9. An electronic device, characterized in that, Including: A memory for storing computer program instructions; And, A processor for executing the computer program instructions to complete the operations of a roadbed disease detection method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, The computer program or the instructions, when executed by a computer, implement a roadbed disease detection method according to any one of claims 1 to 7.

Images